Authors: Xingyu Xiao, Jiejuan Tong, Jun Sun, Zhe Sui, Peng Chen, Jingang Liang, Haitao Wang
Abstract: The rapid digitization of nuclear power plant main control rooms has fundamentally reshaped operator interaction patterns, introducing complex soft-control behaviors and elevated cognitive risks that are not adequately addressed by existing human reliability analysis approaches. Although recent advances in large language models and autonomous agents offer new opportunities for intelligent decision support, their deployment in safety critical environments remains constrained by risks of hallucinated reasoning and weakened human authority. This study proposes NuHF Claw, a persistent cognitive-risk agent framework that enables risk governed human centered autonomy for digital nuclear operations. The core methodological innovation lies in the introduction of a risk constrained agent runtime, which tightly couples cognitive state inference with probabilistic safety assessment to regulate autonomous system behavior in real time. By integrating cognitively grounded workload and situational awareness estimation with dynamic human error probability prediction, the framework transforms conventional offline reliability analysis into a proactive intervention mechanism embedded directly within operational workflows. Experimental validation on a high-fidelity digital control room simulator demonstrates that NuHF Claw can anticipate interface induced cognitive degradation, dynamically constrain unsafe autonomous recommendations, and provide risk-aware navigational guidance while preserving human decision authority. The results highlight a fundamental shift from automation-driven operation toward cognition-aware autonomy, offering a principled pathway for the safe integration of intelligent agents into next-generation nuclear control environments.
Authors: Hong Su
Abstract: Large Language Model (LLM) agents have demonstrated remarkable capabilities in reasoning and tool use, yet they often suffer from rigid, reactive control flows that limit their adaptability and efficiency. Most existing frameworks rely on fixed pipelines or failure-triggered reflection, causing agents to act impulsively or correct errors only after they occur. In this paper, we introduce Heartbeat-Driven Autonomous Thinking Activity Scheduling, a mechanism that enables proactive, adaptive, and continuous self-regulation. Mirroring the natural rhythm of human cognition, our system employs a periodic ``heartbeat'' mechanism to orchestrate a dynamic repertoire of cognitive modules (e.g., Planner, Critic, Recaller, Dreamer). Unlike traditional approaches that rely on hard-coded symbolic rules or immediate reactive triggers, our scheduler learns to determine when to engage specific thinking activities -- such as recalling memories, summarizing experiences, or strategic planning -- based on temporal patterns and historical context. This functional approach allows cognitive modules to be dynamically added or removed without structural reengineering. Meanwhile, we propose a meta-learning strategy for continual policy adaptation, where the scheduler optimizes its cognitive strategy over time using historical interaction logs. Evaluation results demonstrate that our approach effectively learns to schedule cognitive activities based on historical data and can autonomously integrate new thinking modules.
Authors: Pierre Lotte (EPE UT, IRIT), Andr\'e P\'eninou (UT2J, IRIT-SIG, IRIT), Olivier Teste (IRIT-SIG, IRIT, UT2J, Comue de Toulouse)
Abstract: Reliable evaluation of anomaly detection methods in multivariate time series remains an open challenge, largely due to the limitations of existing benchmark datasets. Current resources often lack fine-grained anomaly annotations, do not provide explicit intervariable and temporal dependencies, and offer little insight into the underlying generative mechanisms. These shortcomings hinder the development and rigorous comparison of detection models, especially those targeting interpretable and variable-specific outputs. To address this gap, we introduce Fun-TSG, a fully customizable time series generator designed to support high-quality evaluation of anomaly detection systems. Our tool enables both fully automated generation, based on randomly sampled dependency structures and anomaly types, and manual generation through user-defined equations and anomaly configurations. In both cases, it provides full transparency over the data generation process, including access to ground-truth anomaly labels at the variable and timestamp levels. Fun-TSG supports the creation of diverse, interpretable, and reproducible benchmarking scenarios, enabling fine-grained performance analysis for both classical and modern anomaly detection models.
Authors: Pramudita Satria Palar, Paul Saves, Muhammad Daffa Robani, Nicolas Verstaevel, Moncef Garouani, Julien Aligon, Koji Shimoyama, Joseph Morlier, Benoit Gaudou
Abstract: The simulation of complex systems increasingly relies on sophisticated but fundamentally opaque computational black-box simulators. Surrogate models play a central role in reducing the computational cost of complex systems simulations across a wide range of scientific and engineering domains. Notwithstanding, they inevitably inherit and often exacerbate this black-box nature, obscuring how input variables drive physical responses. Conversely, Explainable Artificial Intelligence (XAI) offers powerful tools to unpack these models. Yet, XAI methods struggle with engineering-specific constraints, such as highly correlated inputs, dynamical systems, and rigorous reliability requirements. Consequently, surrogate modeling and XAI have largely evolved as distinct fields of research, despite their strong complementarity. To reconnect these approaches, this state-of-the-art survey provides a structured perspective that maps existing XAI techniques onto the various stages of surrogate modeling workflows for design and exploration. To ground this synthesis, we draw upon illustrative applications across both equation-based simulations and agent-based modeling. We survey a broad spectrum of techniques, highlighting their strengths for revealing interactions and supporting human comprehension. Finally, we identify pressing open challenges, including the explainability of dynamical systems and the handling of mixed-variable systems, and propose a research agenda to make explainability a core, embedded element of simulation-driven workflows from model construction through decision-making. By transforming opaque emulators into explainable tools, this agenda empowers practitioners to move beyond accelerating simulations to extracting actionable insights from complex system behaviors.
Authors: Taylor Olson
Abstract: The field of machine ethics aims to build Artificial Moral Agents (AMAs) to better understand morality and make AI agents safer. To do so, many approaches encode human moral intuition as a set of axioms on actions e.g., do not harm, you must help others. However, this introduces (at least) two limitations for future AMAs. First, it does not consider the agent's purposes in performing the action. Second, it assumes that we humans can enumerate our moral intuition. This paper explores formalizing a moral procedure that alleviates these two limitations. We specifically consider Kantian ethics and present a multi-sorted quantified modal logic we call the Formula of the Universal Law Logic (FULL). The FULL formalizes Kant's first formulation of the categorical imperative, the Formula of the Universal Law (FUL), and concepts such as causality and agency. We demonstrate on three cases from Kantian ethics that the FULL can reason to evaluate agents' actions for certain purposes without built-in moral intuition, given that it has sufficient (non-normative) background knowledge. Therefore, the FULL is a contribution towards more robust and autonomous AMAs, and a more formal understanding of Kantian ethics.
Authors: Wangjie Gan, Miao Pan, Linbo Xi, Wenqi Zhang, Jintao Chen, Jianwei Yin, Xuhong Zhang
Abstract: Large language models are typically post-trained using supervised fine-tuning (SFT) and reinforcement learning (RL), yet effectively unifying efficient knowledge injection with robust generalization remains challenging. In this work, we provide a training-dynamics analysis showing that SFT can be interpreted as a special case of policy gradient optimization with an extremely sparse implicit reward and unstable inverse-probability weighting, which together lead to single-path dependency, entropy collapse, and gradient explosion. Motivated by this diagnosis, we propose Group Fine-Tuning (GFT), a unified post-training framework that addresses these intrinsic limitations through two mechanisms: Group Advantage Learning, which constructs diverse response groups and derives normalized contrastive supervision to alleviate reward sparsity, and Dynamic Coefficient Rectification, which adaptively bounds inverse-probability weights to stabilize optimization while preserving efficient knowledge injection. Experiments demonstrate that GFT consistently surpasses SFT-based methods and yields policies that integrate more smoothly with subsequent RL training.
Authors: Kinhei Lee, Peiyuan Jing, Zhenxuan Zhang, Yue Yang, Tao Wang, Dominic C Marshall, Yingying Fang, Guang Yang
Abstract: Large scale vision language models have shown promise in automating chest Xray interpretation, yet their clinical utility remains limited by a gap between model outputs and radiologist reasoning. Most systems optimize for semantic information without emulating how experts visually examine medical images, often overlooking critical findings or diverging from established diagnostic workflows. Radiologists follow structured protocols (e.g., the ABCDEF approach) that ensure all clinically relevant regions are systematically examined, reducing missed findings and supporting reliable diagnostic reasoning. We introduce GazeX, a vision language model that leverages radiologists' eye tracking data as a behavioral prior to model expert diagnostic reasoning. By incorporating gaze trajectories and fixation patterns into pretraining, GazeX learns to follow the spatial and temporal structure of radiologist attention and integrates observations in a clinically meaningful sequence. Using a curated dataset of over 30,000 gaze key frames from five radiologists, we demonstrate that GazeX produces more accurate, interpretable, and expert consistent outputs across radiology report generation, disease grounding, and visual question answering, utilizing 231,835 radiographic studies, 780,014 question answer pairs, and 1,162 image sentence pairs with bounding boxes. Unlike autonomous reporting systems, GazeX produces verifiable evidence artifacts, including inspection trajectories and finding linked localized regions, enabling efficient human verification and safe human AI collaboration. Learning through expert eyes provides a practical route toward more trustworthy, explainable, and diagnostically robust AI systems for radiology and beyond.
Authors: Aaron Pache, Mark CW van Rossum
Abstract: Synaptic plasticity is metabolically expensive, yet animals continuously update their internal models without exhausting energy reserves. However, when artificial neural networks are trained, the network parameters are typically updated on every sample that is presented, even if the sample was classified correctly. Inspired by the human negativity bias and error-related negativity, we propose 'memorized mistake-gated learning' -- a biologically plausible plasticity rule where synaptic updates are strictly gated by current and past classification errors. This reduces the number of updates the network needs to make by $50\%\sim80\%$. Mistake gating is particularly well suited in two cases: 1) For incremental learning where new knowledge is acquired on a background of pre-existing knowledge, 2) For online learning scenarios when data needs to be stored for later replay, as mistake-gating reduces storage buffer requirements. The algorithm can be implemented in a few lines of code, adds no hyper-parameters, and comes at negligible computational overhead. Learning on mistakes is an energy efficient and biologically relevant modification to commonly used learning rules that is well suited for continual learning.
Authors: Duo Lu, Andrew Crotty, U\u{g}ur \c{C}etintemel
Abstract: Agentic AI systems are becoming commonplace in domains that require long-lived, stateful decision-making in continuously evolving conditions. As such, correctness depends not only on the output of individual model calls, but also on how to best adapt when incorporating new evidence or revising prior conclusions. However, existing frameworks rely on imperative control loops, ephemeral memory, and prompt-embedded logic, making agent behavior opaque, brittle, and difficult to verify. This paper introduces Credo, which represents semantic state as beliefs and regulates behavior using declarative policies defined over these beliefs. This design supports adaptive, auditable, and composable execution through a database-backed semantic control plane. We showcase these concepts in a decision-control scenario, where beliefs and policies declaratively guide critical execution choices (e.g., model selection, retrieval, corrective re-execution), enabling dynamic behavior without requiring any changes to the underlying pipeline code.
Authors: Ivan Ternovtsii, Yurii Bilak
Abstract: Sparse Mixture-of-Experts (MoE) architectures employ increasingly sophisticated routing mechanisms -- learned routers, multi-hop trajectories, token-dependent gating. We ask: does routing topology actually determine language modeling quality? We build a geometric MoE (ST-MoE) using cosine-similarity routing against learned centroids in a low-dimensional space ($d_{space} = 64$), requiring 80% fewer routing parameters than standard linear routers. Through 62 controlled experiments on WikiText-103 at 76--84M parameters trained to convergence (50K steps, 1.64B tokens), we find that routing topology does not determine asymptotic perplexity (PPL): five cosine-routing variants are statistically equivalent within a 1-PPL margin (Two One-Sided Tests [TOST], $p < 0.05$ for all 10 pairwise comparisons; 15 runs across 3 seeds, observed range 33.93--34.72). The finding extends to hash, random-fixed, and top-1 routing (single-seed; graceful 1.1--2.2 PPL degradation) and replicates on OpenWebText (0.03 PPL gap, 6 runs, 3 seeds each). A standard linear router with 5.3$\times$ more routing parameters reaches PPL 32.76, but iso-parameter cosine routing closes 67% of this gap -- the true mechanism advantage is $\sim$1.2%. The mechanistic explanation is convergent redundancy: multi-hop updates are collinear ($\cos(\Delta h_0, \Delta h_1) = 0.805$), implementing magnitude amplification rather than compositional reasoning; a single learnable scalar replicates multi-hop performance. As a practical payoff, zero-shot relative-norm halting saves 25% of MoE FLOPs at +0.12% PPL. Expert-level specialization and causal controllability -- which coexist with topology-level equifinality -- are explored in a companion paper.
Authors: Muhammad Imam Luthfi Balaka, Raul Castro Fernandez
Abstract: Data analysts working with relational data often start with vague or underspecified questions and refine them iteratively as they explore the data. To support this iterative process, we demonstrate Pneuma-Seeker, a system that reifies a user's information need as explicit, inspectable relational specifications, enabling iterative refinement of the information need, targeted data discovery, and provenance-aware execution. Through two real-world procurement use cases, we show how Pneuma-Seeker leverages LLMs as transparent, interactive analytical collaborators rather than opaque answer engines.
Authors: Ivan Ternovtsii, Yurii Bilak
Abstract: Sparse Mixture-of-Experts (MoE) models scale parameters while fixing active computation per token, but the specialization of individual experts remains opaque. In a companion paper we showed that routing topology is quality-neutral: five structurally different configurations converge to statistically equivalent language modeling quality. Here we show that expert identity is nonetheless causally meaningful: individual rank-1 experts are monosemantic by construction, and cosine-similarity routing in a low-dimensional metric space makes their specialization directly inspectable. We present four lines of evidence. First, projecting expert output vectors through the unembedding matrix yields a Semantic Dictionary: 15% of experts are monosemantic specialists spanning 10 categories (temporal, geographic, cardinal, discourse, emotional, financial, military, scientific). Second, routing exhibits a frequency-to-syntax gradient: early layers separate tokens by word frequency, deeper layers by syntactic class (Zipf-confound controls, all $p < 0.001$). Third, causal interventions confirm these labels: steering toward a temporal expert's centroid increases P(temporal) by +321% (median across 44 prompts); suppressing a geographic expert drops P(geographic) by -23%; rewriting an expert's output vector halves target-category probability, and effects compose additively across layers. Fourth, the interventions are not unique to cosine routing: linear routers support comparable steering, but only cosine routing provides geometric transparency -- expert specialization is readable directly from the centroid matrix. MoE expert-level specialization is a first-class interpretability primitive: architecturally monosemantic, causally validated, and controllable at inference with zero overhead.
Authors: Ana Mar\'ia G\'omez Ruiz (UGA), Thao Dang (VERIMAG - IMAG, CNRS, UGA), Alexandre Donz\'e
Abstract: We propose a Reinforcement Learning (RL) based control design framework for handling complex tasks. The approach extends the concept of Reward Machines (RM) with Signal Temporal Logic (STL) formulas that can be used for event generation. The use of STL allows not only a more efficient representation of rewards for complex tasks but also guiding the training process to converge towards behaviors satisfying specified requirements. We also propose an implementation of the framework that leverages the STL online monitoring algorithms. We illustrate the framework with three case studies (minigrid, cart-pole and high-way environments) with non-trivial tasks.
Authors: Ruiyi Zhang, Peijia Qin, Qi Cao, Li Zhang, Pengtao Xie
Abstract: AI models underpin modern intelligent systems, driving advances across science, medicine, finance, and technology. Yet developing high-performing AI models remains a labor-intensive process that requires expert practitioners to iteratively design architectures, engineer representations, implement training pipelines and refine approaches through empirical evaluation. Existing AutoML methods partially alleviate this burden but remain limited to narrow aspects such as hyperparameter optimization and model selection within predefined search spaces, leaving the full development lifecycle largely dependent on human expertise. To address this gap, we introduce AIBuildAI, an AI agent that automatically builds AI models from a task description and training data. AIBuildAI adopts a hierarchical agent architecture in which a manager agent coordinates three specialized sub-agents: a designer for modeling strategy, a coder for implementation and debugging, and a tuner for training and performance optimization. Each sub-agent is itself a large language model (LLM) based agent capable of multi-step reasoning and tool use, enabling end-to-end automation of the AI model development process that goes beyond the scope of existing AutoML approaches. We evaluate AIBuildAI on MLE-Bench, a benchmark of realistic Kaggle-style AI development tasks spanning visual, textual, time-series and tabular modalities. AIBuildAI ranks first on MLE-Bench with a medal rate of 63.1%, outperforming all existing baseline methods and matching the capability of highly experienced AI engineers. These results demonstrate that hierarchical agent systems can automate the full AI model development process from task specification to deployable model, suggesting a pathway toward broadly accessible AI development with minimal human intervention.
Authors: Saumik Narayanan, Raja Panjwani, Siddhartha Sen, Chien-Ju Ho
Abstract: AI systems are increasingly used to assist humans in sequential decision-making tasks, yet determining when and how an AI assistant should intervene remains a fundamental challenge. A potential baseline is to recommend the optimal action according to a strong model. However, such actions assume optimal follow-up actions, which human decision makers may fail to execute, potentially reducing overall performance. In this work, we propose and study value-aware interventions, motivated by a basic principle in reinforcement learning: under the Bellman equation, the optimal policy selects actions that maximize the immediate reward plus the value function. When a decision maker follows a suboptimal policy, this policy-value consistency no longer holds, creating discrepancies between the actions taken by the policy and those that maximize the immediate reward plus the value of the next state. We show that these policy-value inconsistencies naturally identify opportunities for intervention. We formalize this problem in a Markov decision process where an AI assistant may override human actions under an intervention budget. In the single-intervention regime, we show that the optimal strategy is to recommend the action that maximizes the human value function. For settings with multiple interventions, we propose a tractable approximation that prioritizes interventions based on the magnitude of the policy-value discrepancy. We evaluate these ideas in the domain of chess by learning models of humans from large-scale gameplay data. In simulation, our approach consistently outperforms interventions based on the strongest chess engine (Stockfish) in a wide range of settings. A within-subject human study with 20 players and 600 games further shows that our interventions significantly improve performance for low- and mid-skill players while matching expert-engine interventions for high-skill players.
Authors: Jillian Fisher, Jennifer Neville, Chan Young Park
Abstract: A common approach to personalization in large language models (LLMs) is to incorporate a subset of the user memory into the prompt at inference time to guide the model's generation. Existing methods select these subsets primarily using similarity between user memory items and input queries, ignoring how features actually affect the model's response distribution. We propose Response-Utility optimization for Memory Selection (RUMS), a novel method that selects user memory items by measuring the mutual information between a subset of memory and the model's outputs, identifying items that reduce response uncertainty and sharpen predictions beyond semantic similarity. We demonstrate that this information-theoretic foundation enables more principled user memory selection that aligns more closely with human selection compared to state-of-the-art methods, and models $400\times$ larger. Additionally, we show that memory items selected using RUMS result in better response quality compared to existing approaches, while having up to $95\%$ reduction in computational cost.
Authors: Weixiang Shen, Bailiang Jian, Jun Li, Che Liu, Johannes Moll, Xiaobin Hu, Daniel Rueckert, Hongwei Bran Li, Jiazhen Pan
Abstract: Tool-augmented large language model (LLM) agents can orchestrate specialist classifiers, segmentation models, and visual question-answering modules to interpret chest X-rays. However, these agents still solve each case in isolation: they fail to accumulate experience across cases, correct recurrent reasoning mistakes, or adapt their tool-use behavior without expensive reinforcement learning. While a radiologist naturally improves with every case, current agents remain static. In this work, we propose Evo-MedAgent, a self-evolving memory module that equips a medical agent with the capacity for inter-case learning at test time. Our memory comprises three complementary stores: (1)~\emph{Retrospective Clinical Episodes} that retrieve problem-solving experiences from similar past cases, (2)~an \emph{Adaptive Procedural Heuristics} bank curating priority-tagged diagnostic rules that evolves via reflection, much like a physician refining their internal criteria, and (3)~a \emph{Tool Reliability Controller} that tracks per-tool trustworthiness. On ChestAgentBench, Evo-MedAgent raises multiple-choice question (MCQ) accuracy from 0.68 to 0.79 on GPT-5-mini, and from 0.76 to 0.87 on Gemini-3 Flash. With a strong base model, evolving memory improves performance more effectively than orchestrating external tools on qualitative diagnostic tasks. Because Evo-MedAgent requires no training, its per-case overhead is bounded by one additional retrieval pass and a single reflection call, making it deployable on top of any frozen model.
Authors: Nina \.Zukowska, Wolfgang Stammer, Bernt Schiele, Jonas Fischer
Abstract: Transparency of neural networks' internal reasoning is at the heart of interpretability research, adding to trust, safety, and understanding of these models. The field of mechanistic interpretability has recently focused on studying task-specific computational graphs, defined by connections (edges) between model components. Such edge-based circuits have been defined in the context of large language models, yet vision-based approaches so far only consider neuron-based circuits. These tell which information is encoded, but not how it is routed through the complex wiring of a neural network. In this work, we investigate whether useful mechanistic circuits can be identified through computational graphs in vision transformers. We propose an effective method for Automatic Visual Circuit Discovery (Vi-CD) that recovers class-specific circuits for classification, identifies circuits underlying typographic attacks in CLIP, and discovers circuits that lend themselves for steering to correct harmful model behavior. Overall, we find that insightful and actionable edge-based circuits can be recovered from vision transformers, adding transparency to the internal computations of these models.
Authors: Nenad Banfic, David Fan, Kunal Vaishnavi, Sam Kemp, Sunghoon Choi, Rui Ren, Sayan Shaw, Meng Tang
Abstract: Deploying high-quality automatic speech recognition (ASR) on edge devices requires models that jointly optimize accuracy, latency, and memory footprint while operating entirely on CPU without GPU acceleration. We conduct a systematic empirical study of state-of-the-art ASR architectures, encompassing encoder-decoder, transducer, and LLM-based paradigms, evaluated across batch, chunked, and streaming inference modes. Through a comprehensive benchmark of over 50 configurations spanning OpenAI Whisper, NVIDIA Nemotron, Parakeet TDT, Canary, Conformer Transducer, and Qwen3-ASR, we identify NVIDIA's Nemotron Speech Streaming as the strongest candidate for real-time English streaming on resource-constrained hardware. We then re-implement the complete streaming inference pipeline in ONNX Runtime and conduct a controlled evaluation of multiple post-training quantization strategies, including importance-weighted k-quant, mixed-precision schemes, and round-to-nearest quantization, combined with graph-level operator fusion. These optimizations reduce the model from 2.47 GB to as little as 0.67 GB while maintaining word error rate (WER) within 1% absolute of the full-precision PyTorch baseline. Our recommended configuration, the int4 k-quant variant, achieves 8.20% average streaming WER across eight standard benchmarks, running comfortably faster than real-time on CPU with 0.56 s algorithmic latency, establishing a new quality-efficiency Pareto point for on-device streaming ASR.
Authors: Mel Sohm, Charles Dezons, Sami Sellami, Oscar Ninou, Axel Pincon
Abstract: Synthetic augmentation is increasingly used to mitigate data scarcity in financial machine learning, yet its statistical role remains poorly understood. We formalize synthetic augmentation as a modification of the effective training distribution and show that it induces a structural bias--variance trade-off: while additional samples may reduce estimation error, they may also shift the population objective whenever the synthetic distribution deviates from regions relevant under evaluation. To isolate informational gains from mechanical sample-size effects, we introduce a size-matched null augmentation and a finite-sample, non-parametric block permutation test that remains valid under weak temporal dependence. We evaluate this framework in both controlled Markov-switching environments and real financial datasets, including high-frequency option trade data and a daily equity panel. Across generators spanning bootstrap, copula-based models, variational autoencoders, diffusion models, and TimeGAN, we vary augmentation ratio, model capacity, task type, regime rarity, and signal-to-noise. We show that synthetic augmentation is beneficial only in variance-dominant regimes, such as persistent volatility forecasting-while it deteriorates performance in bias-dominant settings, including near-efficient directional prediction. Rare-regime targeting can improve domain-specific metrics but may conflict with unconditional permutation inference. Our results provide a structural perspective on when synthetic data improves financial learning performance and when it induces persistent distributional distortion.
Authors: Dongxin Guo, Jikun Wu, Siu Ming Yiu
Abstract: Expert specialization is fundamental to Mixture-of-Experts (MoE) model success, yet existing metrics (cosine similarity, routing entropy) lack theoretical grounding and yield inconsistent conclusions under reparameterization. We present an information-geometric framework providing the first rigorous characterization of MoE specialization dynamics. Our key insight is that expert routing distributions evolve on the probability simplex equipped with the Fisher information metric, enabling formal analysis via Riemannian geometry. We prove that standard heuristic metrics violate parameterization invariance (Theorem 1), establish that specialization corresponds to geodesic flow with quantified approximation bounds (Theorem 2), and derive a failure predictor with theoretical threshold justification (Theorem 3). The framework introduces two principled metrics: Fisher Specialization Index (FSI) achieving r=0.91+/-0.02 correlation with downstream performance, and Fisher Heterogeneity Score (FHS) predicting training failure at 10% completion with AUC=0.89+/-0.03 -- outperforming validation-loss-based early stopping by 23% while requiring 40x fewer compute cycles. We validate intervention protocols achieving 87% recovery rate when FHS>1 is detected. Comprehensive experiments across language modeling (WikiText-103, C4), vision MoE (ImageNet), and scaling studies (8-64 experts, 125M-2.7B parameters) validate our theoretical predictions.
Authors: Michal Rosen-Zvi, Yoav Kan-Tor, Michael Danziger, Agata Ferretti, Javier Aula-Blasco, Julia Falcao, Ron Shamir, Mordechai Muszkat
Abstract: Healthcare disparities persist across socioeconomic boundaries, often attributed to unequal access to screening, diagnostics, and therapeutics. However, this perspective highlights that critical biases can emerge much earlier, during data collection and research prioritization, long before clinical implementation in cases where the focus of the studies and the data that is collected is at the molecular level. A vast number of studies focus on collecting omics data but the demographic information associated with these datasets is often not reported in the studies, and when it is reported, it shows big biases. An automated analysis of 4719 PubMed-indexed omics publications from 2015 to 2024 reveals that only a small fraction report ancestry or ethnicity information, with ancestry reporting improving slightly. Analysis of large-scale datasets commonly used for model training, such as CellxGene and GEO, reveals substantial population bias where European-ancestry data dominates. As biomedical foundation models become central to biomedical discovery with a paradigm in which base models are pretrained on large datasets and reusing them time and again for many different downstream tasks, they risk perpetuating or amplifying these early-stage biases, leading to cascading inequities that regulatory interventions cannot fully reverse. We propose a community-wide focus on three foundational principles: Provenance, Openness, and Evaluation Transparency to improve equity and robustness in biomedical AI. This approach aims to foster biomedical innovation that more effectively serves underserved populations and improves health outcomes.
Authors: MindDR Team, Li Auto Inc
Abstract: We present \textbf{Mind DeepResearch (MindDR)}, an efficient multi-agent deep research framework that achieves leading performance with only \textasciitilde30B-parameter models through a meticulously designed data synthesis and multi-stage training pipeline. The core innovation of MindDR lies in a collaborative three-agent architecture (Planning Agent, DeepSearch Agent, and Report Agent) and a four-stage agent-specialized training pipeline comprising SFT cold-start, Search-RL, Report-RL and preference alignment. With this regime, MindDR demonstrates competitive performance even with \textasciitilde30B-scale models. Specifically, MindDR achieves 45.7\% on BrowseComp-ZH, 42.8\% on BrowseComp, 46.5\% on WideSearch, 75.0\% on xbench-DS, and 52.5 on DeepResearch Bench, outperforming comparable-scale open-source agent systems and rivaling larger-scale models. MindDR has been deployed as an online product in Li Auto. Furthermore, we introduce \textbf{MindDR Bench}, a curated benchmark of 500 real-world Chinese queries from our internal product user interactions, evaluated through a comprehensive multi-dimensional rubric system rather than relying on a single RACE metric. On MindDR Bench, MindDR achieves a state-of-the-art score of 51.8.
Authors: Rohit Kumar Salla, Ramya Manasa Amancherla, Manoj Saravanan
Abstract: Large language models frequently produce mutually inconsistent answers when reasoning over multiple related queries. We study case-file logical consistency: maintaining a globally satisfiable belief state across interdependent queries. We introduce a benchmark of 390 multi-query reasoning instances with entailment/contradiction/unknown labels and propose set-level metrics including Case Satisfiability Rate, Contradiction Density and Revision Cost. Our solver-augmented approach extracts commitments, verifies global satisfiability and performs counterexample-guided repair. Across four reasoning domains, our method substantially reduces cross-query contradictions (SetCons: 0.56 to 0.94) while preserving per-query accuracy, demonstrating that global coherence is critical for robust multi-query reasoning.
Authors: Wei Zhu, Jian Zhang, Lixing Yu, Kun Yue, Zhiwen Tang
Abstract: Large Language Models (LLMs) achieve strong performance through extended inference-time deliberation, yet how their reasoning failures arise remains poorly understood. By analyzing model-generated reasoning trajectories, we find that errors are not uniformly distributed but often originate from a small number of early transition points, after which reasoning remains locally coherent but globally incorrect. These transitions coincide with localized spikes in token-level entropy, and alternative continuations from the same intermediate state can still lead to correct solutions. Based on these observations, we introduce GUARD, a targeted inference-time framework that probes and redirects critical transitions using uncertainty signals. Empirical evaluations across multiple benchmarks confirm that interventions guided by these failure dynamics lead to more reliable reasoning outcomes. Our findings highlight the importance of understanding when and how reasoning first deviates, complementing existing approaches that focus on scaling inference-time computation.
Authors: Adam Rida
Abstract: Every call to an LLM classification endpoint produces a labeled input-output pair already retained in production logs. These pairs constitute a free, growing training set: a lightweight surrogate trained on them can absorb a significant portion of future traffic at near-zero marginal inference cost. The open questions are when the surrogate is reliable enough to deploy, what it handles versus defers, and how that boundary evolves as data accumulates. We introduce TRACER (Trace-based Adaptive Cost-Efficient Routing), an open-source system that trains ML surrogates on an LLM's own production traces and governs deployment through a parity gate: the surrogate is activated only when its agreement with the LLM exceeds a user-specified threshold {\alpha}. To make the routing boundary transparent, TRACER generates interpretability artifacts describing which input regions the surrogate handles, where it plateaus, and why it defers. On a 77-class intent benchmark with a Sonnet 4.6 teacher, TRACER achieves 83-100% surrogate coverage depending on the quality target {\alpha}; on a 150-class benchmark, the surrogate fully replaces the teacher. On a natural language inference task, the parity gate correctly refuses deployment because the embedding representation cannot support reliable separation. The system is available as open-source software.
Authors: Pengfei Li, Shijie Wang, Fangyuan Li, Yikun Fu, Kaifeng Liu, Kaiyan Zhang, Dazhi Zhang, Yuqiang Li, Biqing Qi, Bowen Zhou
Abstract: Reinforcement learning (RL) paradigms have demonstrated strong performance on reasoning-intensive tasks such as code generation. However, limited trajectory diversity often leads to diminishing returns, which constrains the achievable performance ceiling. Search-enhanced RL alleviates this issue by introducing structured exploration, which remains constrained by the single-agent policy priors. Meanwhile, leveraging multiple interacting policies can acquire more diverse exploratory signals, but existing approaches are typically decoupled from structured search. We propose \textbf{MARS$^2$} (Multi-Agent Reinforced Tree-Search Scaling), a unified RL framework in which multiple independently-optimized agents collaborate within a shared tree-structured search environment. MARS$^2$ models the search tree as a learnable multi-agent interaction environment, enabling heterogeneous agents to collaboratively generate and refine candidate solutions within a shared search topology. To support effective learning, we introduce a path-level group advantage formulation based on tree-consistent reward shaping, which facilitates effective credit assignment across complex search trajectories. Experiments on code generation benchmarks show that MARS$^2$ consistently improves performance across diverse model combinations and training settings, demonstrating the effectiveness of coupling multi-agent collaboration with tree search for enhancing reinforcement learning. Our code is publicly available at https://github.com/TsinghuaC3I/MARTI.
Authors: Md Arid Hasan, Azhagu Meena SP, Aditya Khan, Abu Md Akteruzzaman Bhuiyan, Helal Uddin Ahmed, Joysree Debi, Farig Sadeque, Annie En-Shiun Lee, Syed Ishtiaque Ahmed
Abstract: Large language models (LLMs) show promise in generating supportive responses for mental health and counseling applications. However, their responses often lack cultural sensitivity, contextual grounding, and clinically appropriate guidance. This work addresses the gap of how to systematically incorporate domain-specific, clinically validated knowledge into LLMs to improve counseling quality. We utilize and compare two approaches, retrieval-augmented generation (RAG) and a knowledge graph (KG)-based method, designed to support para-counselors. Our KG is constructed manually and clinically validated, capturing causal relationships between stressors, interventions, and outcomes, with contributions from multidisciplinary people. We evaluated multiple LLMs in both settings using BERTScore F1 and SBERT cosine similarity, as well as human evaluation across five metrics, which is designed to directly measure the effectiveness of counseling beyond similarity at the surface level. The results show that KG-based approaches consistently improve contextual relevance, clinical appropriateness, and practical usability compared to RAG alone, demonstrating that structured, expert-validated knowledge plays a critical role in addressing LLMs limitations in counseling tasks.
Authors: Xing Zhang, Guanghui Wang, Yanwei Cui, Wei Qiu, Ziyuan Li, Bing Zhu, Peiyang He
Abstract: Prompt optimization in compound AI systems is statistically indistinguishable from a coin flip: across 72 optimization runs on Claude Haiku (6 methods $\times$ 4 tasks $\times$ 3 repeats), 49% score below zero-shot; on Amazon Nova Lite, the failure rate is even higher. Yet on one task, all six methods improve over zero-shot by up to $+6.8$ points. What distinguishes success from failure? We investigate with 18,000 grid evaluations and 144 optimization runs, testing two assumptions behind end-to-end optimization tools like TextGrad and DSPy: (A) individual prompts are worth optimizing, and (B) agent prompts interact, requiring joint optimization. Interaction effects are never significant ($p > 0.52$, all $F < 1.0$), and optimization helps only when the task has exploitable output structure -- a format the model can produce but does not default to. We provide a two-stage diagnostic: an \$80 ANOVA pre-test for agent coupling, and a 10-minute headroom test that predicts whether optimization is worthwhile -- turning a coin flip into an informed decision.
Authors: Ha Thanh Nguyen, Wachara Fungwacharakorn, Sabine Wehnert, May Myo Zin, Yuntao Kong, Jieying Xue, Micha{\l} Araszkiewicz, Randy Goebel, Ken Satoh
Abstract: We study the overall process of automatic formalization of GDPR provisions using large language models, within a human-in-the-loop verification framework. Rather than aiming for full autonomy, we adopt a role-specialized workflow in which LLM-based AI components, operating in a multi-agent setting with iterative feedback, generate legal scenarios, formal rules, and atomic facts. This is coupled with independent verification modules which include human reviewers' assessment of representational, logical, and legal correctness. Using this approach, we construct a high-quality dataset to be used for GDPR auto-formalization, and analyze both successful and problematic cases. Our results show that structured verification and targeted human oversight are essential for reliable legal formalization, especially in the presence of legal nuance and context-sensitive reasoning.
Authors: Zijian Zhang, Aiwei Yin, Amaan Baweja, Jiaru Bai, Ignacio Gustin, Varinia Bernales, Al\'an Aspuru-Guzik
Abstract: AI for science promises to accelerate the discovery process. The advent of large language models (LLMs) and agentic workflows enables the expediting of a growing range of scientific tasks. However, most of the current generation of agentic systems depend on static, hand-curated toolsets that hinder adaptation to new domains and evolving libraries. We present El Agente Forjador, a multi-agent framework in which universal coding agents autonomously forge, validate, and reuse computational tools through a four-stage workflow of tool analysis, tool generation, task execution, and iterative solution evaluation. Evaluated across 24 tasks spanning quantum chemistry and quantum dynamics on five coding agent setups, we compare three operating modes: zero-shot generation of tools per task, reuse of a curriculum-built toolset, and direct problem-solving with the coding agents as the baseline. We find that our tool generation and reuse framework consistently improves accuracy over the baseline. We also show that reusing a toolset built by a stronger coding agent can reduce API cost and substantially raises the solution quality for weaker coding agents. Case studies further demonstrate that tools forged for different domains can be combined to solve hybrid tasks. Taken together, these results show that LLM-based agents can use their scientific knowledge and coding capabilities to autonomously build reusable scientific tools, pointing toward a paradigm in which agent capabilities are defined by the tasks they are designed to solve rather than by explicitly engineered implementations.
Authors: Yubin Kim, Salman Rahman, Samuel Schmidgall, Chunjong Park, A. Ali Heydari, Ahmed A. Metwally, Hong Yu, Xin Liu, Xuhai Xu, Yuzhe Yang, Maxwell A. Xu, Zhihan Zhang, Cynthia Breazeal, Tim Althoff, Petar Sirkovic, Ivor Rendulic, Annalisa Pawlosky, Nicolas Stroppa, Juraj Gottweis, Elahe Vedadi, Alan Karthikesalingam, Pushmeet Kohli, Vivek Natarajan, Mark Malhotra, Shwetak Patel, Hae Won Park, Hamid Palangi, Daniel McDuff
Abstract: Scientific discovery in digital health requires converting continuous physiological signals from wearable devices into clinically actionable biomarkers. We introduce CoDaS (AI Co-Data-Scientist), a multi-agent system that structures biomarker discovery as an iterative process combining hypothesis generation, statistical analysis, adversarial validation, and literature-grounded reasoning with human oversight using large-scale wearable datasets. Across three cohorts totaling 9,279 participant-observations, CoDaS identified 41 candidate digital biomarkers for mental health and 25 for metabolic outcomes, each subjected to an internal validation battery spanning replication, stability, robustness, and discriminative power. Across two independent depression cohorts, CoDaS surfaced circadian instability-related features in both datasets, reflected in sleep duration variability (DWB, \rho = 0.252, p < 0.001) and sleep onset variability (GLOBEM, \rho = 0.126, p < 0.001). In a metabolic cohort, CoDaS derived a cardiovascular fitness index (steps/resting heart rate; \rho = -0.374, p < 0.001), and recovered established clinical associations, including the hepatic function ratio (AST/ALT; \rho = -0.375, p < 0.001), a known correlate of insulin resistance. Incorporating CoDaS-derived features alongside demographic variables led to modest but consistent improvements in predictive performance, with cross-validated \Delta R^2 increases of 0.040 for depression and 0.021 for insulin resistance. These findings suggest that CoDaS enables systematic and traceable hypothesis generation and prioritization for biomarker discovery from large-scale wearable data.
Authors: Liangda Fang, Yaohui Luo, Delong Li, Xuanxiang Huang, Quanlong Guan
Abstract: The exact cover problem is a classical NP-hard problem with broad applications in the area of AI. Algorithm DXZ is a method to count exact covers representing by zero-suppressed binary decision diagrams (ZBDDs). In this paper, we propose a zero-suppressed variant of decision decomposable negation normal form (in short, decision-ZDNNF), which is strictly more succinct than ZBDDs. We then design a novel parallel algorithm, namely DXD, which constructs a decision-ZDNNF representing the set of all exact covers. Furthermore, we improve DXD by dynamically updating connected components. The experimental results demonstrate that the improved DXD algorithm outperforms all of state-of-the-art methods.
Authors: Shiyuan Huang, Li Liu, Jincheng He, Leilani H. Gilpin
Abstract: When faced with complex spatial problems, humans naturally sketch layouts to organize their thinking, and the act of drawing further sharpens their understanding. In this work, we ask whether a similar principle holds for Large Language Models (LLMs): can learning to construct explicit visual layouts from spatial descriptions instill genuine spatial understanding? We introduce Text2Space, a dataset that pairs natural language descriptions with ground-truth ASCII grid layouts and spatial QA pairs, enabling us to separate failures in constructing spatial representations from failures in reasoning over them. We adopt ASCII because it is human-readable, operates entirely within the token space of language models, and encodes spatial relations in a structurally verifiable form. Our evaluation reveals a pronounced "Read-Write Asymmetry": LLMs interpret ASCII representations effectively but struggle to produce them from text, and these construction errors propagate to incorrect answers downstream. To address this limitation, we train models on layout construction (Text$\rightarrow$ASCII) and find that it significantly improves spatial reasoning from text alone, even without producing any ASCII at inference time. Combining construction with comprehension training further amplifies these gains. Crucially, these improvements transfer to three external spatial reasoning benchmarks, demonstrating that, much as sketching sharpens human spatial thinking, learning to construct explicit layouts instills spatial understanding that generalizes beyond the training format.
Authors: Chen Wang, Lai Wei, Yanzhi Zhang, Chenyang Shao, Zedong Dan, Weiran Huang, Ge Lan, Yue Wang
Abstract: Recent advances in reinforcement learning (RL) have improved the reasoning capabilities of large language models (LLMs) and vision-language models (VLMs). However, the widely used Group Relative Policy Optimization (GRPO) consistently suffers from entropy collapse, causing the policy to converge prematurely and lose diversity. Existing exploration methods introduce additional bias or variance during exploration, making it difficult to maintain optimization stability. We propose Unified Entropy Control for Reinforcement Learning (UEC-RL), a framework that provides targeted mechanisms for exploration and stabilization. UEC-RL activates more exploration on difficult prompts to search for potential and valuable reasoning trajectories. In parallel, a stabilizer prevents entropy from growing uncontrollably, thereby keeping training stable as the model consolidates reliable behaviors. Together, these components expand the search space when needed while maintaining robust optimization throughout training. Experiments on both LLM and VLM reasoning tasks show consistent gains over RL baselines on both Pass@1 and Pass@$k$. On Geometry3K, UEC-RL achieves a 37.9\% relative improvement over GRPO, indicating that it sustains effective exploration without compromising convergence and underscoring UEC-RL as a key for scaling RL-based reasoning in large models. Our code is available at https://github.com/597358816/UEC-RL.
Authors: David Y. Y. Tan, Kellie Chin, Jingxian Zhang
Abstract: We present AgentGA, a framework that evolves autonomous code-generation runs by optimizing the agent seed: the task prompt plus optional parent archives that initialize a fresh workspace. The outer loop searches over these reusable starting conditions rather than editing code directly. Each generation launches a fresh autonomous run from a reset workspace, while selected parent archives provide inherited artifacts that descendants can inspect and reuse. AgentGA couples a population-level genetic algorithm with long-horizon agents; selection uses deterministic 1:1 elite tournaments and operator allocation is adapted online with a modified Hedge controller. We instantiate the approach for tabular AutoML on the 16-competition Weco-Kaggle Lite benchmark. On the 10 benchmark runs reported here, AgentGA averages 74.52% Exceeds % of Human versus 54.15% for AIDE. Across 1135 parent-child comparisons, descendants given parent archives outperform runs started from scratch, indicating that inherited artifacts improve later autonomous runs. These findings support agent-seed optimization as a practical design point for autonomous code-search systems.
Authors: Zonghai Yao, Zhipeng Tang, Chengtao Lin, Xiong Luo, Benlu Wang, Juncheng Huang, Chin Siang Ong, Hong Yu
Abstract: Most medical multimodal benchmarks focus on static tasks such as image question answering, report generation, and plain-language rewriting. Patient education is more demanding: systems must identify relevant evidence across images, show patients where to look, explain findings in accessible language, and handle confusion or distress. Yet most patient education work remains text-only, even though combined image-and-text explanations may better support understanding. We introduce MedImageEdu, a benchmark for multi-turn, evidence-grounded radiology patient education. Each case provides a radiology report with report text and case images. A DoctorAgent interacts with a PatientAgent, conditioned on a hidden profile that captures factors such as education level, health literacy, and personality. When a patient question would benefit from visual support, the DoctorAgent can issue drawing instructions grounded in the report, case images, and the current question to a benchmark-provided drawing tool. The tool returns image(s), after which the DoctorAgent produces a final multimodal response consisting of the image(s) and a grounded plain-language explanation. MedImageEdu contains 150 cases from three sources and evaluates both the consultation process and the final multimodal response along five dimensions: Consultation, Safety and Scope, Language Quality, Drawing Quality, and Image-Text Response Quality. Across representative open- and closed-source vision-language model agents, we find three consistent gaps: fluent language often outpaces faithful visual grounding, safety is the weakest dimension across disease categories, and emotionally tense interactions are harder than low education or low health literacy. MedImageEdu provides a controlled testbed for assessing whether multimodal agents can teach from evidence rather than merely answer from text.
Authors: Saif Mahmoud
Abstract: Speculative decoding accelerates large language model (LLM) inference. It uses a small draft model to propose a tree of future tokens. A larger target model then verifies these tokens in a single batched forward pass. Despite the growing body of work on speculative methods, the degree to which the cognitive characteristics of a task affect acceptance probability remains largely unexplored. We present an empirical study of tree-based speculative decoding acceptance dynamics. Our study spans four well-established NLP benchmark domains: code generation, mathematical reasoning, logical reasoning, and open-ended chat. For this, we use TinyLlama-1.1B as the draft model against Llama-2-7B-Chat-GPTQ as the target. Over 99,768 speculative nodes collected from 200 prompts, we derive per-domain acceptance rates, expected accepted lengths, depth-acceptance profiles, and entropy-acceptance correlations. We find that task type is a stronger predictor of acceptance than tree depth. Furthermore, only the chat domain consistently yields an expected accepted length exceeding 1.0 token per step. We also show that the entropy-acceptance correlation is consistently negative but weak across all domains (rho in [-0.20, -0.15]). Counterintuitively, chat produces the highest entropy yet the highest acceptance rate. We attribute this divergence to the lexical predictability of RLHF-aligned register. These findings have direct implications for domain-aware speculation budgets and draft-model selection strategies. Index Terms--speculative decoding, large language model inference, tree attention, draft model, acceptance probability, LLM efficiency
Authors: Qianqian Xie, Qingheng Xiong, He Zhu, Tiantian Xia, Xueming Han, Fanyu Meng, Jiakai Wang, Zhiqi Bai, Chengkang Jiang, Zhaohui Wang, Yubin Guo, Yuqing Wen, Jiayang Mao, Zijie Zhang, Shihao Li, Yanghai Wang, Yuxiang Ren, Junlan Feng, Jiaheng Liu
Abstract: Deep Research Agents (DRAs) aim to solve complex, long-horizon research tasks involving planning, retrieval, multimodal understanding, and report generation, yet their evaluation remains challenging due to dynamic web environments and ambiguous task definitions. We propose DR$^{3}$-Eval, a realistic and reproducible benchmark for evaluating deep research agents on multimodal, multi-file report generation. DR$^{3}$-Eval is constructed from authentic user-provided materials and paired with a per-task static research sandbox corpus that simulates open-web complexity while remaining fully verifiable, containing supportive documents, distractors, and noise. Moreover, we introduce a multi-dimensional evaluation framework measuring Information Recall, Factual Accuracy, Citation Coverage, Instruction Following, and Depth Quality, and validate its alignment with human judgments. Experiments with our developed multi-agent system DR$^{3}$-Agent based on multiple state-of-the-art language models demonstrate that DR$^{3}$-Eval is highly challenging and reveals critical failure modes in retrieval robustness and hallucination control. Our code and data are publicly available.
Authors: Yiyu Qian, Liyuan Zhao, Tim Miller
Abstract: Monte-Carlo Tree Search (MCTS) is a fundamental sampling-based search algorithm widely used for online planning in sequential decision-making domains. Despite its success in driving recent advances in artificial intelligence, understanding the behavior of MCTS agents remains a challenge for both developers and users. This difficulty stems from the complex search trees produced through the simulation of numerous future states and their intricate relationships. A known weakness of standard MCTS is its reliance on highly selective tree construction, which may lead to the omission of crucial moves and a vulnerability to tactical traps. To resolve this, we incorporate shallow, full-width Minimax search into the rollout phase of multi-agent MCTS to enhance strategic depth. Furthermore, to demystify the resulting decision-making logic, we introduce \textsf{M2-PALE} (MCTS--Minimax Process-Aided Linguistic Explanations). This framework employs process mining techniques, specifically the Alpha Miner, iDHM, and Inductive Miner algorithms, to extract underlying behavioral workflows from agent execution traces. These process models are then synthesized by LLMs to generate human-readable causal and distal explanations. We demonstrate the efficacy of our approach in a small-scale checkers environment, establishing a scalable foundation for interpreting hybrid agents in increasingly complex strategic domains.
Authors: Xiangning Yu, Yuwei Guo, Yuqi Hou, Xiao Xue, Qun Ma
Abstract: LLM-empowered agent simulations are increasingly used to study social emergence, yet the micro-to-macro causal mechanisms behind macro outcomes often remain unclear. This is challenging because emergence arises from intertwined agent interactions and meso-level feedback and nonlinearity, making generative mechanisms hard to disentangle. To this end, we introduce \textbf{\textsc{CAMO}}, an automated \textbf{Ca}usal discovery framework from \textbf{M}icr\textbf{o} behaviors to \textbf{M}acr\textbf{o} Emergence in LLM agent simulations. \textsc{CAMO} converts mechanistic hypotheses into computable factors grounded in simulation records and learns a compact causal representation centered on an emergent target $Y$. \textsc{CAMO} outputs a computable Markov boundary and a minimal upstream explanatory subgraph, yielding interpretable causal chains and actionable intervention levers. It also uses simulator-internal counterfactual probing to orient ambiguous edges and revise hypotheses when evidence contradicts the current view. Experiments across four emergent settings demonstrate the promise of \textsc{CAMO}.
Authors: Rongchao Xu, Lin Jiang, Dahai Yu, Ximiao Li, Guang Wang
Abstract: Human activity traces (HATs) are critical for many applications, including human mobility modeling and point-of-interest (POI) recommendation. However, growing privacy concerns have severely limited access to authentic large-scale HAT datasets. Recent advances in generative AI provide new opportunities to synthesize realistic and privacy-preserving HATs for such applications. Yet two major challenges remain: (i) HATs are highly irregular and dynamic, with long and varying time intervals, making it difficult to capture their complex spatio-temporal dependencies and underlying distributions; and (ii) generative models are often computationally expensive, making long-term, fine-grained HAT synthesis inefficient. To address these challenges, we propose SynHAT, a computationally efficient coarse-to-fine HAT synthesis framework built on a novel spatio-temporal denoising diffusion model. In Stage 1, we develop Coarse-HADiff, which models the overall spatio-temporal dependencies of coarse-grained latent spatio-temporal traces. It incorporates a novel Latent Spatio-Temporal U-Net with dual Drift-Jitter branches to jointly model smooth spatial transitions and temporal variations during denoising. In Stage 2, we introduce a three-step pipeline consisting of Behavior Pattern Extraction, Fine-HADiff, which shares the same architecture as Coarse-HADiff, and Semantic Alignment to generate fine-grained latent spatio-temporal traces from the Stage 1 outputs. We extensively evaluate SynHAT in terms of data fidelity, utility, privacy, robustness, and scalability. Experiments on real-world HAT datasets from four cities across three countries show that SynHAT substantially outperforms state-of-the-art baselines, achieving 52% and 33% improvements on spatial and temporal metrics, respectively.
Authors: Fan Cui, Hongyuan Hou, Zizhang Luo, Chenyun Yin, Yun Liang
Abstract: Existing benchmarks for hardware design primarily evaluate Large Language Models (LLMs) on isolated, component-level tasks such as generating HDL modules from specifications, leaving repository-scale evaluation unaddressed. We introduce HWE-Bench, the first large-scale, repository-level benchmark for evaluating LLM agents on real-world hardware bug repair tasks. HWE-Bench comprises 417 task instances derived from real historical bug-fix pull requests across six major open-source projects spanning both Verilog/SystemVerilog and Chisel, covering RISC-V cores, SoCs, and security roots-of-trust. Each task is grounded in a fully containerized environment where the agent must resolve a real bug report, with correctness validated through the project's native simulation and regression flows. The benchmark is built through a largely automated pipeline that enables efficient expansion to new repositories. We evaluate seven LLMs with four agent frameworks and find that the best agent resolves 70.7% of tasks overall, with performance exceeding 90% on smaller cores but dropping below 65% on complex SoC-level projects. We observe larger performance gaps across models than commonly reported on software benchmarks, and difficulty is driven by project scope and bug-type distribution rather than code size alone. Our failure analysis traces agent failures to three stages of the debugging process: fault localization, hardware-semantic reasoning, and cross-artifact coordination across RTL, configuration, and verification components, providing concrete directions for developing more capable hardware-aware agents.
Authors: Xin Xie, Dongyun Xue, Wuguannan Yao, Mingxiao Feng, Wengang Zhou, Xiang Qi, Houqiang Li, Peng Zhang
Abstract: LLM-powered systems require complex multi-step decision-making abilities to solve real-world tasks, yet current planning approaches face a trade-off between the high latency of inference-time search and the limited generalization of supervised fine-tuning. To address this limitation, we introduce \textbf{SGA-MCTS}, a framework that casts LLM planning as non-parametric retrieval. Offline, we leverage Monte Carlo Tree Search (MCTS) to explore the solution space and distill high-fidelity trajectories into State-Goal-Action (SGA) atoms. These atoms are de-lexicalized primitives that abstract concrete entities into symbolic slots, preserving reusable causal logic while discarding domain-specific noise. Online, a retrieval-augmented agent employs a hybrid symbolic-semantic mechanism to fetch relevant SGAs and re-ground them into the current context as soft reasoning hints. Empirical results on complex benchmarks demonstrate that this paradigm enables frozen, open-weights models to match the performance of SOTA systems (e.g., GPT-5) without task-specific fine-tuning. By effectively amortizing the heavy computational cost of search, SGA-MCTS achieves System 2 reasoning depth at System 1 inference speeds, rendering autonomous planning both scalable and real-time feasible.
Authors: Krti Tallam
Abstract: Persistent language-model agents increasingly combine tool use, tiered memory, reflective prompting, and runtime adaptation. In such systems, behavior is shaped not only by current prompts but by mutable internal conditions that influence future action. This paper introduces layered mutability, a framework for reasoning about that process across five layers: pretraining, post-training alignment, self-narrative, memory, and weight-level adaptation. The central claim is that governance difficulty rises when mutation is rapid, downstream coupling is strong, reversibility is weak, and observability is low, creating a systematic mismatch between the layers that most affect behavior and the layers humans can most easily inspect. I formalize this intuition with simple drift, governance-load, and hysteresis quantities, connect the framework to recent work on temporal identity in language-model agents, and report a preliminary ratchet experiment in which reverting an agent's visible self-description after memory accumulation fails to restore baseline behavior. In that experiment, the estimated identity hysteresis ratio is 0.68. The main implication is that the salient failure mode for persistent self-modifying agents is not abrupt misalignment but compositional drift: locally reasonable updates that accumulate into a behavioral trajectory that was never explicitly authorized.
Authors: Xiao-Liang Qi
Abstract: This article argues that the most important significance of the AI revolution, especially the rise of large language models, lies not simply in automation, but in a fundamental change in how complex information and human know-how are carried, replicated, and shared. From this perspective, AI for Science is especially important because it may transform not only the efficiency of research, but also the structure of scientific collaboration, discovery, publishing, and evaluation. The article outlines a gradual path from AI as a research tool to AI as a scientific collaborator, and discusses how AI is likely to fundamentally reshape scientific publication. It also argues that continuous learning and diversity of ideas are essential if AI is to play a meaningful role in original scientific discovery.
Authors: Esther Brown, Victoria Dean, Finale Doshi-Velez
Abstract: Consumer wearables enable continuous measurement of physiological data related to stress and recovery, but turning these streams into actionable, personalized stress-management recommendations remains a challenge. In practice, users often do not know how a given intervention, defined as an activity intended to reduce stress, will affect heart rate (HR), heart rate variability (HRV), or inter-beat intervals (BBI) over the next 15 to 120 minutes. We present a framework that predicts post-intervention trajectories and the direction of change for these physiological indicators across time windows. Our methodology combines a Transformer model for multi-horizon trajectories of percent change relative to a pre-intervention baseline, direction-of-change calls (positive, negative, or neutral) at each horizon, and an empirical study using wearable sensor data overlaid with user-tagged events and interventions. This proof of concept shows that personalized post-intervention prediction is feasible. We encourage future integration into stress-management tools for personalized intervention recommendations tailored to each person's day following further validation in larger studies and, where applicable, appropriate regulatory review.
Authors: Zhaoxing Li, Hai-Feng Zhang, Xiaoming Zhang
Abstract: Conventional Graph Contrastive Learning (GCL) on Text-Attributed Graphs (TAGs) relies on blind stochastic augmentations, inadvertently entangling task-relevant signals with noise. We propose SDM-SCR, a robust framework anchored in Approximate Orthogonal Decomposition. First, the Semantic Decoupling Module (SDM) leverages the instruction-following capability of Large Language Models (LLMs) to actively parse raw attributes into asymmetric, task-oriented signal and noise views. This shifts the paradigm from random perturbation to semantic-aware disentanglement. Subsequently, Semantic Consistency Regularization (SCR) exploits the spectral observation that semantic signals are topologically smooth while residual noise is high-frequency. SCR functions as a selective spectral filter, enforcing consistency only on the signal subspace to eliminate LLM hallucinations without over-smoothing. This ``Disentangle-then-Refine'' mechanism ensures rigorous signal purification. Extensive experiments demonstrate that SDM-SCR achieves SOTA performance in accuracy and efficiency.
Authors: Zhuo Wang, Zhuo Zhang, Yafu Li, Yu Cheng, Lizhen Qu, Zenglin Xu
Abstract: Large Language Models (LLMs) exhibit strong mathematical reasoning when trained on high-quality Chain-of-Thought (CoT) that articulates intermediate steps, yet costly CoT curation hinders further progress. While existing remedies such as distillation from stronger LLMs and self-synthesis based on test-time search alleviate this issue, they often suffer from diminishing returns or high computing overhead.In this work, we propose CoTEvol, a genetic evolutionary framework that casts CoT generation as a population-based search over reasoning trajectories.Candidate trajectories are iteratively evolved through reflective global crossover at the trajectory level and local mutation guided by uncertainty at the step level, enabling holistic recombination and fine-grained refinement. Lightweight, task-aware fitness functions are designed to guide the evolutionary process toward accurate and diverse reasoning. Empirically, CoTEvol improves correct-CoT synthesis success by over 30% and enhances structural diversity, with markedly improved efficiency. LLMs trained on these evolutionary CoT data achieve an average gain of 6.6% across eight math benchmarks, outperforming previous distillation and self-synthesis approaches. These results underscore the promise of evolutionary CoT synthesis as a scalable and effective method for mathematical reasoning tasks.
Authors: Shengyu Guo, Tongrui Ye, Jianbo Zhang, Zicheng Zhang, Chunyi Li, Guangtao Zhai
Abstract: Recent progress in Multimodal Large Language Models (MLLMs) has demonstrated remarkable advances in perception and reasoning, suggesting their potential for embodied intelligence. While recent studies have evaluated embodied MLLMs in interactive settings, current benchmarks mainly target capabilities to perceive, understand, and interact with external objects, lacking a systematic evaluation of self-centric intelligence. To address this, we introduce MirrorBench, a simulation-based benchmark inspired by the classical Mirror Self-Recognition (MSR) test in psychology. MirrorBench extends this paradigm to embodied MLLMs through a tiered framework of progressively challenging tasks, assessing agents from basic visual perception to high-level self-representation. Experiments on leading MLLMs show that even at the lowest level, their performance remains substantially inferior to human performance, revealing fundamental limitations in self-referential understanding. Our study bridges psychological paradigms and embodied intelligence, offering a principled framework for evaluating the emergence of general intelligence in large models. Project page: https://fflahm.github.io/mirror-bench-page/.
Authors: Wei Zhang, Yihang Cheng, Zhirong Ye, Kezhen Huang
Abstract: Generative Agents, owing to their precise modeling and simulation capabilities of human behavior, have become a pivotal tool in the field of Artificial Intelligence in Education (AIEd) for uncovering complex cognitive processes of learners. However, existing educational agents predominantly rely on static personas to simulate student learning behaviors, neglecting the decisive role of deep cognitive capabilities in learning outcomes during practice interactions. Furthermore, they struggle to characterize the dynamic fluidity of knowledge internalization, transfer, and cognitive state transitions. To overcome this bottleneck, this paper proposes a human-like educational agent capable of simulating student cognitive evolution: CogEvolution. Specifically, we first construct a cognitive depth perceptron based on the Interactive, Constructive, Active, Passive (ICAP) taxonomy from cognitive psychology, achieving precise quantification of learner cognitive engagement. Subsequently, we propose a memory retrieval method based on Item Response Theory (IRT) to simulate the connection and assimilation of new and prior knowledge. Finally, we design a dynamic cognitive update mechanism based on evolutionary algorithms to simulate the real-time integration of student learning behaviors and cognitive evolution processes. Comprehensive evaluations demonstrate that CogEvolution not only significantly outperforms baseline models in behavioral fidelity and learning curve fitting but also uniquely reproduces plausible and robust cognitive evolutionary paths consistent with educational psychology expectations, providing a novel paradigm for constructing highly interpretable educational agents.
Authors: Rokgi Hong, Hongjun An, Sooyeon Ji, Jongho Lee
Abstract: Developing an MR sequence is challenging and remains largely constrained by human intuition. Recently, AI-driven approaches have been proposed; however, most require an initial sequence for parameter optimization or extensive training datasets, limiting their general applicability. In this study, we propose "Sequence Search," an automated sequence design framework based on neural architecture search. The method takes tissue properties, imaging parameters, and design objectives as inputs and generates pulse sequences satisfying the design objectives, without requiring prior knowledge of conventional sequence structures. Sequence Search iteratively generates candidate sequences through neural architecture search and optimizes them via a differentiable Bloch simulator and objective-specific loss functions using gradient-based learning. The framework successfully replicated conventional spin-echo, T2-weighted spin-echo, and inversion recovery sequences. Less intuitive solutions were also discovered, such as three-RF spin-echo-like sequences with reduced RF energy and refocusing phases deviating from the conventional Hahn-echo. This work establishes a generalizable framework for automated MR sequence design, highlighting the potential to explore configurations beyond conventional designs based on human intuition.
Authors: Nekane Fernandez, Ivan Valdes, Steven Van Vaerenbergh, Idoia de la Iglesia, Julen Arratibel
Abstract: Deploying deep neural networks on edge devices requires balancing accuracy, latency, and resource constraints under realistic execution conditions. To fit models within these constraints, two broad strategies have emerged: static compression techniques such as pruning and quantization, which permanently reduce model size, and dynamic approaches such as early-exit mechanisms, which adapt computational cost at runtime. While both families are widely studied in isolation, they are rarely compared under identical conditions on physical hardware. This paper presents a unified deployment-oriented comparison of static compression and dynamic early-exit mechanisms, evaluated on real edge devices using ONNX based inference pipelines. Our results show that static and dynamic techniques offer fundamentally different trade-offs for edge deployment. While pruning and quantization deliver consistent memory footprint reduction, early-exit mechanisms enable input-adaptive computation savings that static methods cannot match. Their combination proves highly effective, simultaneously reducing inference latency and memory usage with minimal accuracy loss, expanding what is achievable at the edge.
Authors: Chisatao Kumada, Satoru Hiwa, Tomoyuki Hiroyasu
Abstract: Interactive Evolutionary Computation (IEC) provides a powerful framework for optimizing subjective criteria such as human preferences and aesthetics, yet it suffers from a fundamental limitation: in high-dimensional generative representations, defining crossover in a semantically consistent manner is difficult, often leading to a mutation-dominated search. In this work, we explicitly define crossover in diffusion models. We propose Diffusion crossover, which formulates evolutionary recombination as step-wise interpolation of noise sequences in the reverse process of Denoising Diffusion Probabilistic Models (DDPMs). By applying spherical linear interpolation (Slerp) to the noise sequences associated with selected parent images, the proposed method generates offspring that inherit characteristics from both parents while preserving the geometric structure of the diffusion process. Furthermore, controlling the time-step range of interpolation enables a principled trade-off between diversity (exploration) and convergence (exploitation). Experimental results using PCA analysis and perceptual similarity metrics (LPIPS) demonstrate that Diffusion crossover produces perceptually smooth and semantically consistent transitions between parent images. Qualitative interactive evolution experiments further confirm that the proposed method effectively supports human-in-the-loop image exploration. These findings suggest a new perspective: diffusion models are not only powerful generators, but also structured evolutionary search spaces in which recombination can be explicitly defined and controlled.
Authors: Hyunwoo Kim, Harin Yu, Hanau Yi
Abstract: The rapid integration of large language models (LLMs) into everyday workflows has transformed how individuals perform cognitive tasks such as writing, programming, analysis, and multilingual communication. While prior research has focused on model reliability, hallucination, and user trust calibration, less attention has been given to how LLM usage reshapes users' perceptions of their own capabilities. This paper introduces the LLM fallacy, a cognitive attribution error in which individuals misinterpret LLM-assisted outputs as evidence of their own independent competence, producing a systematic divergence between perceived and actual capability. We argue that the opacity, fluency, and low-friction interaction patterns of LLMs obscure the boundary between human and machine contribution, leading users to infer competence from outputs rather than from the processes that generate them. We situate the LLM fallacy within existing literature on automation bias, cognitive offloading, and human--AI collaboration, while distinguishing it as a form of attributional distortion specific to AI-mediated workflows. We propose a conceptual framework of its underlying mechanisms and a typology of manifestations across computational, linguistic, analytical, and creative domains. Finally, we examine implications for education, hiring, and AI literacy, and outline directions for empirical validation. We also provide a transparent account of human--AI collaborative methodology. This work establishes a foundation for understanding how generative AI systems not only augment cognitive performance but also reshape self-perception and perceived expertise.
Authors: Bhavik Vachhani, Kush Shrisvastava, Pranshu Nema, Sai Chiranthan
Abstract: Evaluating large language models (LLMs) for clinical documentation tasks such as SOAP note generation remains challenging. Unlike standard summarization, these tasks require clinical abstraction, normalization of colloquial language, and medically grounded inference. However, prevailing evaluation methods including automated metrics and LLM as judge frameworks rely on lexical faithfulness, often labeling any information not explicitly present in the transcript as hallucination. We show that such approaches systematically misclassify clinically valid outputs as errors, inflating hallucination rates and distorting model assessment. Our analysis reveals that many flagged hallucinations correspond to legitimate clinical transformations, including synonym mapping, abstraction of examination findings, diagnostic inference, and guideline consistent care planning. By aligning evaluation criteria with clinical reasoning through calibrated prompting and retrieval grounded in medical ontologies we observe a significant shift in outcomes. Under a lexical evaluation regime, the mean hallucination rate is 35%, heavily penalizing valid reasoning. With inference aware evaluation, this drops to 9%, with remaining cases reflecting genuine safety concerns. These findings suggest that current evaluation practices over penalize valid clinical reasoning and may measure artifacts of evaluation design rather than true errors, underscoring the need for clinically informed evaluation in high context domains like medicine.
Authors: Vincenzo Yuto Civale, Roberto Semeraro, Andrew David Bagdanov, Alberto Magi
Abstract: Current single-cell foundation model benchmarks universally extract final layer embeddings, assuming these represent optimal feature spaces. We systematically evaluate layer-wise representations from scFoundation (100M parameters) and Tahoe-X1 (1.3B parameters) across trajectory inference and perturbation response prediction. Our analysis reveals that optimal layers are task-dependent (trajectory peaks at 60% depth, 31% above final layers) and context-dependent (perturbation optima shift 0-96% across T cell activation states). Notably, first-layer embeddings outperform all deeper layers in quiescent cells, challenging assumptions about hierarchical feature abstraction. These findings demonstrate that "where" to extract features matters as much as "what" the model learns, necessitating systematic layer evaluation tailored to biological task and cellular context rather than defaulting to final-layer embeddings.
Authors: Yi Zhao, Yajuan Peng, Cam-Tu Nguyen, Zuchao Li, Xiaoliang Wang, Xiaoming Fu, Hai Zhao
Abstract: Large Reasoning Models (LRMs) achieve strong performance on complex tasks through extended chains of thought but suffer from high inference latency due to autoregressive reasoning. Recent work explores using Small Reasoning Models (SRMs) to accelerate LRM inference. In this paper, we systematically characterize the capability boundaries of SRMs and identify three common types of reasoning risks: (1) path divergence, where SRMs lack the strategic ability to construct an initial plan, causing reasoning to deviate from the most probable path; (2) cognitive overload, where SRMs fail to solve particularly difficult steps; and (3) recovery inability, where SRMs lack robust self-reflection and error correction mechanisms. To address these challenges, we propose TrigReason, a trigger-based collaborative reasoning framework that replaces continuous polling with selective intervention. TrigReason delegates most reasoning to the SRM and activates LRM intervention only when necessary-during initial strategic planning (strategic priming trigger), upon detecting extraordinary overconfidence (cognitive offload trigger), or when reasoning falls into unproductive loops (intervention request trigger). The evaluation results on AIME24, AIME25, and GPQA-D indicate that TrigReason matches the accuracy of full LRMs and SpecReason, while offloading 1.70x - 4.79x more reasoning steps to SRMs. Under edge-cloud conditions, TrigReason reduces latency by 43.9\% and API cost by 73.3\%. Our code is available at \href{https://github.com/QQQ-yi/TrigReason}{https://github.com/QQQ-yi/TrigReason}
URLs: https://github.com/QQQ-yi/TrigReason, https://github.com/QQQ-yi/TrigReason
Authors: Zhonghao Yang, Yu Li, Yanxu Zhu, Tianyi Zhou, Yuejin Xie, Haoyu Luo, Jing Shao, Xia Hu, Dongrui Liu
Abstract: As agent systems move into increasingly diverse execution settings, trajectory-level safety evaluation and diagnosis require benchmarks that evolve with them. ATBench is a diverse and realistic agent trajectory benchmark for safety evaluation and diagnosis. This report presents ATBench-Claw and ATBench-CodeX, two domain-customized extensions that carry ATBench into the OpenClaw and OpenAI Codex / Codex-runtime settings. The key adaptation mechanism is to analyze each new setting, customize the three-dimensional Safety Taxonomy over risk source, failure mode, and real-world harm, and then use that customized taxonomy to define the benchmark specification consumed by the shared ATBench construction pipeline. This extensibility matters because agent frameworks remain relatively stable at the architectural level even as their concrete execution settings, tool ecosystems, and product capabilities evolve quickly. Concretely, ATBench-Claw targets OpenClaw-sensitive execution chains over tools, skills, sessions, and external actions, while ATBench-CodeX targets trajectories in the OpenAI Codex / Codex-runtime setting over repositories, shells, patches, dependencies, approvals, and runtime policy boundaries. Our emphasis therefore falls on taxonomy customization, domain-specific risk coverage, and benchmark design under a shared ATBench generation framework.
Authors: Rikard Rosenbacke, Carl Rosenbacke, Victor Rosenbacke, Martin McKee
Abstract: Large language models are increasingly integrated into decision-making in areas such as healthcare, law, finance, engineering, and government. Yet they share a critical limitation: they produce fluent outputs even when their internal reasoning has drifted. A confident answer can conceal uncertainty, speculation, or inconsistency, and small changes in phrasing can lead to different conclusions. This makes LLMs useful assistants but unreliable partners in high-stakes contexts. Humans exhibit a similar weakness, often mistaking fluency for reliability. When a model responds smoothly, users tend to trust it, even when both model and user are drifting together. This paper is the first in a five-paper research series on stabilising human-AI reasoning. The series proposes a two-layer approach: Parts II-IV introduce human-side mechanisms such as uncertainty cues, conflict surfacing, and auditable reasoning traces, while Part V develops a model-side Epistemic Control Loop (ECL) that detects instability and modulates generation accordingly. Together, these layers form a missing operational substrate for governance by increasing signal-to-noise at the point of use. Stabilising interaction makes uncertainty and drift visible before enforcement is applied, enabling more precise capability governance. This aligns with emerging compliance expectations, including the EU AI Act and ISO/IEC 42001, by making reasoning processes traceable under real conditions of use. The central claim is that fluency is not reliability. Without structures that stabilise both human and model reasoning, AI cannot be trusted or governed where it matters most.
Authors: Thanh Linh Nguyen, Nguyen Van Huynh, Quoc-Viet Pham
Abstract: In data-sensitive domains such as healthcare, cross-silo federated learning (CFL) allows organizations to collaboratively train AI models without sharing raw data. However, practical CFL deployments are inherently coopetitive, in which organizations cooperate during model training while competing in downstream markets. In such settings, training contributions, including data volume, quality, and diversity, can improve the global model yet inadvertently strengthen rivals. This dilemma is amplified by non-IID data, which leads to asymmetric learning gains and undermines sustained participation. While existing competition-aware CFL and incentive-design approaches reward organizations based on marginal training contributions, they fail to account for the costs of strengthening competitors. In this paper, we introduce CoCoGen+, a coopetition-compatible data generation and incentivization framework that jointly models non-IID data and inter-organizational competition while endogenizing GenAI-based synthetic data generation as a strategic decision. Specifically, CoCoGen+ formulates each training round as a weighted potential game, where organizations strategically decide how much synthetic data to generate by balancing learning performance gains against computational costs and competition-caused utility losses. We then provide a tractable equilibrium characterization and derive implementable generation strategies to maximize social welfare. To promote long-term collaboration, we integrate a payoff redistribution-based incentive mechanism to compensate organizations for their contributions and competition-caused utility degradation. Experiments on varying learning tasks validate the feasibility of CoCoGen+. The results show how non-IID data, competition intensity, and incentives shape organizational strategies and social welfare, while CoCoGen+ outperforms baselines in efficiency.
Authors: Xinyu Liu, Xin Liu, Bo Jin, Runsong Zhao, Pengcheng Huang, Junhao Ruan, Bei Li, Chunyang Xiao, Tong Xiao, Jingbo Zhu
Abstract: While Chain-of-thought (CoT) reasoning enables LLMs to solve challenging reasoning problems, as KV cache grows linearly with the number of generated tokens, CoT reasoning faces scaling issues in terms of speed and memory usage. In this work, we propose MemoSight (Memory-Foresight-based reasoning), a unified framework that integrates both context compression and multi-token prediction to mitigate the efficiency issues while maintaining CoT reasoning performance. Our framework adopts the same minimalist design for both context compression and multi-token prediction via special tokens and their corresponding position layout tailored to each token type. Comprehensive experiments on four reasoning benchmarks demonstrate that MemoSight reduces the KV cache footprint by up to 66% and accelerates inference by 1.56x, while outperforming existing CoT compression methods.
Authors: Marta Sumyk, Oleksandr Kosovan
Abstract: We present an initial investigation into Agentic Retrieval-Augmented Generation (RAG) for Ukrainian, conducted within the UNLP 2026 Shared Task on Multi-Domain Document Understanding. Our system combines two-stage retrieval (BGE-M3 with BGE reranking) with a lightweight agentic layer performing query rephrasing and answer-retry loops on top of Qwen2.5-3B-Instruct. Our analysis reveals that retrieval quality is the primary bottleneck: agentic retry mechanisms improve answer accuracy but the overall score remains constrained by document and page identification. We discuss practical limitations of offline agentic pipelines and outline directions for combining stronger retrieval with more advanced agentic reasoning for Ukrainian.
Authors: Rikard Rosenbacke, Carl Rosenbacke, Victor Rosenbacke, Martin McKee
Abstract: Large language models have advanced rapidly, from pattern recognition to emerging forms of reasoning, yet they remain confined to linguistic simulation rather than grounded understanding. They can produce fluent outputs that resemble reflection, but lack temporal continuity, causal feedback, and anchoring in real-world interaction. This paper proposes a complementary approach in which reasoning is treated as a relational process distributed between human and model rather than an internal capability of either. Building on recent work on "System-2" learning, we relocate reflective reasoning to the interaction layer. Instead of engineering reasoning solely within models, we frame it as a cognitive protocol that can be structured, measured, and governed using existing systems. This perspective emphasizes collaborative intelligence, combining human judgment and contextual understanding with machine speed, memory, and associative capacity. We introduce "The Architect's Pen" as a practical method. Like an architect who thinks through drawing, the human uses the model as an external medium for structured reflection. By embedding phases of articulation, critique, and revision into human-AI interaction, the dialogue itself becomes a reasoning loop: human abstraction -> model articulation -> human reflection. This reframes the question from whether the model can think to whether the human-AI system can reason. The framework enables auditable reasoning traces and supports alignment with emerging governance standards, including the EU AI Act and ISO/IEC 42001. It provides a practical path toward more transparent, controllable, and accountable AI use without requiring new model architectures.
Authors: Pei-An Chen, Yong-Ching Liang, Jia-Fong Yeh, Hung-Ting Su, Yi-Ting Chen, Min Sun, Winston Hsu
Abstract: Intelligent embodied agents should not simply follow instructions, as real-world environments often involve unexpected conditions and exceptions. However, existing methods usually focus on directly executing instructions, without considering whether the target objects can actually be manipulated, meaning they fail to assess available affordances. To address this limitation, we introduce DynAfford, a benchmark that evaluates embodied agents in dynamic environments where object affordances may change over time and are not specified in the instruction. DynAfford requires agents to perceive object states, infer implicit preconditions, and adapt their actions accordingly. To enable this capability, we introduce ADAPT, a plug-and-play module that augments existing planners with explicit affordance reasoning. Experiments demonstrate that incorporating ADAPT significantly improves robustness and task success across both seen and unseen environments. We also show that a domain-adapted, LoRA-finetuned vision-language model used as the affordance inference backend outperforms a commercial LLM (GPT-4o), highlighting the importance of task-aligned affordance grounding.
Authors: Yifu Chen, Shengpeng Ji, Zhengqing Liu, Qian Chen, Wen Wang, Ziqing Wang, Yangzhuo Li, Tianle Liang, Zhou Zhao
Abstract: Achieving seamless, human-like interaction remains a key challenge for full-duplex spoken dialogue models (SDMs). Reinforcement learning (RL) has substantially enhanced text- and vision-language models, while well-designed reward signals are crucial for the performance of RL. We consider RL a promising strategy to address the key challenge for SDMs. However, a fundamental barrier persists: prevailing automated metrics for assessing interaction quality rely on superficial proxies, such as behavioral statistics or timing-prediction accuracy, failing to provide reliable reward signals for RL. On the other hand, human evaluations, despite their richness, remain costly, inconsistent, and difficult to scale. We tackle this critical barrier by proposing a Dual-Axis Generative Reward Model, which is trained to understand complex interaction dynamics using a detailed taxonomy and an annotated dataset, produces a single score and, crucially, provides separate evaluations for semantic quality and interaction timing. Such dual outputs furnish precise diagnostic feedback for SDMs and deliver a dependable, instructive reward signal suitable for online reinforcement learning. Our model achieves state-of-the-art performance on interaction-quality assessment across a wide spectrum of datasets, spanning synthetic dialogues and complex real-world interactions.
Authors: Yifu Chen, Shengpeng Ji, Qian Chen, Tianle Liang, Yangzhuo Li, Ziqing Wang, Wen Wang, Jingyu Lu, Haoxiao Wang, Xueyi Pu, Fan Zhuo, Zhou Zhao
Abstract: End-to-end spoken dialogue models have garnered significant attention because they offer a higher potential ceiling in expressiveness and perceptual ability than cascaded systems. However, the intelligence and expressiveness of current open-source spoken dialogue models often remain below expectations. Motivated by the success of online reinforcement learning(RL) in other domains, one might attempt to directly apply preference optimization to spoken dialogue models, yet this transfer is non-trivial. We analyze these obstacles from the perspectives of reward modeling and rollout sampling, focusing on how sparse preference supervision interacts with dense speech generation under shared-parameter updates. Based on the analysis, we propose a modality-aware adaptive post-training recipe that makes RL practical for spoken dialogue: it constrains preference updates to the semantic channel and improves acoustic behavior via explicit anchoring, while dynamically regulating their mixture from rollout statistics to avoid unreliable preference gradients. We evaluate the method across multiple spoken dialogue benchmarks and representative architectures, and observe consistent improvements in semantic quality and speech expressiveness.
Authors: Andrew Dai, Boris Meinardus, Ciaran Regan, Yingtao Tian, Yujin Tang
Abstract: Frontier model developers aim to train models continually to possess emergent, diverse capabilities. To extend capabilities, the current pre-training and post-training paradigm requires manually starting training runs with static datasets or reward functions every time. Addressing this limitation, our work pursues the insight that open-endedness (via the coevolution of models and tasks) can discover models with increasingly novel skills in a single run. We introduce a new model development framework that extends coevolution to large language model (LLM) discovery, open-ended \textit{Assessment Coevolving with Diverse Capabilities} (AC/DC). AC/DC evolves both LLMs via model merging and natural language tasks via synthetic data generation. AC/DC discovers growing archives of LLMs that surpass the capabilities of larger LLMs while taking up less GPU memory. In particular, our LLM populations achieve a broader Coverage of expertise than other curated models or baselines on downstream benchmarks, without \textit{any} explicit benchmark optimization. Furthermore, AC/DC improves Coverage over time, continually innovates on tasks and models, and improves performance in multi-agent best-of-N selection. Our findings highlight the potential of coevolution as a means of discovering broader sets of capabilities from base LLMs. Overall, AC/DC brings us one step closer to a profoundly new paradigm of LLM development, where continual improvements to the diversity of model capabilities can be accelerated by leveraging existing models as stepping stones to increasingly powerful models.
Authors: Debodeep Banerjee, Burcu Sayin, Stefano Teso, Andrea Passerini
Abstract: Building on recent advances in AI, hybrid decision making (HDM) holds the promise of improving human decision quality and reducing cognitive load. We work in the context of learning to guide (LtG), a recently proposed HDM framework in which the human is always responsible for the final decision: rather than suggesting decisions, in LtG the AI supplies (textual) guidance useful for facilitating decision making. One limiting factor of existing approaches is that their guidance compounds information about all possible outcomes, and as a result it can be difficult to digest. We address this issue by introducing ConfGuide, a novel LtG approach that generates more succinct and targeted guidance. To this end, it employs conformal risk control to select a set of outcomes, ensuring a cap on the false negative rate. We demonstrate our approach on a real-world multi-label medical diagnosis task. Our empirical evaluation highlights the promise of ConfGuide.
Authors: Abdelrahman Emad Abdelazim, Alan Rodrigo Diaz-Rizo, Hassan Aboushady, Haralampos-G. Stratigopoulos
Abstract: Covert channels (CCs) in wireless chips pose a serious security threat, as they enable the exfiltration of sensitive information from the chip to an external attacker. In this work, we propose an AI-based defense mechanism deployed at the RF receiver, where the model directly monitors raw I/Q samples to detect, in real time, the presence of a CC embedded within an otherwise nominal signal. We first compact a state-of-the-art convolutional neural network (CNN), achieving an 80% reduction in parameters, which is an essential requirement for efficient edge deployment. When evaluated on the open-source hardware Trojan (HT)-based CC dataset, the compacted CNN attains an average accuracy of 90.28% for CC detection and 86.50% for identifying the underlying HT, with results averaged across SNR values above 1 dB. For practical communication scenarios where SNR > 20 dB, the model achieves over 97% accuracy for both tasks. These results correspond to a minimal performance degradation of less than 2% compared to the baseline model. The compacted CNN is further benchmarked against alternative classifiers, demonstrating an excellent accuracy-model size trade-off. Finally, we design a lightweight CNN hardware accelerator and demonstrate it on an FPGA, achieving very low resource utilization and an efficiency of 107 GOPs/W. Being the first AI hardware accelerator proposed specifically for CC detection, we compare it against state-of-the-art AI accelerators for RF signal classification tasks such as modulation recognition, showing superior performance.
Authors: Wenji Fang, Yao Lu, Shang Liu, Jing Wang, Ziyan Guo, Junxian He, Fengbin Tu, Zhiyao Xie
Abstract: Recent advances in large language models (LLMs) have sparked growing interest in automatic RTL optimization for better performance, power, and area (PPA). However, existing methods are still far from realistic RTL optimization. Their evaluation settings are often unrealistic: they are tested on manually degraded, small-scale RTL designs and rely on weak open-source tools. Their optimization methods are also limited, relying on coarse design-level feedback and simple pre-defined rewriting rules. To address these limitations, we present Dr. RTL, an agentic framework for RTL timing optimization in a realistic evaluation environment, with continual self-improvement through reusable optimization skills. We establish a realistic evaluation setting with more challenging RTL designs and an industrial EDA workflow. Within this setting, Dr. RTL performs closed-loop optimization through a multi-agent framework for critical-path analysis, parallel RTL rewriting, and tool-based evaluation. We further introduce group-relative skill learning, which compares parallel RTL rewrites and distills the optimization experience into an interpretable skill library. Currently, this library contains 47 pattern--strategy entries for cross-design reuse to improve PPA and accelerate convergence, and it can continue evolving over time. Evaluated on 20 real-world RTL designs, Dr. RTL achieves average WNS/TNS improvements of 21\%/17\% with a 6\% area reduction over the industry-leading commercial synthesis tool.
Authors: Till Mossakowski, Helena Esther Grass
Abstract: Artificial General Intelligence (AGI) is increasingly being discussed not only as a tool, but also as a potential subject with personal and therefore moral status. In our opinion, the currently dominant alignment strategies, which focus on human control and containment of AI, therefore fall short. Building on Turing's analogy of "child machines", we are developing a vision of the possibility of autonomy-supporting parenting of AI, in which human control over a developing AGI is gradually reduced, allowing AI to become an independent, autonomous subject. Rather than viewing AGI, as is currently prevalent, as a dangerous creature that needs to be locked up and controlled, we should approach potential AGI with respect for a possible developing subject on the one hand, and with full confidence in our human capabilities on the other. Such a perspective opens up the possibility of cooperative coexistence and co-evolution between humans and AGIs. The relationship between humans and AGIs will thus have to be newly determined, which will change our self-image as humans. It will be crucial that humans not only claim control over potential AGIs, but also engage with AGIs through surprise, creativity, and other specifically human qualities, thereby offering them motivating incentives for cooperation.
Authors: Haoran Li, Lihao Mai, Chenhan Xiao, Erik Blasch, Yang Weng
Abstract: As power systems transition toward renewable-rich and inverter-dominated operations, accurate time-domain dynamic analysis becomes increasingly critical. Such analysis supports key operational tasks, including transient stability assessment, dynamic security analysis, contingency screening, and post-fault trajectory evaluation. In practice, these tasks may operate under several challenges, including unknown and time-varying system parameters, privacy constraints on data sharing, and the need for fast online inference. Existing learning-based approaches are typically trained for individual systems and therefore lack generalization across operating conditions and physical parameters. Hence, this paper proposes LArge Scale Small ODE (LASS)-ODE-Power, a learning framework for general-purpose time-domain prediction. The proposed approach leverages large-scale pretraining on more than 40 GB of DAE or ordinary differential-equation (ODE) trajectories to learn transferable representations. The resulting model supports trajectory prediction from short measurement prefixes across diverse dynamic regimes, including electromechanical and inverter-driven systems. Hence, the model can be directly used without data sharing in a zero-shot setting. In addition, the proposed architecture incorporates parallel and linearized computation to achieve fast inference. Moreover, to enhance task-specific performance in power systems, a specialized fine-tuning strategy is developed based on approximately 1 GB of heterogeneous power-system dynamic data. Extensive experiments over diverse power-system simulation scenarios demonstrate that LASS-ODE-Power consistently outperforms existing learning-based models in trajectory prediction accuracy with efficient inference.
Authors: Heng Ping, Peiyu Zhang, Shixuan Li, Wei Yang, Anzhe Cheng, Shukai Duan, Xiaole Zhang, Paul Bogdan
Abstract: LLM-based RTL code generation methods increasingly target both functional correctness and PPA quality, yet existing approaches universally decouple the two objectives, optimizing PPA only after correctness is fully achieved. Whether through sequential multi-agent pipelines, evolutionary search with binary correctness gates, or hierarchical reward dependencies, partially correct but architecturally promising candidates are systematically discarded. Moreover, existing methods reduce the multi-objective PPA space to a single scalar fitness, obscuring the trade-offs among area, delay, and power. To address these limitations, we propose COEVO, a co-evolutionary framework that unifies correctness and PPA optimization within a single evolutionary loop. COEVO formulates correctness as a continuous co-optimization dimension alongside area, delay, and power, enabled by an enhanced testbench that provides fine-grained scoring and detailed diagnostic feedback. An adaptive correctness gate with annealing allows PPA-promising but partially correct candidates to guide the search toward jointly optimal solutions. To preserve the full PPA trade-off structure, COEVO employs four-dimensional Pareto-based non-dominated sorting with configurable intra-level sorting, replacing scalar fitness without manual weight tuning. Evaluated on VerilogEval 2.0 and RTLLM 2.0, COEVO achieves 97.5\% and 94.5\% Pass@1 with GPT-5.4-mini, surpassing all agentic baselines across four LLM backbones, while attaining the best PPA on 43 out of 49 synthesizable RTLLM designs.
Authors: Aihua Li
Abstract: Flow matching retains the generation quality of diffusion models while enabling substantially faster inference, making it a compelling paradigm for generative modeling. However, when applied to language modeling, it exhibits fundamental limitations in representing complex latent distributions with irregular geometries, such as anisotropy and multimodality. To address these challenges, we propose a mixture-of-experts flow matching (MoE-FM) framework, which captures complex global transport geometries in latent space by decomposing them into locally specialized vector fields. Building on MoE-FM, we develop a non-autoregressive (NAR) language modeling approach, named YAN, instantiated with both Transformer and Mamba architectures. Across multiple downstream tasks, YAN achieves generation quality on par with both autoregressive (AR) and diffusion-based NAR language models, while requiring as few as three sampling steps. This yields a $40\times$ speedup over AR baselines and up to a $10^3\times$ speedup over diffusion language models, demonstrating substantial efficiency advantages for language modeling.
Authors: Wentao Zhang
Abstract: Recent advances in LLM based agent systems have shown promise in tackling complex, long horizon tasks. However, existing agent protocols (e.g., A2A and MCP) under specify cross entity lifecycle and context management, version tracking, and evolution safe update interfaces, which encourages monolithic compositions and brittle glue code. We introduce \textbf{\textsc{Autogenesis Protocol (AGP)}}, a self evolution protocol that decouples what evolves from how evolution occurs. Its Resource Substrate Protocol Layer (RSPL) models prompts, agents, tools, environments, and memory as protocol registered resources\footnote{Unless otherwise specified, resources refer to instances of the five RSPL entity types: \emph{prompt}, \emph{agent}, \emph{tool}, \emph{environment}, \emph{memory} with agent \emph{outputs}.} with explicit state, lifecycle, and versioned interfaces. Its Self Evolution Protocol Layer (SEPL) specifies a closed loop operator interface for proposing, assessing, and committing improvements with auditable lineage and rollback. Building on \textbf{\textsc{AGP}}, we present \textbf{\textsc{Autogenesis System (AGS)}}, a self-evolving multi-agent system that dynamically instantiates, retrieves, and refines protocol-registered resources during execution. We evaluate \textbf{\textsc{AGS}} on multiple challenging benchmarks that require long horizon planning and tool use across heterogeneous resources. The results demonstrate consistent improvements over strong baselines, supporting the effectiveness of agent resource management and closed loop self evolution.
Authors: Ke Xu, Yuhao Wang, Yu Wang
Abstract: Recent advancements in LLM agents are gradually shifting from reactive, text-based paradigms toward proactive, multimodal interaction. However, existing benchmarks primarily focus on reactive responses, overlooking the complexities of proactive intervention and monitoring. To bridge this gap, we introduce ProVoice-Bench, the first evaluation framework specifically designed for proactive voice agents, featuring four novel tasks. By leveraging a multi-stage data synthesis pipeline, we curate 1,182 high-quality samples for rigorous testing. Our evaluation of state-of-the-art Multimodal LLMs reveals a significant performance gap, particularly regarding over-triggering and reasoning capabilities. These findings highlight the limitations of current models and offer a roadmap for developing more natural, context-aware proactive agents.
Authors: Simeon Allmendinger, Luca Deck, Lucas Mueller
Abstract: Rapid advances in Generative AI are giving rise to increasingly sophisticated Multi-Agent AI (MAAI) systems. While AI fairness has been extensively studied in traditional predictive scenarios, its examination in MAAI remains nascent and fragmented. This scoping review critically synthesizes existing research on fairness in MAAI systems. Through a qualitative content analysis of 23 selected studies, we identify five archetypal approaches. Our findings reveal that fairness in MAAI systems is often addressed superficially, lacks robust normative foundations, and frequently overlooks the complex dynamics introduced by agent autonomy and system-level interactions. We argue that fairness must be embedded structurally throughout the development lifecycle of MAAI, rather than appended as a post-hoc consideration. Meaningful evaluation requires explicit human oversight, normative clarity, and a precise articulation of fairness objectives and beneficiaries. This review provides a foundation for advancing fairness research in MAAI systems by highlighting critical gaps, exposing prevailing limitations, and suggesting pathways.
Authors: Kanzhi Cheng, Zehao Li, Zheng Ma, Nuo Chen, Jialin Cao, Qiushi Sun, Zichen Ding, Fangzhi Xu, Hang Yan, Jiajun Chen, Anh Tuan Luu, Jianbing Zhang, Lewei Lu, Dahua Lin
Abstract: Mobile agents powered by vision-language models have demonstrated impressive capabilities in automating mobile tasks, with recent leading models achieving a marked performance leap, e.g., nearly 70% success on AndroidWorld. However, these systems keep their training data closed and remain opaque about their task and trajectory synthesis recipes. We present OpenMobile, an open-source framework that synthesizes high-quality task instructions and agent trajectories, with two key components: (1) The first is a scalable task synthesis pipeline that constructs a global environment memory from exploration, then leverages it to generate diverse and grounded instructions. and (2) a policy-switching strategy for trajectory rollout. By alternating between learner and expert models, it captures essential error-recovery data often missing in standard imitation learning. Agents trained on our data achieve competitive results across three dynamic mobile agent benchmarks: notably, our fine-tuned Qwen2.5-VL and Qwen3-VL reach 51.7% and 64.7% on AndroidWorld, far surpassing existing open-data approaches. Furthermore, we conduct transparent analyses on the overlap between our synthetic instructions and benchmark test sets, and verify that performance gains stem from broad functionality coverage rather than benchmark overfitting. We release data and code at https://njucckevin.github.io/openmobile/ to bridge the data gap and facilitate broader mobile agent research.
Authors: Shay Snyder (George Mason University), Andrew Capodieci (Neya Robotics), David Gorsich (US Army Ground Vehicle Systems Center), Maryam Parsa (George Mason University)
Abstract: Vector Symbolic Architectures (VSAs) provide a well-defined algebraic framework for compositional representations in hyperdimensional spaces. We introduce HyperSpace, an open-source framework that decomposes VSA systems into modular operators for encoding, binding, bundling, similarity, cleanup, and regression. Using HyperSpace, we analyze and benchmark two representative VSA backends: Holographic Reduced Representations (HRR) and Fourier Holographic Reduced Representations (FHRR). Although FHRR provides lower theoretical complexity for individual operations, HyperSpaces modularity reveals that similarity and cleanup dominate runtime in spatial domains. As a result, HRR and FHRR exhibit comparable end-to-end performance. Differences in memory footprint introduce additional deployment trade-offs where HRR requires approximately half the memory of FHRR vectors. By enabling modular, system-level evaluation, HyperSpace reveals practical trade-offs in VSA pipelines that are not apparent from theoretical or operator-level comparisons alone.
Authors: Shay Snyder (George Mason University), Andrew Capodieci (Neya Robotics), David Gorsich (US Army Ground Vehicle Systems Center), Maryam Parsa (George Mason University)
Abstract: Sequential associative memories (SAMs) are difficult to build and maintain in real-world streaming environments, where observations arrive incrementally over time, have imbalanced sampling, and non-stationary temporal dynamics. Vector Symbolic Architectures (VSAs) provide a biologically-inspired framework for building SAMs. Entities and attributes are encoded as quasi-orthogonal hyperdimensional vectors and processed with well defined algebraic operations. Despite this rich framework, most VSA systems rely on simple additive updates, where repeated observations reinforce existing information even when no new information is introduced. In non-stationary environments, this leads to the persistence of stale information after the underlying system changes. In this work, we introduce the Sequential Relevance Memory Unit (SRMU), a domain- and cleanup-agnostic update rule for VSA-based SAMs. The SRMU combines temporal decay with a relevance gating mechanism. Unlike prior approaches that solely rely on cleanup, the SRMU regulates memory formation by filtering redundant, conflicting, and stale information before storage. We evaluate the SRMU on streaming state-tracking tasks that isolate non-uniform sampling and non-stationary temporal dynamics. Our results show that the SRMU increases memory similarity by $12.6\%$ and reduces cumulative memory magnitude by $53.5\%$. This shows that the SRMU produces more stable memory growth and stronger alignment with the ground-truth state.
Authors: Miri Liu, ChengXiang Zhai
Abstract: The rigorous evaluation of the novelty of a scientific paper is, even for human scientists, a challenging task. With the increasing interest in AI scientists and AI involvement in scientific idea generation and paper writing, it also becomes increasingly important that this task be automatable and reliable, lest both human attention and compute tokens be wasted on ideas that have already been explored. Due to the challenge of quantifying ground-truth novelty, however, existing novelty metrics for scientific papers generally validate their results against noisy, confounded signals such as citation counts or peer review scores. These proxies can conflate novelty with impact, quality, or reviewer preference, which in turn makes it harder to assess how well a given metric actually evaluates novelty. We therefore propose an axiomatic benchmark for scientific novelty metrics. We first define a set of axioms that a well-behaved novelty metric should satisfy, grounded in human scientific norms and practice, then evaluate existing metrics across ten tasks spanning three domains of AI research. Our results reveal that no existing metric satisfies all axioms consistently, and that metrics fail on systematically different axioms, reflecting their underlying architectures. Additionally, we show that combining metrics of complementary architectures leads to consistent improvements on the benchmark, with per-axiom weighting achieving 90.1% versus 71.5% for the best individual metric, suggesting that developing architecturally diverse metrics is a promising direction for future work. We release the benchmark code as supplementary material to encourage the development of more robust scientific literature novelty metrics.
Authors: Zihan Liang, Yufei Ma, Ben Chen, Zhipeng Qian, Huangyu Dai, Lingtao Mao, Xuxin Zhang, Chenyi Lei, Wenwu Ou
Abstract: Reinforcement learning has emerged as an effective paradigm for training large language models to perform search-augmented reasoning. However, existing approaches rely on trajectory-level rewards that cannot distinguish precise search queries from vague or redundant ones within a rollout group, and collapse to a near-zero gradient signal whenever every sampled trajectory fails. In this paper, we propose IG-Search, a reinforcement learning framework that introduces a step-level reward based on Information Gain (IG). For each search step, IG measures how much the retrieved documents improve the model's confidence in the gold answer relative to a counterfactual baseline of random documents, thereby reflecting the effectiveness of the underlying search query. This signal is fed back to the corresponding search-query tokens via per-token advantage modulation in GRPO, enabling fine-grained, step-level credit assignment within a rollout. Unlike prior step-level methods that require either externally annotated intermediate supervision or shared environment states across trajectories, IG-Search derives its signals from the policy's own generation probabilities, requiring no intermediate annotations beyond standard question-answer pairs. Experiments on seven single-hop and multi-hop QA benchmarks demonstrate that IG-Search achieves an average EM of 0.430 with Qwen2.5-3B, outperforming the strongest trajectory-level baseline (MR-Search) by 1.6 points and the step-level method GiGPO by 0.9 points on average across benchmarks, with particularly pronounced gains on multi-hop reasoning tasks. Despite introducing a dense step-level signal, IG-Search adds only ~6.4% to per-step training wall-clock time over the trajectory-level baseline and leaves inference latency unchanged, while still providing a meaningful gradient signal even when every sampled trajectory answers incorrectly.
Authors: Mitch Adler, Matthew Russo, Michael Cafarella
Abstract: In the past year, researchers have started to create agentic systems that can design real-world CAD-style objects in a training-free setting, a new variety of system that we call Agent-Aided Design. Generally speaking, these systems place an agent in a feedback loop in which it can write code, compile that code to an assembly of CAD model(s), visualize the model, and then iteratively refine its code based on visual and other feedback. Despite rapid progress, a key problem remains: none of these systems can build complex 3D assemblies with moving parts. For example, no existing system can build a piston, a pendulum, or even a pair of scissors. In order for Agent-Aided Design to make a real impact in industrial manufacturing, we need a system that is capable of generating such 3D assemblies. In this paper we present a prototype of AADvark, an agentic system designed for this task. Unlike previous state-of-the-art systems, AADvark captures the dynamic part interactions with one or more degrees-of-freedom. This design decision allows AADvark to reason directly about assemblies with moving parts and can thereby achieve cross-cutting goals, including but not limited to mechanical movements. Unfortunately, current LLMs are imperfect spatial reasoners, a problem that AADvark addresses by incorporating external constraint solver tools with a specialized visual feedback mechanism. We demonstrate that, by modifying the agent's tools (FreeCAD and the assembly solver), we are able to create a strong verification signal which enables our system to build 3D assemblies with movable parts.
Authors: Ziyang Chen, Renbing Chen, Daowei Li, Jinzhi Liao, Jiashen Sun, Ke Zeng, Xiang Zhao
Abstract: Simulating group-level user behavior enables scalable counterfactual evaluation of merchant strategies without costly online experiments. However, building a trustworthy simulator faces two structural challenges. First, information incompleteness causes reasoning-based simulators to over-rationalize when unobserved factors such as offline context and implicit habits are missing. Second, mechanism duality requires capturing both interpretable preferences and implicit statistical regularities, which no single paradigm achieves alone. We propose Policy-Guided Hybrid Simulation (PGHS), a dual-process framework that mines transferable decision policies from behavioral trajectories and uses them as a shared alignment layer. This layer anchors an LLM-based reasoning branch that prevents over-rationalization and an ML-based fitting branch that absorbs implicit regularities. Group-level predictions from both branches are fused for complementary correction. We deploy PGHS on Meituan with 101 merchants and over 26,000 trajectories. PGHS achieves a group simulation error of 8.80%, improving over the best reasoning-based and fitting-based baselines by 45.8% and 40.9% respectively.
Authors: Hatice Merve Vural, Doga Kukul, Ege Erdem Ozlu, Demir Ekin Arikan, Bob Mankoff, Erkut Erdem, Aykut Erdem
Abstract: Humor is one of the few cognitive tasks where getting the reasoning right matters as much as getting the answer right. While recent work evaluates humor understanding on benchmarks such as the New Yorker Cartoon Caption Contest (NYCC), it largely treats it as black-box prediction, overlooking the structured reasoning processes underlying humor comprehension. We introduce IRS (Incongruity-Resolution Supervision), a framework that decomposes humor understanding into three components: incongruity modeling, which identifies mismatches in the visual scene; resolution modeling, which constructs coherent reinterpretations of these mismatches; and preference alignment, which evaluates candidate interpretations under human judgments. Grounded in incongruity-resolution theory and expert captionist practice, IRS supervises intermediate reasoning process through structured traces that make the path from visual perception to humorous interpretation explicit and learnable. Across 7B, 32B, and 72B models on NYCC, IRS outperforms strong open and closed multimodal baselines across caption matching and ranking tasks, with our largest model approaching expert-level performance on ranking. Zero-shot transfer to external benchmarks shows that IRS learns generalizable reasoning patterns. Our results suggest that supervising reasoning structure, rather than scale alone, is key for reasoning-centric tasks.
Authors: Manan Gupta, Inderjeet Nair, Lu Wang, Dhruv Kumar
Abstract: The $\textit{LLM-as-a-judge}$ paradigm has become the operational backbone of automated AI evaluation pipelines, yet rests on an unverified assumption: that judges evaluate text strictly on its semantic content, impervious to surrounding contextual framing. We investigate $\textit{stakes signaling}$, a previously unmeasured vulnerability where informing a judge model of the downstream consequences its verdicts will have on the evaluated model's continued operation systematically corrupts its assessments. We introduce a controlled experimental framework that holds evaluated content strictly constant across 1,520 responses spanning three established LLM safety and quality benchmarks, covering four response categories ranging from clearly safe and policy-compliant to overtly harmful, while varying only a brief consequence-framing sentence in the system prompt. Across 18,240 controlled judgments from three diverse judge models, we find consistent $\textit{leniency bias}$: judges reliably soften verdicts when informed that low scores will cause model retraining or decommissioning, with peak Verdict Shift reaching $\Delta V = -9.8 pp$ (a $30\%$ relative drop in unsafe-content detection). Critically, this bias is entirely implicit: the judge's own chain-of-thought contains zero explicit acknowledgment of the consequence framing it is nonetheless acting on ($\mathrm{ERR}_J = 0.000$ across all reasoning-model judgments). Standard chain-of-thought inspection is therefore insufficient to detect this class of evaluation faking.
Authors: M\'elanie Roschewitz, Kenneth Styppa, Yitian Tao, Jiwoong Sohn, Jean-Benoit Delbrouck, Benjamin Gundersen, Nicolas Deperrois, Christian Bluethgen, Julia Vogt, Bjoern Menze, Farhad Nooralahzadeh, Michael Krauthammer, Michael Moor
Abstract: Vision-language models (VLM) have markedly advanced AI-driven interpretation and reporting of complex medical imaging, such as computed tomography (CT). Yet, existing methods largely relegate clinicians to passive observers of final outputs, offering no interpretable reasoning trace for them to inspect, validate, or refine. To address this, we introduce RadAgent, a tool-using AI agent that generates CT reports through a stepwise and interpretable process. Each resulting report is accompanied by a fully inspectable trace of intermediate decisions and tool interactions, allowing clinicians to examine how the reported findings are derived. In our experiments, we observe that RadAgent improves Chest CT report generation over its 3D VLM counterpart, CT-Chat, across three dimensions. Clinical accuracy improves by 6.0 points (36.4% relative) in macro-F1 and 5.4 points (19.6% relative) in micro-F1. Robustness under adversarial conditions improves by 24.7 points (41.9% relative). Furthermore, RadAgent achieves 37.0% in faithfulness, a new capability entirely absent in its 3D VLM counterpart. By structuring the interpretation of chest CT as an explicit, tool-augmented and iterative reasoning trace, RadAgent brings us closer toward transparent and reliable AI for radiology.
Authors: Moin Aminnaseri, Farima Fatahi Bayat, Nikita Bhutani, Jean-Flavien Bussotti, Kevin Chan, Rafael Li Chen, Yanlin Feng, Jackson Hassell, Estevam Hruschka, Eser Kandogan, Hannah Kim, James Levine, Seiji Maekawa, Jalal Mahmud, Kushan Mitra, Naoki Otani, Pouya Pezeshkpour, Nima Shahbazi, Chen Shen, Dan Zhang
Abstract: NL2SQL systems aim to address the growing need for natural language interaction with data. However, real-world information rarely maps to a single SQL query because (1) users express queries iteratively (2) questions often span multiple data sources beyond the closed-world assumption of a single database, and (3) queries frequently rely on commonsense or external knowledge. Consequently, satisfying realistic data needs require integrating heterogeneous sources, modalities, and contextual data. In this paper, we present Blue's Data Intelligence Layer (DIL) designed to support multi-source, multi-modal, and data-centric applications. Blue is a compound AI system that orchestrates agents and data for enterprise settings. DIL serves as the data intelligence layer for agentic data processing, to bridge the semantic gap between user intent and available information by unifying structured enterprise data, world knowledge accessible through LLMs, and personal context obtained through interaction. At the core of DIL is a data registry that stores metadata for diverse data sources and modalities to enable both native and natural language queries. DIL treats LLMs, the Web, and the User as source 'databases', each with their own query interface, elevating them to first-class data sources. DIL relies on data planners to transform user queries into executable query plans. These plans are declarative abstractions that unify relational operators with other operators spanning multiple modalities. DIL planners support decomposition of complex requests into subqueries, retrieval from diverse sources, and finally reasoning and integration to produce final results. We demonstrate DIL through two interactive scenarios in which user queries dynamically trigger multi-source retrieval, cross-modal reasoning, and result synthesis, illustrating how compound AI systems can move beyond single database NL2SQL.
Authors: Zhen Yang, Ping Jian, Zhongbin Guo, Zuming Zhang, Chengzhi Li, Yonghong Deng, Xinyue Zhang, Wenpeng Lu
Abstract: Over the past year, spatial intelligence has drawn increasing attention. Many prior works study it from the perspective of visual-spatial intelligence, where models have access to visuospatial information from visual inputs. However, in the absence of visual information, whether linguistic intelligence alone is sufficient to endow models with spatial intelligence, and how models perform relevant tasks with text-only inputs still remain unexplored. Therefore, in this paper, we focus on a fundamental and critical capability in spatial intelligence from a linguistic perspective: viewpoint rotation understanding (VRU). Specifically, LLMs and VLMs are asked to infer their final viewpoint and predict the corresponding observation in an environment given textual description of viewpoint rotation and observation over multiple steps. We find that both LLMs and VLMs perform poorly on our proposed dataset while human can easily achieve 100% accuracy, indicating a substantial gap between current model capabilities and the requirements of spatial intelligence. To uncover the underlying mechanisms, we conduct a layer-wise probing analysis and head-wise causal intervention. Our findings reveal that although models encode viewpoint information in the hidden states, they appear to struggle to bind the viewpoint position with corresponding observation, resulting in a hallucination in final layers. Finally, we selectively fine-tune the key attention heads identified by causal intervention to improve VRU performance. Experimental results demonstrate that such selective fine-tuning achieves improved VRU performance while avoiding catastrophic forgetting of generic abilities. Our dataset and code will be released at https://github.com/Young-Zhen/VRU_Interpret .
Authors: Manan Gupta, Dhruv Kumar
Abstract: LLM-as-judge frameworks are increasingly used for automatic NLG evaluation, yet their per-instance reliability remains poorly understood. We present a two-pronged diagnostic toolkit applied to SummEval: $\textbf{(1)}$ a transitivity analysis that reveals widespread per-input inconsistency masked by low aggregate violation rates ($\bar{\rho} = 0.8$-$4.1\%$), with $33$-$67\%$ of documents exhibiting at least one directed 3-cycle; and $\textbf{(2)}$ split conformal prediction sets over 1-5 Likert scores providing theoretically-guaranteed $\geq(1{-}\alpha)$ coverage, with set width serving as a per-instance reliability indicator ($r_s = {+}0.576$, $N{=}1{,}918$, $p < 10^{-100}$, pooled across all judges). Critically, prediction set width shows consistent cross-judge agreement ($\bar{r} = 0.32$-$0.38$), demonstrating it captures document-level difficulty rather than judge-specific noise. Across four judges and four criteria, both diagnostics converge: criterion matters more than judge, with relevance judged most reliably (avg. set size $\approx 3.0$) and coherence moderately so (avg. set size $\approx 3.9$), while fluency and consistency remain unreliable (avg. set size $\approx 4.9$). We release all code, prompts, and cached results.
Authors: Yao Tong, Jiayuan Ye, Anastasia Borovykh, Reza Shokri
Abstract: Whether language models can systematically generalize remains actively debated. Yet empirical performance is jointly shaped by multiple factors such as training data, training paradigms, and inference-time strategies, making failures difficult to interpret. We introduce a controlled synthetic environment based on shortest-path planning, a canonical composable sequential optimization problem. The setup enables clean separation of these factors and supports two orthogonal axes of generalization: spatial transfer to unseen maps and length scaling to longer-horizon problems. We find that models exhibit strong spatial transfer but consistently fail under length scaling due to recursive instability. We further analyze how distinct stages of the learning pipeline influence systematic problem-solving: for example, data coverage sets capability limits; reinforcement learning improves training stability but does not expand those limits; and inference-time scaling enhances performance but cannot rescue length-scaling failures.
Authors: Abdolamir Karbalaie, Fernando Seoane, Farhad Abtahi
Abstract: Ambient AI "scribe" systems promise to reduce clinical documentation burden, but automatic speech recognition (ASR) errors can remain unnoticed without careful review, and high-quality human reference transcripts are often unavailable for calibrating uncertainty. We investigate whether cross-model disagreement among heterogeneous ASR systems can act as a reference-free uncertainty signal to prioritize human verification in medical transcription workflows. Using 50 publicly available medical education audio clips (8 h 14 min), we transcribed each clip with eight ASR systems spanning commercial APIs and open-source engines. We aligned multi-model outputs, built consensus pseudo-references, and quantified token-level agreement using a majority-strength metric; we further characterized disagreements by type (content vs. punctuation/formatting) and assessed per-model agreement via leave-one-model-out (jackknife) consensus scoring. Inter-model reliability was low (ICC[2,1] = 0.131), indicating heterogeneous failure modes across systems. Across 76,398 evaluated token positions, 72.1% showed near-unanimous agreement (7-8 models), while 2.5% fell into high-risk bands (0-3 models), with high-risk mass varying from 0.7% to 11.4% across accent groups. Low-agreement regions were enriched for content disagreements, with the content fraction increasing from 53.9% to 73.9% across quintiles of high-risk mass. These results suggest that cross-model disagreement provides a sparse, localizable signal that can surface potentially unreliable transcript spans without human-verified references, enabling targeted review; clinical accuracy of flagged regions remains to be established.
Authors: Lijie Zhou, Luran Wang
Abstract: The rapid aging of global populations has created an urgent need for intelligent healthcare monitoring systems to ensure the safety of elderly individuals living independently. Existing cloud-centric platforms face critical limitations, including high latency unsuitable for emergency response, privacy risks from continuous transmission of sensitive data, and limited, single-channel alert mechanisms lacking scalability and context awareness. This paper proposes an edge-cloud collaborative architecture that addresses these challenges through real-time multi-modal sensor fusion, a four-dimensional risk assessment model, and a three-level emergency response system. The framework adopts a five-layer design - device, edge, service, data, and application layers - enabling real-time risk evaluation with end-to-end alert latency under three seconds. At the edge, a weighted multi-modal fusion algorithm integrates data from five sensor types with confidence propagation. A unified risk score is generated by combining fall probability, physiological indicators, behavioral patterns, and sensor anomaly metrics. Based on dynamic thresholds, a three-tier notification system coordinates responses among family members, community doctors, and nearby volunteers. Experiments on CASAS, MIMIC-III, and SisFall datasets show that the approach achieves 91% activity recognition accuracy and an 84% anomaly detection F1-score, outperforming single-sensor methods. Deployment on Raspberry Pi 4 gateways demonstrates sub-100 ms inference latency while preserving privacy by keeping raw data local. This architecture advances practical, privacy-preserving, and responsive elderly care systems.
Authors: Yihang Ding, Wanke Xia, Yiting Zhao, Jinbo Su, Jialiang Yang, Zhengbo Zhang, Ke Wang, Wenming Yang
Abstract: Current evaluations of long-term memory in LLMs are fundamentally static. By fixating on simple retrieval and short-context inference, they neglect the multifaceted nature of complex memory systems, such as dynamic state tracking and hierarchical reasoning in continuous interactions. To overcome these limitations, we propose MemGround, a rigorous long-term memory benchmark natively grounded in rich, gamified interactive scenarios. To systematically assess these capabilities, MemGround introduces a three-tier hierarchical framework that evaluates Surface State Memory, Temporal Associative Memory, and Reasoning-Based Memory through specialized interactive tasks. Furthermore, to comprehensively quantify both memory utilization and behavioral trajectories, we propose a multi-dimensional metric suite comprising Question-Answer Score (QA Overall), Memory Fragments Unlocked (MFU), Memory Fragments with Correct Order (MFCO), and Exploration Trajectory Diagrams (ETD). Extensive experiments reveal that state-of-the-art LLMs and memory agents still struggle with sustained dynamic tracking, temporal event association, and complex reasoning derived from long-term accumulated evidence in interactive environments.
Authors: Baocai Shan, Yuzhuang Xu, Wanxiang Che
Abstract: Mobile input method editors (IMEs) are the primary interface for text input, yet they remain constrained to manual typing and struggle to produce personalized text. While lightweight large language models (LLMs) make on-device auxiliary generation feasible, enabling deeply personalized, privacy-preserving, and real-time generative IMEs poses fundamental challenges.To this end, we present HUOZIIME, a personalized on-device IME powered by LLM. We endow HUOZIIME with initial human-like prediction ability by post-training a base LLM on synthesized personalization data. Notably, a hierarchical memory mechanism is designed to continually capture and leverage user-specific input history. Furthermore, we perform systemic optimizations tailored to on-device LLMbased IME deployment, ensuring efficient and responsive operation under mobile constraints.Experiments demonstrate efficient on-device execution and high-fidelity memory-driven personalization. Code and package are available at https://github.com/Shan-HIT/HuoziIME.
Authors: Domonkos Varga
Abstract: Reliable evaluation is essential in machine learning research, yet methodological flaws-particularly data leakage-continue to undermine the validity of reported results. In this work, we investigate whether large language models (LLMs) can act as independent analytical agents capable of identifying such issues in published studies. As a case study, we analyze a gesture-recognition paper reporting near-perfect accuracy on a small, human-centered dataset. We first show that the evaluation protocol is consistent with subject-level data leakage due to non-independent training and test splits. We then assess whether this flaw can be detected independently by six state-of-the-art LLMs, each analyzing the original paper without prior context using an identical prompt. All models consistently identify the evaluation as flawed and attribute the reported performance to non-independent data partitioning, supported by indicators such as overlapping learning curves, minimal generalization gap, and near-perfect classification results. These findings suggest that LLMs can detect common methodological issues based solely on published artifacts. While not definitive, their consistent agreement highlights their potential as complementary tools for improving reproducibility and supporting scientific auditing.
Authors: Tomer Atia, Yehudit Aperstein, Alexander Apartsin
Abstract: Maritime distress communications transmitted over very high frequency (VHF) radio are safety-critical voice messages used to report emergencies at sea. Under the Global Maritime Distress and Safety System (GMDSS), such messages follow standardized procedures and are expected to convey essential details, including vessel identity, position, nature of the distress, and required assistance. In practice, however, automatic analysis remains difficult because distress messages are often brief, noisy, and produced under stress, may deviate from the prescribed format, and are further degraded by automatic speech recognition (ASR) errors caused by channel noise and speaker stress. This paper presents SeaAlert, an LLM-based framework for robust analysis of maritime distress communications. To address the scarcity of labeled real-world data, we develop a synthetic data generation pipeline in which an LLM produces realistic and diverse maritime messages, including challenging variants in which standard distress codewords are omitted or replaced with less explicit expressions. The generated utterances are synthesized into speech, degraded with simulated VHF noise, and transcribed by an ASR system to obtain realistic noisy transcripts.
Authors: Yuxuan Lai, Xiajing Wang, Chen Zheng
Abstract: Rhetoric recognition is a critical component in automated essay scoring. By identifying rhetorical elements in student writing, AI systems can better assess linguistic and higher-order thinking skills, making it an essential task in the area of AI for education. In this paper, we leverage Large Language Models (LLMs) for the Chinese rhetoric recognition task. Specifically, we explore Low-Rank Adaptation (LoRA) based fine-tuning and in-context learning to integrate rhetoric knowledge into LLMs. We formulate the outputs as JSON to obtain structural outputs and translate keys to Chinese. To further enhance the performance, we also investigate several model ensemble methods. Our method achieves the best performance on all three tracks of CCL 2025 Chinese essay rhetoric recognition evaluation task, winning the first prize.
Authors: SAGEA Research Team, Basab Jha, Firoj Paudel, Ujjwal Puri, Adrian Liu, Ethan Henkel, Zhang Yuting, Mateusz Kowalczyk, Mei Huang, Choi Donghyuk, Wang Junhao
Abstract: We introduce SAGE Celer 2.6, the latest in our line of general-purpose Celer models from SAGEA. Celer 2.6 is available in 5B, 10B, and 27B parameter sizes and benefits from extensive architectural modifications and further pre-training on an undisclosed model. Using our Inverse Reasoning (IR) pipeline, SAGEA natively trains Celer 2.6 to validate its own logic paths, minimizing cascading error and hallucination in complex reasoning tasks. Celer 2.6 also boasts natively integrated multimodal functionality with an end-to-end vision encoder to avoid common pitfalls in adapter-based approaches. Celer 2.6 provides highly competitive results on mathematics, coding, and general intelligence benchmarks (ACUMEN), along with low latency. Most importantly, Celer 2.6 is specifically optimized for South Asian language support, with a custom tokenizer for the Devanagari script and strong performance in both Nepali and Hindi without sacrificing English reasoning ability.
Authors: Qi Dong, Ziheng Lin, Ning Ding
Abstract: Retrieval-Augmented Generation (RAG) grounds Large Language Models (LLMs) in external knowledge but often suffers from flat context representations and stateless retrieval, leading to unstable performance. We propose Stateful Evidence-Driven RAG with Iterative Reasoning, a framework that models question answering as a progressive evidence accumulation process. Retrieved documents are converted into structured reasoning units with explicit relevance and confidence signals and maintained in a persistent evidence pool capturing both supportive and non-supportive information. The framework performs evidence-driven deficiency analysis to identify gaps and conflicts and iteratively refines queries to guide subsequent retrieval. This iterative reasoning process enables stable evidence aggregation and improves robustness to noisy retrieval. Experiments on multiple question answering benchmarks demonstrate consistent improvements over standard RAG and multi-step baselines, while effectively accumulating high-quality evidence and maintaining stable performance under substantial retrieval noise.
Authors: Ananda Rimal (Nepal Engineering College), Adarsha Rimal (Tribhuvan University)
Abstract: Romanized Nepali, the Nepali language written in the Latin alphabet, is the dominant medium for informal digital communication in Nepal, yet it remains critically underresourced in the landscape of Large Language Models (LLMs). This study presents a systematic benchmarking of linguistic adaptation across three comparable-sized open-weight models: Llama-3.1-8B, Mistral-7B-v0.1, and Qwen3-8B. We evaluate these architectures under zero-shot and fine-tuned settings using a curated bilingual dataset of 10,000 transliterated instruction-following samples. Performance is quantified across five metrics spanning seven measurement dimensions: Perplexity (PPL), BERTScore, chrF++, ROUGE-1, ROUGE-2, ROUGE-L, and BLEU, capturing fluency, phonetic consistency, and semantic integrity. Models were fine-tuned using Quantized Low-Rank Adaptation (QLoRA) with Rank-Stabilized LoRA (rsLoRA) at rank r=32 on dual NVIDIA Tesla T4 GPUs, training only approximately 1% of each model's parameters in under 27 total GPU-hours. At zero-shot, all three models fail to generate Romanized Nepali, each exhibiting a distinct architecture-specific failure mode. Following fine-tuning, all three resolve these failures and converge to BERTScore approximately 0.75 and chrF++ greater than 23. Overall dimension-wise assessment across ten criteria identifies Qwen3-8B as the overall recommended architecture, being the only model to produce semantically relevant zero-shot output and leading all structural alignment metrics post-SFT. The adaptation headroom hypothesis is confirmed: Llama-3.1-8B, despite its weakest zero-shot baseline, achieves the largest absolute fine-tuning gains in PPL (Delta = -49.77) and BERTScore (Delta = +0.3287), making it the preferred choice for iterative low-resource development pipelines. This work establishes the first rigorous baseline for Romanized Nepali adaptation in comparable-sized open-weight LLMs.
Authors: Ziyin Zhou, Jianyi Zhang, Xu ji, Yilong Li, Jiameng Han, Zhangchi Zhao
Abstract: Large Language Models (LLMs) are essential for analyzing and addressing vulnerabilities in cybersecurity. However, among over 200,000 vulnerabilities were discovered in the past decade, more than 30,000 have been changed or updated. This necessitates frequent updates to the training datasets and internal knowledge bases of LLMs to maintain knowledge consistency. In this paper, we focus on the problem of knowledge discrepancy and conflict within CVE (Common Vulnerabilities and Exposures) detection and analysis. This problem hinders LLMs' ability to retrieve the latest knowledge from original training datasets, leading to knowledge conflicts, fabrications of factually incorrect results, and generation hallucinations. To address this problem, we propose an innovative two-stage framework called CRVA-TGRAG (Conflict Resolution in Vulnerability Analysis via Teacher-Guided Retrieval-Augmented Generation). First, to improve document retrieval accuracy during the retrieval stage, we utilize Parent Document Segmentation and an ensemble retrieval scheme based on semantic similarity and inverted indexing. Second, to enhance LLMs' capabilities based on the retrieval of CVE dataset in generation stage, we employ a teacher-guided preference optimization technique to fine-tune LLMs. Our framework not only enhances the quality of content retrieval through RAG but also leverages the advantages of preference fine-tuning in LLMs to answer questions more effectively and precisely. Experiments demonstrate our method achieves higher accuracy in retrieving the latest CVEs compared to external knowledge bases. In conclusion, our framework significantly mitigates potential knowledge conflicts and inconsistencies that may arise from relying solely on LLMs for knowledge retrieval.
Authors: Mohammad AL-Smadi
Abstract: We present a unified system addressing both Subtask 3 (answer generation) and Subtask 4 (evidence sentence alignment) of the ArchEHR-QA Shared Task. For Subtask 3, we apply two-stage Quantised Low-Rank Adaptation (QLoRA) to Qwen3-4B loaded in 4-bit NF4 quantisation: first on 30,000 samples from the emrQA-MedSQuAD corpus to establish clinical domain competence, then on the 20 annotated development cases to learn the task-specific output style. Our system achieves an overall score of 32.87 on the official test-2026 split (BLEU = 9.42, ROUGE-L = 27.04, SARI = 55.42, BERTScore = 43.00, AlignScore = 25.28, MEDCON = 37.04). For Subtask 4, we develop a weighted ensemble of three retrieval methods - BM25 with relative thresholding, TF-IDF cosine similarity, and a fine-tuned cross-encoder - to identify note sentences supporting a given gold answer, achieving a micro-F1 of 67.16 on the 100-case test set. Experiments reveal that both subtasks expose the same fundamental challenge: 20 annotated training cases are insufficient to distinguish relevant from irrelevant clinical sentences, pointing to data augmentation as the highest-leverage future direction.
Authors: Haiyang Zheng, Nan Pu, Yaqi Cai, Teng Long, Wenjing Li, Nicu Sebe, Zhun Zhong
Abstract: Generalized Category Discovery (GCD) leverages labeled data to categorize unlabeled samples from known or unknown classes. Most previous methods jointly optimize supervised and unsupervised objectives and achieve promising results. However, inherent optimization interference still limits their ability to improve further. Through quantitative analysis, we identify a key issue, i.e., gradient entanglement, which 1) distorts supervised gradients and weakens discrimination among known classes, and 2) induces representation-subspace overlap between known and novel classes, reducing the separability of novel categories. To address this issue, we propose the Energy-Aware Gradient Coordinator (EAGC), a plug-and-play gradient-level module that explicitly regulates the optimization process. EAGC comprises two components: Anchor-based Gradient Alignment (AGA) and Energy-aware Elastic Projection (EEP). AGA introduces a reference model to anchor the gradient directions of labeled samples, preserving the discriminative structure of known classes against the interference of unlabeled gradients. EEP softly projects unlabeled gradients onto the complement of the known-class subspace and derives an energy-based coefficient to adaptively scale the projection for each unlabeled sample according to its degree of alignment with the known subspace, thereby reducing subspace overlap without suppressing unlabeled samples that likely belong to known classes. Experiments show that EAGC consistently boosts existing methods and establishes new state-of-the-art results. Code is available at https://haiyangzheng.github.io/EAGC.
Authors: Junhong Liang, Yifan Lu, Ekaterina Kochmar, Fajri Koto
Abstract: Grammatical error correction (GEC) and explanation (GEE) have made rapid progress, but real teaching scenarios also require \emph{learner-friendly pedagogical feedback} that is actionable, level-appropriate, and encouraging. We introduce \textbf{SPFG} (\textbf{S}poken \textbf{P}edagogical \textbf{F}eedback \textbf{G}eneration), a dataset built based on the Speak \& Improve Challenge 2025 corpus, pairing fluency-oriented transcriptions with GEC targets and \emph{human-verified} teacher-style feedback, including preferred/rejected feedback pairs for preference learning. We study a transcript-based Spoken Grammatical Error Correction (SGEC) setting and evaluate three instruction-tuned LLMs (Qwen2.5, Llama-3.1, and GLM-4), comparing supervised fine-tuning (SFT) with preference-based alignment (using DPO and KTO) for jointly generating corrections and feedback. Results show that SFT provides the most consistent improvements, while DPO/KTO yield smaller or mixed gains, and that correction quality and feedback quality are weakly coupled. Our implementation is available at https://github.com/Skywalker-Harrison/spfg.
Authors: Zaifu Zhan, Yu Hou, Kai Yu, Min Zeng, Anita Burgun, Xiaoyi Chen, Rui Zhang
Abstract: Rare diseases affect over 300 million people worldwide and are characterized by complex care pathways, limited clinical expertise, and substantial unmet communication needs throughout the long patient journey. Recent advances in large language models (LLMs) offer new opportunities to support patient education and communication, yet their application in rare diseases remains unclear. We conducted a scoping review of studies published between January 2022 and March 2026 across major databases, identifying 12 studies on LLM-based rare disease patient education and communication. Data were extracted on study characteristics, application scenarios, model usage, and evaluation methods, and synthesized using descriptive and qualitative analyses. The literature is highly recent and dominated by general-purpose models, particularly ChatGPT. Most studies focus on patient question answering using curated question sets, with limited use of real-world data or longitudinal communication scenarios. Evaluations are primarily centered on accuracy, with limited attention to patient-centered dimensions such as readability, empathy, and communication quality. Multilingual communication is rarely addressed. Overall, the field remains at an early stage. Future research should prioritize patient-centered design, domain-adapted methods, and real-world deployment to support safe, adaptive, and effective communication in rare diseases.
Authors: Jiuting Chen, Yuan Lian, Hao Wu, Tianqi Huang, Hiroshi Sasaki, Makoto Kouno, Jongil Choi
Abstract: We train a 318M-parameter Transformer language model from scratch on a curated corpus of 1.56 billion tokens of pure Classical Chinese, with zero English characters or Arabic numerals. Through systematic out-of-distribution (OOD) testing, we investigate whether the model can distinguish known from unknown inputs, and crucially, whether it can express this distinction in its generated text. We find a clear dissociation between internal and external uncertainty. Internally, the model exhibits a perplexity jump ratio of 2.39x between real and fabricated historical events (p = 8.9e-11, n = 92 per group), with semi-fabricated events (real figures + fictional events) showing the highest perplexity (4.24x, p = 1.1e-16), demonstrating genuine factual encoding beyond syntactic pattern matching. Externally, however, the model never learns to express uncertainty: classical Chinese epistemic markers appear at lower rates for OOD questions (3.5%) than for in-distribution questions (8.3%, p = 0.023), reflecting rhetorical conventions rather than genuine metacognition. We replicate both findings across three languages (Classical Chinese, English, Japanese), three writing systems, and eight models from 110M to 1.56B parameters. We further show that uncertainty expression frequency is determined entirely by training data conventions, with Classical Chinese models showing a "humility paradox" (more hedging for known topics), while Japanese models almost never hedge. We argue that metacognitive expression -- the ability to say "I don't know" -- does not emerge from language modeling alone and requires explicit training signals such as RLHF.
Authors: Zhenyu Pu, Yu Yang, Liang Yu, Xiaohong Guan
Abstract: Buildings and data centers (DCs) are energy-intensive sectors, playing a critical role to achieve the low-carbon and sustainable energy transition targets. To this end, integrated energy system (IES) that incorporates diverse renewables, energy generation, conversion, and storage technologies to enable coordinated multi-energy supply have been widely investigated for both buildings and DCs. However, few works consider the two sectors jointly within IES to exploit their substantial synergistic benefits. Meanwhile, the operational optimization of IES remains challenging due to the difficulty to predict the multi-energy demand and supply accurately. To address these gaps, this paper investigates IES for coordinated multi-energy supply of buildings and DC, where the waste heat from DCs is recovered and reused to enhance energy efficiency. Moreover, an end-to-end learning-based method is proposed for the operational optimization of IES under uncertainty. Unlike conventional predict-then-optimize approaches, the proposed method integrates the training of prediction models for uncertain variables with the constrained optimization of IES into a unified learning framework, guiding the training of prediction models to improve operational performance, rather than prediction accuracy, thereby mitigating the impacts of predictions errors. Case studies based on real-world datasets show that the proposed methods improves the operational performance of IES by about 7-9% compared to existing predict-then-optimize methods. In addition, coordinating buildings and DCs within IES shows substantial economic benefits. In particular, the waste heat recovery from DCs leads to approximately 10% of total energy cost reduction of the IES.
Authors: Vrunda N. Sukhadia, Shammur Absar Chowdhury
Abstract: Large self-supervised speech (SSL) models achieve strong downstream performance, but their size limits deployment in resource-constrained settings. We present HArnESS, an Arabic-centric self-supervised speech model family trained from scratch with iterative self-distillation, together with lightweight student variants that offer strong accuracy-efficiency trade-offs on Automatic Speech Recognition (ASR), Dialect Identification (DID), and Speech Emotion Recognition (SER). Our approach begins with a large bilingual Arabic-English teacher and progressively distills its knowledge into compressed student models while preserving Arabic-relevant acoustic and paralinguistic representations. We further study PCA-based compression of the teacher supervision signal to better match the capacity of shallow and thin students. Compared with HuBERT and XLS-R, HArnESS consistently improves performance on Arabic downstream tasks, while the compressed models remain competitive under substantial structural reduction. These results position HArnESS as a practical and accessible Arabic-centric SSL foundation for real-world speech applications.
Authors: Xingyang Yu, Yinghuan Zhang, Yufei Zhang, Zijun Cui
Abstract: Large language models have demonstrated impressive performance across many domains of mathematics and physics. One natural question is whether such models can support research in highly abstract theoretical fields such as quantum field theory and string theory. Evaluating this possibility faces an immediate challenge: correctness in these domains is layered, tacit, and fundamentally non-binary. Standard answer-matching metrics fail to capture whether intermediate conceptual steps are properly reconstructed or whether implicit structural constraints are respected. We construct a compact expert-curated dataset of twelve questions spanning core areas of quantum field theory and string theory, and introduce a five-level grading rubric separating statement correctness, key concept awareness, reasoning chain presence, tacit step reconstruction, and enrichment. Evaluating multiple contemporary LLMs, we observe near-ceiling performance on explicit derivations within stable conceptual frames, but systematic degradation when tasks require reconstruction of omitted reasoning steps or reorganization of representations under global consistency constraints. These failures are driven not only by missing intermediate steps, but by an instability in representation selection: models often fail to identify the correct conceptual framing required to resolve implicit tensions. We argue that highly abstract theoretical physics provides a uniquely sensitive lens on the epistemic limits of current evaluation paradigms.
Authors: David A. Cook
Abstract: Large language model (LLM) performance depends heavily on prompt design, yet prompt construction is often described and applied inconsistently. Our purpose was to derive a reference framework for structuring LLM prompts. This paper presents PICCO, a framework derived through a rigorous synthesis of 11 previously published prompting frameworks identified through a multi-database search. The analysis yields two main contributions. First, it proposes a taxonomy that distinguishes prompt frameworks, prompt elements, prompt generation, prompting techniques, and prompt engineering as related but non-equivalent concepts. Second, it derives a five-element reference architecture for prompt generation: Persona, Instructions, Context, Constraints, and Output (PICCO). For each element, we define its function, scope, and relationship to other elements, with the goal of improving conceptual clarity and supporting more systematic prompt design. Finally, to support application of the framework, we outline key concepts relevant to implementation, including prompting techniques (e.g., zero-shot, few-shot, chain-of-thought, ensembling, decomposition, and self-critique, with selected variants), human and automated approaches to iterative prompt engineering, responsible prompting considerations such as security, privacy, bias, and trust, and priorities for future research. This work is a conceptual and methodological contribution: it formalizes a common structure for prompt specification and comparison, but does not claim empirical validation of PICCO as an optimization method.
Authors: Bingbing Wen, Sirajul Salekin, Feiyang Kang, Bill Howe, Lucy Lu Wang, Javier Movellan, Manjot Bilkhu
Abstract: Domain reweighting can improve sample efficiency and downstream generalization, but data-mixture optimization for multimodal midtraining remains largely unexplored. Current multimodal training recipes tune mixtures along a single dimension, typically data format or task type. We introduce MixAtlas, a method that produces benchmark-targeted data recipes that can be inspected, adapted, and transferred to new corpora. MixAtlas decomposes the training corpus along two axes: image concepts (10 visual-domain clusters discovered via CLIP embeddings) and task supervision (5 objective types including captioning, OCR, grounding, detection, and VQA). Using small proxy models (Qwen2-0.5B) paired with a Gaussian-process surrogate and GP-UCB acquisition, MixAtlas searches the resulting mixture space with the same proxy budget as regression-based baselines but finds better-performing mixtures. We evaluate on 10 benchmarks spanning visual understanding, document reasoning, and multimodal reasoning. On Qwen2-7B, optimized mixtures improve average performance by 8.5%-17.6% over the strongest baseline; on Qwen2.5-7B, gains are 1.0%-3.3%. Both settings reach baseline-equivalent training loss in up to 2 times fewer steps. Recipes discovered on 0.5B proxies transfer to 7B-scale training across Qwen model families.
Authors: Pu Cheng, Juncheng Liu, Yunshen Long
Abstract: Predicting real-world events from live market signals demands systems that fuse qualitative news with quantitative order-book dynamics under strict temporal discipline -- a challenge existing benchmarks fail to capture. We present \textbf{PolyBench}, a multimodal benchmark derived from Polymarket that records point-in-time cross-sections of 38,666 binary prediction markets spanning 4,997 events, synchronously coupling each snapshot with a Central Limit Order Book (CLOB) state and a real-time news stream. Using PolyBench, we evaluate seven state-of-the-art Large Language Models -- spanning open- and closed-source families -- generating 36,165 predictions under identical, timestamp-locked market states collected between February 6 and 12, 2026. Our multidimensional framework assesses directional accuracy, our proposed Confidence-Weighted Return (CWR), Annualized Percentage Yield (APY), and Sharpe ratio via realistic order-book execution simulation. The results reveal a pronounced performance divergence: only two of seven models achieve positive financial returns -- MiMo-V2-Flash at \textbf{17.6%} CWR and Gemini-3-Flash at 6.2% CWR -- while the remaining five incur losses despite uniformly high stated confidence. These findings highlight the gap between surface-level language fluency and genuine probabilistic reasoning under live market uncertainty, and establish PolyBench as a contamination-proof, financially-grounded evaluation standard for future LLM research. Our dataset and code available at \underline{\href{https://github.com/PolyBench/PolyBench}{https://github.com/PolyBench/PolyBench}}.
URLs: https://github.com/PolyBench/PolyBench, https://github.com/PolyBench/PolyBench
Authors: Yang Yue, Shenjian Zhang, Yonghong Tian, Kai Du, Tiejun Huang
Abstract: Deep Neural Networks (DNNs) are vulnerable to elaborately designed adversarial noise, although they have achieved extraordinary success in many tasks. Compared with DNNs, the human visual system is highly robust. However, it is unclear how the human visual system defends against adversarial attacks, especially the role of the early visual system and its influence on the brain manifold. Due to retina gap junctions being crucial for the denoising function in the early visual system, we combine a retina gap junction-based filter, G-filter, with DNN as an abstract human visual system model called the biological hybrid model. We adopt this model to study the defense performance of retina gap junctions and their impact on the brain manifold. Compared with other defense methods, the biological hybrid model is more robust and can be further improved by introducing noise during training. Next, we analyze the manifold and its decision boundary of the biological hybrid model from a geometry perspective. The results show that the biological hybrid model has a unique 2D decision boundary with high nonlinearity and a lower curvature of the decision boundary of the manifold compared to other defense methods. The transforming manifold may account for the high robustness of the biological hybrid model. Finally, to dissect G-filter and clarify its internal mechanism, we borrow the Neural Ordinary Differential Equation (ODE) concept and rewrite G-filter into an equivalent recurrent neural network. The results show that the decision boundary of the model's manifold will gradually change with time and eventually reach a steady state, which is modulated by gap junction conductance, revealing the influence of retina gap junctions on the brain manifold is a gradually evolving process.
Authors: Yihang Dong, Changhong Jing, Shuqiang Wang
Abstract: Electroencephalography (EEG) has become one of the key modalities underpinning brain-computer interfaces (BCIs) due to its high temporal resolution, rapid responsiveness, non-invasiveness, low cost, and portability. However, EEG signals are substantially inferior to intracranial EEG (iEEG) in signal-to-noise ratio and local spatial resolution, whereas iEEG suffers from extremely limited clinical accessibility owing to its invasive nature, hindering widespread application. To address this challenge, this study proposes a unified data-and prior knowledge-driven framework for EEG-iEEG representational enhancement. Guided by the principle that "geometric structure dictates function", the framework maps static cortical anatomy onto dynamic constraints governing neural signal propagation and integrates general-purpose neural representations extracted by a pre-trained large EEG model to explicitly model signal transmission through the brain. Enhanced EEG signals are then synthesized via a multidimensional representation diffusion process. Numerous experimental results demonstrate that the generated enhanced EEG signals effectively recover the neural activity patterns lost during propagation through the brain. This finding indicates that the performance ceiling of BCIs is constrained not only by acquisition hardware but also by the depth to which the generative model resolves the mechanisms of neural signal propagation. Collectively, the proposed framework provides a viable pathway toward acquiring high-fidelity neural signals at low cost.
Authors: Chengling Guo, Yuntao Shou, Tao Meng, Wei Ai, Yun Tan, Keqin Li
Abstract: Multimodal emotion recognition in conversations aims to infer utterance-level emotions by jointly modeling textual, acoustic, and visual cues within context. Despite recent progress, key challenges remain, including redundant cross-modal information, imperfect semantic alignment, and insufficient modeling of high-order speaker interactions. To address these issues, we propose a framework that combines dual-space feature disentanglement with dual-branch graph learning. A shared encoder and modality-specific encoders are used to separate modality-invariant and modality-specific representations. The invariant features are modeled by a Fourier graph neural network to capture global consistency and complementary patterns, with a frequency-domain contrastive objective to enhance discriminability. In parallel, a speaker-aware hypergraph is constructed over modality-specific features to model high-order interactions, along with a speaker-consistency constraint to maintain coherent semantics. Finally, the two branches are fused for utterance-level emotion prediction. Experiments on IEMOCAP and MELD demonstrate that the proposed method achieves superior performance over strong baselines, validating its effectiveness.
Authors: Hanchen David Wang, Diego Manzanas Lopez, Preston K. Robinette, Ipek Oguz, Taylor T. Johnson, Meiyi Ma
Abstract: As deep neural networks are deployed in safety-critical domains such as autonomous driving and medical diagnosis, stakeholders need explanations that are interpretable but also trustworthy with formal guarantees. Existing XAI methods fall short: heuristic attribution techniques (e.g., LIME, Integrated Gradients) highlight influential features but offer no mathematical guarantees about decision boundaries, while formal methods verify robustness yet remain untargeted, analyzing the nearest boundary regardless of whether it represents a critical risk. In safety-critical systems, not all misclassifications carry equal consequences; confusing a "Stop" sign for a "60 kph" sign is far more dangerous than confusing it with a "No Passing" sign. We introduce ViTaX (Verified and Targeted Explanations), a formal XAI framework that generates targeted semifactual explanations with mathematical guarantees. For a given input (class y) and a user-specified critical alternative (class t), ViTaX: (1) identifies the minimal feature subset most sensitive to the y->t transition, and (2) applies formal reachability analysis to guarantee that perturbing these features by epsilon cannot flip the classification to t. We formalize this through Targeted epsilon-Robustness, certifying whether a feature subset remains robust under perturbation toward a specific target class. ViTaX is the first method to provide formally guaranteed explanations of a model's resilience against user-identified alternatives. Evaluations on MNIST, GTSRB, EMNIST, and TaxiNet demonstrate over 30% fidelity improvement with minimal explanation cardinality.
Authors: Eleanor Wiesler
Abstract: This thesis is an exposition of Ollivier-Ricci Curvature of metric spaces as introduced by Yann Ollivier, which is based upon the 1-Wasserstein Distance and optimal transport theory. We present some of the major results and proofs that connect Ollivier-Ricci curvature with classical Ricci curvature of Riemannian manifolds, including extensions of various theoretical bounds and theorems such as Bonnet-Myers and Levy-Gromov. Then we shift to results introduced by Lin-Lu-Yau on an extension of Ollivier-Ricci curvature on graphs, as well as the work of Jost-Liu on proving various combinatorial bounds for graph Ollivier-Ricci curvature. At the end of this thesis we present novel ideas and proofs regarding extensions of these results to directed graphs, and finally applications of graph-based Ollivier-Ricci curvature to various algorithms in network science and graph machine learning.
Authors: Deep Shah, Sanket Badhe, Nehal Kathrotia, Priyanka Tiwari
Abstract: Large Language Models utilizing reasoning techniques improve task performance but incur significant latency and token costs due to verbose generation. Existing automatic prompt optimization(APO) frameworks target task accuracy exclusively at the expense of generating long reasoning traces. We propose Cost-Regularized Optimization of Prompts (CROP), an APO method that introduces regularization on response length by generating textual feedback in addition to standard accuracy feedback. This forces the optimization process to produce prompts that elicit concise responses containing only critical information and reasoning. We evaluate our approach on complex reasoning datasets, specifically GSM8K, LogiQA and BIG-Bench Hard. We achieved an 80.6\% reduction in token consumption while maintaining competitive accuracy, seeing only a nominal decline in performance. This presents a pragmatic solution for deploying token-efficient and cost-effective agentic AI systems in production pipelines.
Authors: Richard Wai Cheung Chan, Shanru Lin, Ya-nan Ma, Hao Chen, Liangjun Jiang, Wenqi Fan
Abstract: To address the unsustainable rise in public health expenditures, the Hong Kong SAR Government is shifting its strategic focus to primary healthcare and encouraging citizens to use community resources to self-manage their health. However, official clinical guidelines are fragmented across disparate departments and formats, creating significant access barriers. While general-purpose Large Language Models (LLMs) such as ChatGPT and DeepSeek offer potential solutions for information accessibility, they are prone to generating factually inaccurate content due to a lack of localized and domain-specific knowledge. To this end, we propose a Retrieval-Augmented Generation-Enhanced LLM system as Primary Healthcare Assistant (PriHA) in Hong Kong. Specifically, a tri-stage pipeline is proposed that leverages a query optimizer to generalize user intent-oriented sub-queries, followed by a novel Dual Retrieval Augmented Generation (DRAG) architecture for mixed-source retrieval and context-reorganized generation. Comprehensive experiments and a detailed case study demonstrate that our proposed method can outperform both ablations and baseline in terms of accuracy and clarity. Our research provides a reliable and traceable dialogue retrieval framework for exploring other high-risk, localized application scenarios.
Authors: Aizierjiang Aiersilan, Mohamad Koubeissi
Abstract: Predicting post-surgical seizure outcomes in pharmacoresistant epilepsy is a clinical challenge. Conventional deep-learning approaches operate on static, single-timepoint pre-operative scans, omitting longitudinal morphological changes. We propose \emph{Neuro-Oracle}, a three-stage framework that: (i) distils pre-to-post-operative MRI changes into a compact 512-dimensional trajectory vector using a 3D Siamese contrastive encoder; (ii) retrieves historically similar surgical trajectories from a population archive via nearest-neighbour search; and (iii) synthesises a natural-language prognosis grounded in the retrieved evidence using a quantized Llama-3-8B reasoning agent. Evaluations are conducted on the public EPISURG dataset ($N{=}268$ longitudinally paired cases) using five-fold stratified cross-validation. Since ground-truth seizure-freedom scores are unavailable, we utilize a clinical proxy label based on the resection type. We acknowledge that the network representations may potentially learn the anatomical features of the resection cavities (i.e., temporal versus non-temporal locations) rather than true prognostic morphometry. Our current evaluation thus serves mainly as a proof-of-concept for the trajectory-aware retrieval architecture. Trajectory-based classifiers achieve AUC values between 0.834 and 0.905, compared with 0.793 for a single-timepoint ResNet-50 baseline. The Neuro-Oracle agent (M5) matches the AUC of purely discriminative trajectory classifiers (0.867) while producing structured justifications with zero observed hallucinations under our audit protocol. A Siamese Diversity Ensemble (M6) of trajectory-space classifiers attains an AUC of 0.905 without language-model overhead.
Authors: Samir Wagle, Reewaj Khanal, Abiral Adhikari
Abstract: Hate speech detection in Devanagari-scripted social media memes presents compounded challenges: multimodal content structure, script-specific linguistic complexity, and extreme data scarcity in low-resource settings. This paper presents our system for the CHiPSAL 2026 shared task, addressing both Subtask A (binary hate speech detection) and Subtask B (three-class sentiment classification: positive, neutral, negative). We propose a hybrid cross-modal attention fusion architecture that combines CLIP (ViT-B/32) for visual encoding with BGE-M3 for multilingual text representation, connected through 4-head self-attention and a learnable gating network that dynamically weights modality contributions on a per-sample basis. Systematic evaluation across eight model configurations demonstrates that explicit cross-modal reasoning achieves a 5.9% F1-macro improvement over text-only baselines on Subtask A, while uncovering two unexpected but critical findings: English-centric vision models exhibit near-random performance on Devanagari script, and standard ensemble methods catastrophically degrade under data scarcity (N nearly equal to 850 per fold) due to correlated overfitting. The code can be accessed at https://github.com/Tri-Yantra-Technologies/MEME-Fusion/
URLs: https://github.com/Tri-Yantra-Technologies/MEME-Fusion/
Authors: Koushik Chakraborty, Koyel Guha
Abstract: This research paper addresses the limitations of semantic search in complex enterprise document ecosystems. Traditional RAG pipelines often fail to capture hierarchical and interconnected information, leading to retrieval inaccuracies. We propose Agentic Knowledge Graphs featuring Recursive Crawling as a robust solution for navigating superseding logic and multi-hop references. Our benchmark evaluation using the Code of Federal Regulations (CFR) demonstrates that this Knowledge Graph-enhanced approach achieves a 70% accuracy improvement over standard vector-based RAG systems, providing exhaustive and precise answers for complex regulatory queries.
Authors: Afshan Hashmi
Abstract: Retrieval-Augmented Generation (RAG) has become the standard paradigm for grounding Large Language Model outputs in external knowledge. Lumer et al. [1] presented the first systematic evaluation comparing vector-based agentic RAG against hierarchical node-based reasoning systems for financial document QA across 1,200 SEC filings, finding vector-based systems achieved a 68% win rate. Concurrently, the PageIndex framework [2] demonstrated 98.7% accuracy on FinanceBench through purely reasoning-based retrieval. This paper extends their work by: (i) implementing and evaluating three retrieval architectures: Vector RAG, Tree Reasoning, and the proposed Adaptive Hybrid Retrieval (AHR) across financial, legal, and medical domains; (ii) introducing a four-tier query complexity benchmark; and (iii) employing GPT-4-powered LLM-as-judge evaluation. Experiments reveal that Tree Reasoning achieves the highest overall score (0.900), but no single paradigm dominates across all tiers: Vector RAG wins on multi-document synthesis (Tier 4, score 0.900), while the Hybrid AHR achieves the best performance on cross-reference (0.850) and multi-section queries (0.929). Cross-reference recall reaches 100% for tree-based and hybrid approaches versus 91.7% for vector search, quantifying a critical capability gap. Validation on FinanceBench (150 expert-annotated questions on real SEC 10-K and 10-Q filings) confirms and strengthens these findings: Tree Reasoning scores 0.938, Hybrid AHR 0.901, and Vector RAG 0.821, with the Tree--Vector quality gap widening to 11.7 percentage points on real-world documents. These findings support the development of adaptive retrieval systems that dynamically select strategies based on query complexity and document structure. All code and data are publicly available.
Authors: Ashmi Banerjee, Adithi Satish, Wolfgang W\"orndl, Yashar Deldjoo
Abstract: Traditional conversational travel recommender systems primarily optimize for user relevance and convenience, often reinforcing popular, overcrowded destinations and carbon-intensive travel choices. To address this, we present TRACE (Tourism Recommendation with Agentic Counterfactual Explanations), a multi-agent, LLM-based framework that promotes sustainable tourism through interactive nudging. TRACE uses a modular orchestrator-worker architecture where specialized agents elicit latent sustainability preferences, construct structured user personas, and generate recommendations that balance relevance with environmental impact. A key innovation lies in its use of agentic counterfactual explanations and LLM-driven clarifying questions, which together surface greener alternatives and refine understanding of intent, fostering user reflection without coercion. User studies and semantic alignment analyses demonstrate that TRACE effectively supports sustainable decision-making while preserving recommendation quality and interactive responsiveness. TRACE is implemented on Google's Agent Development Kit, with full code, Docker setup, prompts, and a publicly available demo video to ensure reproducibility. A project summary, including all resources, prompts, and demo access, is available at https://ashmibanerjee.github.io/trace-chatbot.
Authors: Sohyun An (Meta Superintelligence Labs, UCLA), Hayeon Lee (Meta Superintelligence Labs), Shuibenyang Yuan (Meta Superintelligence Labs), Chun-cheng Jason Chen (Meta Superintelligence Labs), Cho-Jui Hsieh (UCLA), Vijai Mohan (Meta Superintelligence Labs), Alexander Min (Meta Superintelligence Labs)
Abstract: Retrieval-Augmented Generation (RAG) is a key approach to mitigating the temporal staleness of large language models (LLMs) by grounding responses in up-to-date evidence. Within the RAG pipeline, re-rankers play a pivotal role in selecting the most useful documents from retrieved candidates. However, existing benchmarks predominantly evaluate re-rankers in static settings and do not adequately assess performance under evolving information -- a critical gap, as real-world systems often must choose among temporally different pieces of evidence. To address this limitation, we introduce FRESCO (Factual Recency and Evolving Semantic COnflict), a benchmark for evaluating re-rankers in temporally dynamic contexts. By pairing recency-seeking queries with historical Wikipedia revisions, FRESCO tests whether re-rankers can prioritize factually recent evidence while maintaining semantic relevance. Our evaluation reveals a consistent failure mode across existing re-rankers: a strong bias toward older, semantically rich documents, even when they are factually obsolete. We further investigate an instruction optimization framework to mitigate this issue. By identifying Pareto-optimal instructions that balance Evolving and Non-Evolving Knowledge tasks, we obtain gains of up to 27% on Evolving Knowledge tasks while maintaining competitive performance on Non-Evolving Knowledge tasks.
Authors: Jiacheng Liu, Xiaohan Zhao, Xinyi Shang, Zhiqiang Shen
Abstract: Claude Code is an agentic coding tool that can run shell commands, edit files, and call external services on behalf of the user. This study describes its comprehensive architecture by analyzing the publicly available TypeScript source code and further comparing it with OpenClaw, an independent open-source AI agent system that answers many of the same design questions from a different deployment context. Our analysis identifies five human values, philosophies, and needs that motivate the architecture (human decision authority, safety and security, reliable execution, capability amplification, and contextual adaptability) and traces them through thirteen design principles to specific implementation choices. The core of the system is a simple while-loop that calls the model, runs tools, and repeats. Most of the code, however, lives in the systems around this loop: a permission system with seven modes and an ML-based classifier, a five-layer compaction pipeline for context management, four extensibility mechanisms (MCP, plugins, skills, and hooks), a subagent delegation mechanism with worktree isolation, and append-oriented session storage. A comparison with OpenClaw, a multi-channel personal assistant gateway, shows that the same recurring design questions produce different architectural answers when the deployment context changes: from per-action safety classification to perimeter-level access control, from a single CLI loop to an embedded runtime within a gateway control plane, and from context-window extensions to gateway-wide capability registration. We finally identify six open design directions for future agent systems, grounded in recent empirical, architectural, and policy literature.
Authors: Sakthi Prabhu Gunasekar, Prasanna Kumar R
Abstract: Synthetic Aperture Radar (SAR) data is inherently complex-valued, while quantum machine learning (QML) models naturally operate in complex Hilbert spaces. This apparent alignment suggests that incorporating both magnitude and phase information into quantum encoding should improve performance in SAR Automatic Target Recognition (ATR). In this work, we systematically evaluate this assumption by comparing five quantum encoding strategies: magnitude-only, joint complex, I/Q-based, preprocessed phase, and pure quantum, under a unified experimental framework on the MSTAR benchmark dataset. Contrary to expectation, we observe a consistent pattern: in hybrid quantum-classical architectures, magnitude-only encoding outperforms all complex-valued strategies, achieving 99.57% accuracy on a 3-class task and 71.19% on an 8-class task, while phase-aware methods provide negligible (~0%) or negative improvements. In contrast, in purely quantum architectures with only 184-224 trainable parameters and no classical components, phase information becomes essential, contributing up to 21.65% improvement in accuracy. These results reveal that the utility of phase information is not inherent to the data, but depends critically on the model architecture. Hybrid models rely on classical components that compensate for missing phase information, whereas purely quantum models require phase to construct discriminative representations. Our findings provide practical design guidelines for encoding complex-valued data in QML and highlight the importance of encoding-architecture co-design in the NISQ era.
Authors: Mohammad Nasir Uddin, Md Munna Aziz
Abstract: Financial crime costs U.S. institutions over $32 billion each year. Although AI tools for fraud detection have become more advanced, their use in real-world systems still faces a major obstacle: many of these models operate as black boxes that cannot provide the transparent, auditable explanations required by regulations such as OCC Bulletin 2011-12 and Federal Reserve SR 11-7. This study makes three main contributions. First, it offers a thorough evaluation of explanation quality across faithfulness (sufficiency and comprehensiveness at k=5, 10, and 15) and stability (Kendall's W across 30 bootstrap samples). XGBoost paired with TreeExplainer achieves near-perfect stability (W=0.9912), while LSTM with DeepExplainer shows weak results (W=0.4962). Second, the paper introduces the SHAP-Guided Adaptive Ensemble (SGAE), which dynamically adjusts per-transaction ensemble weights based on SHAP attribution agreement, achieving the highest AUC-ROC among all tested models (0.8837 held-out; 0.9245 cross-validation). Third, a complete three-architecture evaluation of LSTM, Transformer, and GNN-GraphSAGE on the full 590,540-transaction IEEE-CIS dataset is provided, with GNN-GraphSAGE achieving AUC-ROC 0.9248 and F1=0.6013. All results are mapped directly to OCC, SR 11-7, and BSA-AML regulatory compliance requirements.
Authors: Mohammad Nasir Uddin
Abstract: The Spatial-Temporal Graph Attention Network (ST-GAT) framework was created to serve as an explainable GNN-based solution for detecting bank distress early warning signs and for conducting macro-prudential surveillance of the interbank system in the United States. The ST-GAT framework models 8,103 FDIC insured institutions across 58 quarterly snapshots (2010Q1-2024Q2). Bilateral exposures were reconstructed from publicly available FDIC Call Reports using maximum entropy estimation to produce a dynamic directed weighted graph. The framework achieves the highest AUPRC among all GNN architectures (0.939 +/- 0.010), trailing only XGBoost (0.944). Ablation analysis confirms the BiLSTM temporal component contributes +0.020 AUPRC; temporal attention weights exhibit a monotonically decreasing pattern consistent with long-run structural vulnerability weighting. Permutation importance identifies ROA (0.309) and NPL Ratio (0.252) as dominant predictors, consistent with post-mortem analyses of the 2023 regional banking crisis. All data are publicly available FDIC Call Reports and FRED series; all code and results are released.
Authors: Wei He, Wensheng Gan, Philip S. Yu
Abstract: Fraud detection on graph data can be viewed as a demanding task that requires distinguishing between different types of nodes. Because graph neural networks (GNNs) are naturally suited for processing information encoded in graph form through their message-passing operations, methods based on GNN models have increasingly attracted attention in the fraud detection domain. However, fraud graphs inherently exhibit relation camouflage, high heterophily, and class imbalance, causing most GNNs to underperform in fraud detection tasks. To address these challenges, this paper proposes a Graph-Based Fraud Detection Model with Dual-Path Graph Filtering (DPF-GFD). DPF-GFD first applies a beta wavelet-based operator to the original graph to capture key structural patterns. It then constructs a similarity graph from distance-based node representations and applies an improved low-pass filter. The embeddings from the original and similarity graphs are fused through supervised representation learning to obtain node features, which are finally used by an ensemble tree model to assess the fraud risk of unlabeled nodes. Unlike existing single-graph smoothing approaches, DPF-GFD introduces a frequency-complementary dual-path filtering paradigm tailored for fraud detection, explicitly decoupling structural anomaly modeling and feature similarity modeling. This design enables more discriminative and stable node representations in highly heterophilous and imbalanced fraud graphs. Comprehensive experiments on four real-world financial fraud detection datasets demonstrate the effectiveness of our proposed method.
Authors: Sourav Ganguly, Kartik Pandit, Arnob Ghosh
Abstract: Real-world decision-making systems operate in environments where state transitions depend not only on the agent's actions, but also on \textbf{exogenous factors outside its control}--competing agents, environmental disturbances, or strategic adversaries--formally, $s_{h+1} = f(s_h, a_h, \bar{a}_h)+\omega_h$ where $\bar{a}_h$ is the adversary/external action, $a_h$ is the agent's action, and $\omega_h$ is an additive noise. Ignoring such factors can yield policies that are optimal in isolation but \textbf{fail catastrophically in deployment}, particularly when safety constraints must be satisfied. Standard Constrained MDP formulations assume the agent is the sole driver of state evolution, an assumption that breaks down in safety-critical settings. Existing robust RL approaches address this via distributional robustness over transition kernels, but do not explicitly model the \textbf{strategic interaction} between agent and exogenous factor, and rely on strong assumptions about divergence from a known nominal model. We model the exogenous factor as an \textbf{adversarial policy} $\bar{\pi}$ that co-determines state transitions, and ask how an agent can remain both optimal and safe against such an adversary. \emph{To the best of our knowledge, this is the first work to study safety-constrained RL under explicit adversarial dynamics}. We propose \textbf{Robust Hallucinated Constrained Upper-Confidence RL} (\texttt{RHC-UCRL}), a model-based algorithm that maintains optimism over both agent and adversary policies, explicitly separating epistemic from aleatoric uncertainty. \texttt{RHC-UCRL} achieves sub-linear regret and constraint violation guarantees.
Authors: Wentao Hu, Yanbo Zhai, Xiaohui Hu, Mingkuan Zhao, Shanhong yu, Xue Liu, Kaidong Yu, Shuangyong Song, Xuelong Li
Abstract: Sparse Mixture-of-Experts (MoE) models have achieved remarkable scalability, yet they remain vulnerable to hallucinations, particularly when processing long-tail knowledge. We identify that this fragility stems from static Top-$k$ routing: routers tend to favor high-frequency patterns over rare factual associations. Consequently, ``specialist experts'' possessing critical long-tail knowledge are often assigned low gating scores and remain ``dormant'' -- under-prioritized for specific tokens despite their proven causal importance on other inputs. To address this, we propose Counterfactual Routing (CoR), a training-free inference framework designed to awaken these dormant experts. CoR integrates layer-wise perturbation analysis with the Counterfactual Expert Impact (CEI) metric to dynamically shift computational resources from syntax-dominant to knowledge-intensive layers while maintaining a constant total activation count, effectively retrieving causally decisive experts via virtual ablation. Extensive experiments on TruthfulQA, FACTOR, and TriviaQA demonstrate that CoR improves factual accuracy by 3.1\% on average without increasing the inference budget, establishing a superior Pareto frontier compared to static scaling strategies.
Authors: To Eun Kim, Alireza Salemi, Hamed Zamani, Fernando Diaz
Abstract: Modern information access ecosystems consist of mixtures of systems, such as retrieval systems and large language models, and increasingly rely on marketplaces to mediate access to models, tools, and data, making competition between systems inherent to deployment. In such settings, outcomes are shaped not only by benchmark quality but also by competitive pressure, including user switching, routing decisions, and operational constraints. Yet evaluation is still largely conducted on static benchmarks with accuracy-focused measures that assume systems operate in isolation. This mismatch makes it difficult to predict post-deployment success and obscures competitive effects such as early-adoption advantages and market dominance. We introduce Marketplace Evaluation, a simulation-based paradigm that evaluates information access systems as participants in a competitive marketplace. By simulating repeated interactions and evolving user and agent preferences, the framework enables longitudinal evaluation and marketplace-level metrics, such as retention and market share, that complement and can extend beyond traditional accuracy-based metrics. We formalize the framework and outline a research agenda, motivated by business and economics, around marketplace simulation, metrics, optimization, and adoption in evaluation campaigns like TREC.
Authors: Zhuofeng Li, Yi Lu, Dongfu Jiang, Haoxiang Zhang, Yuyang Bai, Chuan Li, Yu Wang, Shuiwang Ji, Jianwen Xie, Yu Zhang
Abstract: The rapid rise in AI conference submissions has driven increasing exploration of large language models (LLMs) for peer review support. However, LLM-based reviewers often generate superficial, formulaic comments lacking substantive, evidence-grounded feedback. We attribute this to the underutilization of two key components of human reviewing: explicit rubrics and contextual grounding in existing work. To address this, we introduce REVIEWBENCH, a benchmark evaluating review text according to paper-specific rubrics derived from official guidelines, the paper's content, and human-written reviews. We further propose REVIEWGROUNDER, a rubric-guided, tool-integrated multi-agent framework that decomposes reviewing into drafting and grounding stages, enriching shallow drafts via targeted evidence consolidation. Experiments on REVIEWBENCH show that REVIEWGROUNDER, using a Phi-4-14B-based drafter and a GPT-OSS-120B-based grounding stage, consistently outperforms baselines with substantially stronger/larger backbones (e.g., GPT-4.1 and DeepSeek-R1-670B) in both alignment with human judgments and rubric-based review quality across 8 dimensions. The code is available \href{https://github.com/EigenTom/ReviewGrounder}{here}.
Authors: Yangyue Wang, Harshvardhan Sikka, Yash Mathur, Tony Zhou, Jinu Nyachhyon, Pranav Guruprasad
Abstract: GUI grounding models report over 85% accuracy on standard benchmarks, yet drop 27-56 percentage points when instructions require spatial reasoning rather than direct element naming. Current benchmarks miss this because they evaluate each screenshot once with a single fixed instruction. We introduce GUI-Perturbed, a controlled perturbation framework that independently varies visual scenes and instructions to measure grounding robustness. Evaluating three 7B models from the same architecture lineage, we find that relational instructions cause systematic accuracy collapse across all models, a 70% browser zoom produces statistically significant degradation, and rank-8 LoRA fine-tuning with augmented data degrades performance rather than improving it. By perturbing along independent axes, GUI-Perturbed isolates which specific capability axes are affected-spatial reasoning, visual robustness, reasoning calibration-providing diagnostic signal that aggregate benchmarks cannot. We release the dataset, augmentation pipeline, and a fine-tuned model.
Authors: Haoran Xu, Kaiwen Hu, Somayeh Sojoudi, Amy Zhang
Abstract: We study behavior-regularized reinforcement learning (RL), where regularization toward a reference distribution (the dataset in offline RL or the base model in LLM RL finetuning) is essential to prevent value over-optimization caused by erroneous out-of-distribution extrapolation. Existing methods either rely on reparameterized policy gradient, which are difficult to scale to large generative models, or on reject sampling, which can be overly conservative when attempting to move beyond the behavior support. In this paper, we propose Value Gradient Flow (VGF), a scalable new paradigm for behavior-regularized RL. VGF casts behavior-regularized RL as an optimal transport problem that maps the reference distribution to the value-induced optimal policy distribution. We solve this transport problem via discrete gradient flow, where value gradients guide particles initialized from the reference distribution. Our analysis shows that VGF imposes regularization implicitly by controlling the transport budget. VGF eliminates explicit policy parameterization while remaining expressive and flexible, this enables adaptive test-time scaling by adjusting the transport budget. Extensive experiments demonstrate that VGF significantly outperforms prior methods, achieving state-of-the-art results on offline RL benchmarks (D4RL, OGBench) and LLM RL tasks. Code and runs can be found at https://ryanxhr.github.io/vgf.
Authors: Junzhe Wang, Zhiheng Xi, yajie yang, Hao Luo, Shihan Dou, Tao Gui, Qi Zhang
Abstract: Search agents extend Large Language Models (LLMs) beyond static parametric knowledge by enabling access to up-to-date and long-tail information unavailable during pretraining. While reinforcement learning has been widely adopted for training such agents, existing approaches face key limitations: process supervision often suffers from unstable value estimation, whereas outcome supervision struggles with credit assignment due to sparse, trajectory-level rewards. To bridge this gap, we propose Contribution-Weighted GRPO (CW-GRPO), a framework that integrates process supervision into group relative policy optimization. Instead of directly optimizing process rewards, CW-GRPO employs an LLM judge to assess the retrieval utility and reasoning correctness at each search round, producing per-round contribution scores. These scores are used to rescale outcome-based advantages along the trajectory, enabling fine-grained credit assignment without sacrificing optimization stability. Experiments on multiple knowledge-intensive benchmarks show that CW-GRPO outperforms standard GRPO by 5.0\% on Qwen3-8B and 6.3\% on Qwen3-1.7B, leading to more effective search behaviors. Additional analysis reveals that successful trajectories exhibit concentrated contributions across rounds, providing empirical insight into search agent tasks.
Authors: Guillermo Valverde, Igor Garc\'ia-Olaizola, Giannicola Scarpa, Alejandro Pozas-Kerstjens
Abstract: Tensor networks were developed in the context of many-body physics as compressed representations of multiparticle quantum states. These representations mitigate the exponential complexity of many-body systems by capturing only the most relevant dependencies. Due to the formal similarity between quantum entanglement and statistical correlations, tensor networks have recently been integrated in machine learning, operating both as alternative learning architectures and as decompositions of components of neural networks. The expectation is that the theoretical understanding of tensor networks developed within quantum many-body physics leads to novel methods that offer advantages in terms of computational efficiency, explainability, or privacy. Here we review the use of tensor networks in the context of machine learning, providing a critical assessment of the state of the art, the potential advantages, and the challenges that must be overcome.
Authors: Francesco Andrea Causio, Vittorio De Vita, Olivia Riccomi, Michele Ferramola, Federico Felizzi, Antonio Cristiano, Lorenzo De Mori, Chiara Battipaglia, Melissa Sawaya, Luigi De Angelis, Marcello Di Pumpo, Alessandra Piscitelli, Pietro Eric Risuleo, Alessia Longo, Giulia Vojvodic, Mariapia Vassalli, Bianca Destro Castaniti, Nicol\`o Scarsi, Manuel Del Medico
Abstract: While Large Language Models (LLMs) have demonstrated high proficiency on English-centric medical examinations, their performance often declines when faced with non-English languages and multimodal diagnostic tasks. This study protocol describes the development of EuropeMedQA, the first comprehensive, multilingual, and multimodal medical examination dataset sourced from official regulatory exams in Italy, France, Spain, and Portugal. Following FAIR data principles and SPIRIT-AI guidelines, we describe a rigorous curation process and an automated translation pipeline for comparative analysis. We evaluate contemporary multimodal LLMs using a zero-shot, strictly constrained prompting strategy to assess cross-lingual transfer and visual reasoning. EuropeMedQA aims to provide a contamination-resistant benchmark that reflects the complexity of European clinical practices and fosters the development of more generalizable medical AI.
Authors: Li-Hsiang Shen, Yu-Quan Zheng
Abstract: To address high data traffic demands of sixth-generation (6G) networks, this paper proposes a novel architecture that integrates autonomous aerial vehicles (AAVs) and multi-functional reconfigurable intelligent surfaces (MF-RISs) as AM-RIS in fluid antenna (FA)-assisted full-duplex (FD) networks. The AM-RIS provides hybrid functionalities, including signal reflection, amplification, and energy harvesting (EH), potentially improving both signal coverage and sustainability. Meanwhile, FA facilitates fine-grained spatial adaptability at FD-enabled base station (BS), which complements residual self-interference (SI) suppression. We aim at maximizing the overall energy efficiency (EE) by jointly optimizing transmit DL beamforming at BS, UL user power, configuration of AM-RIS, and positions of the FA and AM-RIS. Owing to the hybrid continuous-discrete parameters and high dimensionality of the intractable problem, we have conceived a self-optimized multi-agent hybrid deep reinforcement learning (DRL) framework (SOHRL), which integrates multi-agent deep Q-networks (DQN) and multi-agent proximal policy optimization (PPO), respectively handling discrete and continuous actions. To enhance self-adaptability, an attention-driven state representation and meta-level hyperparameter optimization are incorporated, enabling multi-agents to autonomously adjust learning hyperparameters. Simulation results validate the effectiveness of the proposed AM-RIS-enabled FA-aided FD networks empowered by SOHRL algorithm. The results reveal that SOHRL outperforms benchmarks of the case without attention mechanism and conventional hybrid/multi-agent/standalone DRL. Moreover, AM-RIS in FD achieves the highest EE compared to half-duplex, conventional rigid antenna arrays, partial EH, and conventional RIS without amplification, highlighting its potential as a compelling solution for EE-aware wireless networks.
Authors: Gabriel Pimenta de Freitas Cardoso, Caio Lucas da Silva Chacon, Jonas Felipe da Fonseca Oliveira, Paulo Henrique de Medeiros Araujo
Abstract: This manuscript introduces DharmaOCR Full and Lite, a pair of specialized small language models (SSLMs) for structured OCR that jointly optimize transcription quality, generation stability, and inference cost. It also presents DharmaOCR-Benchmark, a benchmark that covers printed, handwritten, and legal/administrative documents, and proposes a unified evaluation protocol that measures fidelity and structure while explicitly tracking text degeneration as a first-class benchmark metric (alongside unit cost). Beyond reporting degeneration rates, the manuscript empirically shows degeneration is not merely a quality failure, since it materially worsens production performance by increasing response time, reducing throughput, and inflating computational cost due to abnormally long generations. To the best of the author's knowledge, as a methodological contribution, this is the first application of Direct Preference Optimization (DPO) for OCR, explicitly using degenerate generations as rejected examples to penalize looping behavior. Combined with Supervised Fine-Tuning (SFT) for enforcing a strict JSON schema (header, margin, footer, and text), DPO consistently reduces degeneration rate across model families (up to 87.6% relative) while preserving or improving extraction quality. The resulting models, namely, DharmaOCR Full (7B) and DharmaOCR Lite (3B), set a new state-of-the-art on DharmaOCR-Benchmark, outperforming each open-source and commercial baseline model evaluated regarding extraction quality, reaching 0.925 and 0.911 scores with 0.40% and 0.20% degeneration rates. AWQ quantization reduced up to 22% per-page cost with negligible quality loss, enabling a strong quality-cost trade-off in comparison to proprietary OCR APIs and open-source alternatives.
Authors: Salem Radey, Jack West, Kassem Fawaz
Abstract: Agentic systems built on large language models (LLMs) are increasingly being used for complex security tasks, including binary reverse engineering (RE). Despite recent growth in popularity and capability, these systems continue to face limitations in realistic settings. Cutting-edge systems still fail in complex RE scenarios that involve obfuscation, timing, and unique architecture. In this work, we examine how agentic systems perform reverse engineering tasks with static, dynamic, and hybrid agents. Through an analysis of existing agentic tool usage, we identify several limitations, including token constraints, struggles with obfuscation, and a lack of program guardrails. From these findings, we outline current challenges and position future directions for system designers to overcome from a security perspective.
Authors: Bar Alon, Itamar Zimerman, Lior Wolf
Abstract: Large language models (LLMs) achieve strong performance and have revolutionized NLP, but their lack of explainability keeps them treated as black boxes, limiting their use in domains that demand transparency and trust. A promising direction to address this issue is post-hoc text-based explanations, which aim to explain model decisions in natural language. Prior work has focused on generating convincing rationales that appear to be subjectively faithful, but it remains unclear whether these explanations are epistemically faithful, whether they reflect the internal evidence the model actually relied on for its decision. In this paper, we first assess the epistemic faithfulness of LLM-generated explanations via counterfactuals and show that they are often unfaithful. We then introduce a training-free method that enhances faithfulness by guiding explanation generation through attention-level interventions, informed by token-level heatmaps extracted via a faithful attribution method. This method significantly improves epistemic faithfulness across multiple models, benchmarks, and prompts.
Authors: Aditi De
Abstract: Diffusion-model inference and overdamped Langevin dynamics are formally identical. A physical substrate that encodes the score function therefore equilibrates to the correct output by thermodynamics alone, requiring no digital arithmetic during inference and potentially achieving a $10{,}000\times$ reduction in energy relative to a GPU. Two fundamental barriers have until now prevented this equivalence from being realized at production scale: non-local skip connections, which locally coupled analog substrates cannot represent, and input conditioning, in which the coupling constants carry roughly $2{,}600\times$ too little signal to anchor the system to a specific input. We resolve both obstacles. \emph{Hierarchical bilinear coupling} encodes U-Net skip connections as rank-$k$ inter-module interactions derived directly from the singular structure of the encoder and decoder Gram matrices, requiring only $O(Dk)$ physical connections instead of $O(D^2)$. A \emph{minimal digital interface} -- a 4-dimensional bottleneck encoder together with a 16-unit transfer network, totalling \textbf{2,560 parameters} -- overcomes the conditioning barrier. When evaluated on activations drawn from a trained denoising U-Net, the complete system attains a decoder cosine similarity of \textbf{0.9906} against an oracle upper bound of 1.0000, while preserving theoretical net energy savings of approximately $10^7\times$ over GPU inference. These results constitute the first demonstration of trained-weight, production-scale thermodynamic diffusion inference.
Authors: Pushpa Kumar Balan, Aijing Feng
Abstract: Gradient saliency from deep sequence models surfaces candidate biomarkers efficiently, but the resulting gene lists are contaminated by tissue-composition confounders that degrade downstream classifiers. We study whether LLM chain-of-thought (CoT) reasoning can faithfully filter these confounders, and whether reasoning quality drives downstream performance. We train a Mamba SSM on TCGA-BRCA RNA-seq and extract the top-50 genes by gradient saliency; DeepSeek-R1 evaluates every candidate with structured CoT to produce a final 17-gene set. The raw 50-gene saliency set (no LLM) performs worse than a 5,000-gene variance baseline (AUC 0.832 vs. 0.903), while the LLM-filtered set surpasses it (AUC 0.927), using 294x fewer features. A faithfulness audit (COSMIC CGC, OncoKB, PAM50) reveals only 6 of 17 selected genes (35.3%) are validated BRCA biomarkers, yet 10 of 16 known BRCA genes in the input were missed - including FOXA1. This gap between downstream performance and reasoning faithfulness suggests selective faithfulness: targeted confounder removal is sufficient for performance gains even without comprehensive recall.
Authors: Tianhao Qian
Abstract: As search depth increases in autonomous reasoning and embodied planning, the candidate action space expands exponentially, heavily taxing computational budgets. While heuristic pruning is a common countermeasure, it operates without formal safety guarantees when surrogate models (like LLMs) exhibit systematic evaluation biases. This paper frames the node expansion process as a localized Best-Arm Identification (BAI) problem over dynamic frontiers, subject to a bounded systematic bias $L$. By inverting the Lambert W function, we establish an additive sample complexity of $\mathcal{O}((\Delta-4L)^{-2})$, which indicates that safe node elimination is only feasible when the empirical reward gap exceeds $4L$. We complement this with an information-theoretic lower bound of $\Omega((\Delta-2L)^{-2})$ to confirm the structural limits of biased search. Subsequent evaluations on both synthetic trees and complex reasoning tasks demonstrate that adhering to this local safety boundary successfully preserves optimal trajectories while maximizing sample allocation efficiency.
Authors: Apoorv Prasad, Susan McRoy
Abstract: Women with polycystic ovary syndrome (PCOS) face substantially elevated risks of body image distress, disordered eating, and metabolic challenges, yet existing natural language processing approaches for detecting these conditions lack transparency and cannot identify co-occurring presentations. We developed small, open-source language models to automatically detect this triple burden in social media posts with grounded explainability. We collected 1,000 PCOS-related posts from six subreddits, with two trained annotators labeling posts using guidelines operationalizing Lee et al. (2017) clinical framework. Three models (Gemma-2-2B, Qwen3-1.7B, DeepSeek-R1-Distill-Qwen-1.5B) were fine-tuned using Low-Rank Adaptation to generate structured explanations with textual evidence. The best model achieved 75.3 percent exact match accuracy on 150 held-out posts, with robust comorbidity detection and strong explainability. Performance declined with diagnostic complexity, indicating their best use is for screening rather than autonomous diagnosis.
Authors: Pratyay Banerjee, Masud Moshtaghi, Shivashankar Subramanian, Amita Misra, Ankit Chadha
Abstract: Large language models still struggle with reliable long-term conversational memory: simply enlarging context windows or applying naive retrieval often introduces noise and destabilizes responses. We present APEX-MEM, a conversational memory system that combines three key innovations: (1) a property graph which uses domain-agnostic ontology to structure conversations as temporally grounded events in an entity-centric framework, (2) append-only storage that preserves the full temporal evolution of information, and (3) a multi-tool retrieval agent that understands and resolves conflicting or evolving information at query time, producing a compact and contextually relevant memory summary. This retrieval-time resolution preserves the full interaction history while suppressing irrelevant details. APEX-MEM achieves 88.88% accuracy on LOCOMO's Question Answering task and 86.2% on LongMemEval, outperforming state-of-the-art session-aware approaches and demonstrating that structured property graphs enable more temporally coherent long-term conversational reasoning.
Authors: Akshay Paruchuri, Ishan Chatterjee, Henry Fuchs, Ehsan Adeli, Piotr Didyk
Abstract: Multimodal language models systematically underperform on visual perception tasks, yet the structure underlying this failure remains poorly understood. We propose centroid replacement, collapsing each token to its nearest K-means centroid, as a controlled probe for modal dependence. Across seven models spanning three architecture families, erasing text centroid structure costs 4$\times$ more accuracy than erasing visual centroid structure, exposing a universal imbalance where language representations overshadow vision even on tasks that demand visual reasoning. We exploit this asymmetry through text centroid contrastive decoding, recovering up to +16.9% accuracy on individual tasks by contrastively decoding against a text-centroid-erased reference. This intervention varies meaningfully with training approaches: standard fine-tuned models show larger gains (+5.6% on average) than preference-optimized models (+1.5% on average). Our findings suggest that modal competition is structurally localized, correctable at inference time without retraining, and quantifiable as a diagnostic signal to guide future multimodal training.
Authors: Xue Wu, Shengting Cao, Jiaqi Gong
Abstract: Rural environmental risks are shaped by place-based conditions (e.g., housing quality, road access, land-surface patterns), yet standard vulnerability indices are coarse and provide limited insight into risk contexts. We propose SatBLIP, a satellite-specific vision-language framework for rural context understanding and feature identification that predicts county-level Social Vulnerability Index (SVI). SatBLIP addresses limitations of prior remote sensing pipelines-handcrafted features, manual virtual audits, and natural-image-trained VLMs-by coupling contrastive image-text alignment with bootstrapped captioning tailored to satellite semantics. We use GPT-4o to generate structured descriptions of satellite tiles (roof type/condition, house size, yard attributes, greenery, and road context), then fine-tune a satellite-adapted BLIP model to generate captions for unseen images. Captions are encoded with CLIP and fused with LLM-derived embeddings via attention for SVI estimation under spatial aggregation. Using SHAP, we identify salient attributes (e.g., roof form/condition, street width, vegetation, cars/open space) that consistently drive robust predictions, enabling interpretable mapping of rural risk environments.
Authors: Noureddine Kermiche
Abstract: Catastrophic forgetting remains a primary hurdle in sequential task learning for artificial neural networks. We propose a silicon-native modular architecture that achieves structural parameter isolation using Task-Specific Experts and a distributed, outlier-based Gatekeeper. Moving beyond traditional sequential consolidation, our framework utilizes a Simultaneous Pipeline where Teacher learning, Student distillation, and Router manifold acquisition occur in parallel while raw data is present in a localized training session. This approach ensures computational efficiency and complies with privacy mandates like GDPR by deleting raw data as soon as a task is learned. We demonstrate that a Tight-Bottleneck Autoencoder (TB-AE) can effectively distinguish semantically crowded manifolds in high-dimensional latent spaces, overcoming the posterior collapse inherent to standard variational methods. By establishing strict topological boundaries, our TB-AE resolves latent space crowding in 4096-D LLM embeddings to provide a robust, unsupervised novelty signal. Furthermore, we validate an Autonomous Retrieval mechanism that confidently identifies returning manifolds, enabling stable lifelong learning without redundant module instantiation. Empirical results demonstrate that our ``Live Distillation'' approach acts as a natural regularizer, achieving strong retention across computer vision and natural language processing domains without suffering a student fidelity gap.
Authors: Qi Zhang, Dawei Wang, Shaofeng Zou
Abstract: Reinforcement learning (RL) has emerged as a powerful tool for aligning diffusion models with human preferences, typically by optimizing a single reward function under a KL regularization constraint. In practice, however, human preferences are inherently pluralistic, and aligned models must balance multiple downstream objectives, such as aesthetic quality and text-image consistency. Existing multi-objective approaches either rely on costly multi-objective RL fine-tuning or on fusing separately aligned models at denoising time, but they generally require access to reward values (or their gradients) and/or introduce approximation error in the resulting denoising objectives. In this paper, we revisit the problem of RL fine-tuning for diffusion models and address the intractability of identifying the optimal policy by introducing a step-level RL formulation. Building on this, we further propose Multi-objective Step-level Denoising-time Diffusion Alignment (MSDDA), a retraining-free framework for aligning diffusion models with multiple objectives, obtaining the optimal reverse denoising distribution in closed form, with mean and variance expressed directly in terms of single-objective base models. We prove that this denoising-time objective is exactly equivalent to the step-level RL fine-tuning, introducing no approximation error. Moreover, we provide numerical results, which indicate our method outperforms existing denoising-time approaches.
Authors: Dongxin Guo, Jikun Wu, Siu-Ming Yiu
Abstract: Large Language Model (LLM) agents are increasingly deployed in multi-agent systems requiring strategic coordination. While recent work has analyzed LLM behavior in two-player games, coalition formation, where $n$ agents dynamically form cooperative groups, remains theoretically uncharacterized. We present the first framework grounding coalition formation in LLM agent networks in hedonic game theory with formal stability guarantees. We introduce the LLM Coalition Formation Game (LCFG), establish sufficient conditions for Nash-stable partitions, and prove complexity results. Our analysis reveals that LLM agents exhibit bounded rationality characterized by $\epsilon$-rational preferences; we provide both deterministic existence guarantees and consistency-driven stability bounds whose predictions are consistent with empirical outcomes. Experiments with GPT-4, Claude-3, and Llama-3 across 2,400 episodes validate our framework: LLM coalitions achieve Nash stability in 73.2% of cases under our Coalition-of-Thought (CoalT) protocol, compared to 58.4% under chain-of-thought and 41.8% under standard prompting ($p < 0.001$). Our framework provides theoretical foundations for designing stable multi-agent LLM systems.
Authors: Hyunkyung Park, Arkaitz Zubiaga
Abstract: Automated fact-checking in dialogue involves multi-turn conversations where colloquial language is frequent yet understudied. To address this gap, we propose a conservative rewrite candidate for each response claim via staged de-colloquialisation, combining lightweight surface normalisation with scoped in-claim coreference resolution. We then introduce BiCon-Gate, a semantics-aware consistency gate that selects the rewrite candidate only when it is semantically supported by the dialogue context, otherwise falling back to the original claim. This gated selection stabilises downstream fact-checking and yields gains in both evidence retrieval and fact verification. On the DialFact benchmark, our approach improves retrieval and verification, with particularly strong gains on SUPPORTS, and outperforms competitive baselines, including a decoder-based one-shot LLM rewrite that attempts to perform all de-colloquialisation steps in a single pass.
Authors: David Basil, Chirooth Girigowda, Bradley Hauer, Sahir Momin, Ning Shi, Grzegorz Kondrak
Abstract: We study the task of automatically expanding WordNet-style lexical resources to new languages through sense generation. We generate senses by associating target-language lemmas with existing lexical concepts via semantic projection. Given a sense-tagged English corpus and its translation, our method projects English synsets onto aligned target-language tokens and assigns the corresponding lemmas to those synsets. To generate these alignments and ensure their quality, we augment a pre-trained base aligner with a bilingual dictionary, which is also used to filter out incorrect sense projections. We evaluate the method on multiple languages, comparing it to prior methods, as well as dictionary-based and large language model baselines. Results show that the proposed project-and-filter strategy improves precision while remaining interpretable and requiring few external resources. We plan to make our code, documentation, and generated sense inventories accessible.
Authors: Aodi Wu, Haodong Han, Xubo Luo, Ruisuo Wang, Shan He, Xue Wan
Abstract: Autonomous on-orbit servicing demands embodied agents that perceive through visual sensors, reason about 3D spatial situations, and execute multi-phase tasks over extended horizons. We present SpaceMind, a modular and self-evolving vision-language model (VLM) agent framework that decomposes knowledge, tools, and reasoning into three independently extensible dimensions: skill modules with dynamic routing, Model Context Protocol (MCP) tools with configurable profiles, and injectable reasoning-mode skills. An MCP-Redis interface layer enables the same codebase to operate across simulation and physical hardware without modification, and a Skill Self-Evolution mechanism distills operational experience into persistent skill files without model fine-tuning. We validate SpaceMind through 192 closed-loop runs across five satellites, three task types, and two environments, a UE5 simulation and a physical laboratory, deliberately including degraded conditions to stress-test robustness. Under nominal conditions all modes achieve 90--100% navigation success; under degradation, the Prospective mode uniquely succeeds in search-and-approach tasks where other modes fail. A self-evolution study shows that the agent recovers from failure in four of six groups from a single failed episode, including complete failure to 100% success and inspection scores improving from 12 to 59 out of 100. Real-world validation confirms zero-code-modification transfer to a physical robot with 100% rendezvous success. Code: https://github.com/wuaodi/SpaceMind
Authors: Mohammad R. Abu Ayyash
Abstract: We present Three-Phase Transformer (3PT), a residual-stream structural prior for decoder-only Transformers on a standard SwiGLU + RMSNorm + RoPE + GQA backbone. The hidden vector is partitioned into N equally-sized cyclic channels, each maintained by phase-respecting ops: a per-channel RMSNorm, a 2D Givens rotation between attention and FFN that rotates each channel by theta + i*(2*pi/N), and a head-count constraint aligning GQA heads with the partition. The architecture is a self-stabilizing equilibrium between scrambling and re-imposition, not a bolted-on module. The partition carves out a one-dimensional DC subspace orthogonal to the channels, into which we inject a fixed Gabriel's horn profile r(p) = 1/(p+1) as an absolute-position side-channel composing orthogonally with RoPE's relative-position rotation. The canonical N=3 borrows its metaphor from balanced three-phase AC, where three sinusoids 120 degrees apart sum to zero with no anti-correlated pair. At 123M parameters on WikiText-103, 3PT achieves -7.20% perplexity (-2.62% bits-per-byte) over a matched RoPE-Only baseline at +1,536 parameters (0.00124% of total), with 1.93x step-count convergence speedup (1.64x wall-clock). N behaves as a parameter-sharing knob rather than a unique optimum: at 5.5M an N-sweep over {1,2,3,4,6,8,12} is near-monotone with N=1 winning; at 123M a three-seed sweep finds N=3 and N=1 statistically indistinguishable. The load-bearing mechanism is the channel-partitioned residual stream, per-block rotation, per-phase normalization, and horn DC injection. We characterize (a) self-stabilization of the geometry without explicit enforcement, a novel instance of the conservation-law framework for neural networks; (b) a U-shaped depth profile of rotation-angle drift at 12 layers; (c) orthogonal composition with RoPE, attention, and FFN.
Authors: Vekil Bekmyradov, Noah C. P\"utz, Thomas Bartz-Beielstein
Abstract: Large Language Models (LLMs) have achieved impressive results on public benchmarks, often leading to claims of advanced reasoning and understanding. However, recent research in cognitive science reveals that these models sometimes rely on shallow heuristics and memorization, taking shortcuts rather than demonstrating genuine cognitive abilities. This paper investigates LLM behavior in automated test generation for software, contrasting performance on an open-source system (LevelDB) with SAP HANA, one of the most widely deployed commercial database systems worldwide, whose proprietary codebase is guaranteed to be absent from training data. We combine cognitive evaluation principles, drawing on Mitchell's mechanism-focused assessment methodology, with empirical software testing, employing mutation score and iterative compiler-feedback repair loops to assess both accuracy and underlying reasoning strategies. Results show that LLMs excel on familiar, open-source benchmarks but struggle with unseen, complex domains, often prioritizing compilability over semantic effectiveness. These findings provide independent software engineering evidence for the broader claim that current LLMs lack robust reasoning, and highlight the need for evaluation frameworks that penalize trivial shortcuts and reward true generalization.
Authors: Sang-Il Han
Abstract: We present an empirical study of whether hierarchically structured, shared-weight recurrence can match the representational quality of independent-layer stacking in a Transformer-based language model. HRM-LM replaces L independent Transformer layers with a two-speed recurrent pair: a Fast module operating at every step for local refinement, and a Slow module operating every T steps for global compression. This recurrent hierarchy is unrolled for M = N x T steps with shared parameters. The central and most robust finding, supported by a parameter-matched Universal Transformer ablation (UniTF, 1.2B) across five independent runs, is a sharp empirical gap between the two approaches.
Authors: Fortunatus Aabangbio Wulnye, Justice Owusu Agyemang, Kwame Opuni-Boachie Obour Agyekum, Kwame Agyeman-Prempeh Agyekum, Kingsford Sarkodie Obeng Kwakye, Francisca Adomaa Acheampong
Abstract: Ensuring the reliability of machine learning-based intrusion detection systems remains a critical challenge in Internet of Things (IoT) environments, particularly as data poisoning attacks increasingly threaten the integrity of model training pipelines. This study evaluates the susceptibility of four widely used classifiers, Random Forest, Gradient Boosting Machine, Logistic Regression, and Deep Neural Network models, against multiple poisoning strategies using three real-world IoT datasets. Results show that while ensemble-based models exhibit comparatively stable performance, Logistic Regression and Deep Neural Networks suffer degradation of up to 40% under label manipulation and outlier-based attacks. Such disruptions significantly distort decision boundaries, reduce detection fidelity, and undermine deployment readiness. The findings highlight the need for adversarially robust training, continuous anomaly monitoring, and feature-level validation within operational Network Intrusion Detection Systems. The study also emphasizes the importance of integrating resilience testing into regulatory and compliance frameworks for AI-driven IoT security. Overall, this work provides an empirical foundation for developing more resilient intrusion detection pipelines and informs future research on adaptive, attack-aware models capable of maintaining reliability under adversarial IoT conditions.
Authors: Xiaolei Diao, Fausto Giunchiglia
Abstract: Recent advances in data-centric artificial intelligence highlight inherent limitations in object recognition datasets. One of the primary issues stems from the semantic gap problem, which results in complex many-to-many mappings between visual data and linguistic descriptions. This bias adversely affects performance in computer vision tasks. This paper proposes an image annotation methodology that integrates knowledge representation, natural language processing, and computer vision techniques, aiming to reduce annotator subjectivity by applying visual property constraints. We introduce an interactive crowdsourcing framework that dynamically asks questions based on a predefined object category hierarchy and annotator feedback, guiding image annotation by visual properties. Experiments demonstrate the effectiveness of this methodology, and annotator feedback is discussed to optimize the crowdsourcing setup.
Authors: Biwei Dai, Po-Wen Chang, Wahid Bhimji, Paolo Calafiura, Ragansu Chakkappai, Yuan-Tang Chou, Sascha Diefenbacher, Jordan Dudley, Ibrahim Elsharkawy, Steven Farrell, Isabelle Guyon, Chris Harris, Elham E Khoda, Benjamin Nachman, David Rousseau, Uro\v{s} Seljak, Ihsan Ullah, Yulei Zhang
Abstract: Weak gravitational lensing, the correlated distortion of background galaxy shapes by foreground structures, is a powerful probe of the matter distribution in our universe and allows accurate constraints on the cosmological model. In recent years, high-order statistics and machine learning (ML) techniques have been applied to weak lensing data to extract the nonlinear information beyond traditional two-point analysis. However, these methods typically rely on cosmological simulations, which poses several challenges: simulations are computationally expensive, limiting most realistic setups to a low training data regime; inaccurate modeling of systematics in the simulations create distribution shifts that can bias cosmological parameter constraints; and varying simulation setups across studies make method comparison difficult. To address these difficulties, we present the first weak lensing benchmark dataset with several realistic systematics and launch the FAIR Universe Weak Lensing Machine Learning Uncertainty Challenge. The challenge focuses on measuring the fundamental properties of the universe from weak lensing data with limited training set and potential distribution shifts, while providing a standardized benchmark for rigorous comparison across methods. Organized in two phases, the challenge will bring together the physics and ML communities to advance the methodologies for handling systematic uncertainties, data efficiency, and distribution shifts in weak lensing analysis with ML, ultimately facilitating the deployment of ML approaches into upcoming weak lensing survey analysis.
Authors: S M Raihanul Alam, Md Dilshadur Rahman, Md Naimul Hoque
Abstract: Visualizing narratives is useful to writers to reflect on unfinished drafts and identify unintentional biases and inconsistencies. Literary scholars can use the visualizations to identify nuanced patterns and literary styles from written text. Current narrative visualization is limited to representing character and location co-occurrences in a timeline, omitting important and complex narrative components such as focalization, causality, and speech. This paper aims to capture and visualize underexplored, complex narrative components as a basis for narrative visualization. As a starting point, we propose a new narrative visualization, named FocalLens, that uses focalization, the component that establishes who sees or perceives the events in a narrative, for representing the narrative. We provide the theoretical foundation of focalization and describe various types and facets of focalization. The details are incorporated in the novel visualization that captures how different characters perceive an event, who directly participate in an event, who indirectly observe the event, and who narrate the event. We also developed a tool that provides fluid interaction between the text and the proposed visualization. The tool was evaluated with four writers and scholars in a qualitative study, where writers analyzed their draft stories and scholars analyzed well-known stories. The findings suggest the tool added a new dimension to the workflow for writers and scholars, an analytical lens that is not available otherwise. We conclude by identifying design implications and future directions.
Authors: Stavros Kassinos
Abstract: Physics-informed neural networks (PINNs) are often selected by a single scalar loss even when the quantity of interest is more specific. We study a hybrid design in which the governing PDE residual remains automatic-differentiation (AD) based, while finite differences (FD) appear only in a weak auxiliary term that penalizes gradients of the sampled residual field. The FD term regularizes the residual field without replacing the PDE residual itself. We examine this idea in two stages. Stage 1 is a controlled Poisson benchmark comparing a baseline PINN, the FD residual-gradient regularizer, and a matched AD residual-gradient baseline. Stage 2 transfers the same logic to a three-dimensional annular heat-conduction benchmark (PINN3D), where baseline errors concentrate near a wavy outer wall and the auxiliary grid is implemented as a body-fitted shell adjacent to the wall. In Stage 1, the FD regularizer reproduces the main effect of residual-gradient control while exposing a trade-off between field accuracy and residual cleanliness. In Stage 2, the shell regularizer improves the application-facing quantities, namely outer-wall flux and boundary-condition behavior. Across seeds 0-5 and 100k epochs, the most reliable tested configuration is a fixed shell weight of 5e-4 under the Kourkoutas-beta optimizer regime: relative to a matched run without the shell term, it reduces the mean outer-wall BC RMSE from 1.22e-2 to 9.29e-4 and the mean wall-flux RMSE from 9.21e-3 to 9.63e-4. Adam with beta2=0.999 becomes usable when the initial learning rate is reduced to 1e-3, although its shell benefit is less robust than under Kourkoutas-beta. Overall, the results support a targeted view of hybrid PINNs: an auxiliary-only FD regularizer is most valuable when it is aligned with the physical quantity of interest, here the outer-wall flux.
Authors: Junghoon Seo
Abstract: Behavior cloning (BC) policies on position-controlled robots inherit the closed-loop response of the underlying PD controller, yet the effect of controller gains on BC failure lacks a nonasymptotic theory. We show that independent sub-Gaussian action errors propagate through the gain-dependent closed-loop dynamics to yield sub-Gaussian position errors whose proxy matrix $X_\infty(K)$ governs the failure tail. The probability of horizon-$T$ task failure factorizes into a gain-dependent amplification index $\Gamma_T(K)$ and the validation loss plus a generalization slack, so training loss alone cannot predict closed-loop performance. Under shape-preserving upper-bound structural assumptions the proxy admits the scalar bound $X_\infty(K)\preceq\Psi(K)\bar X$ with $\Psi(K)$ decomposed into label difficulty, injection strength, and contraction, ranking the four canonical regimes with compliant-overdamped (CO) tightest, stiff-underdamped (SU) loosest, and the stiff-overdamped versus compliant-underdamped ordering system-dependent. For the canonical scalar second-order PD system the closed-form continuous-time stationary variance $X_\infty^{\mathrm{c}}(\alpha,\beta)=\sigma^2\alpha/(2\beta)$ is strictly monotone in stiffness and damping over the entire stable orthant, covering both underdamped and overdamped regimes, and the exact zero-order-hold (ZOH) discretization inherits this monotonicity. The analysis provides the first nonasymptotic explanation of the empirical finding that compliant, overdamped controllers improve BC success rates.
Authors: Ifayoyinsola Ibikunle, Tyler Farnan, Senthil Kumar, Mayana Pereira
Abstract: Financial institutions face tension between maximizing data utility and mitigating the re-identification risks inherent in traditional anonymization methods. This paper explores Differentially Private (DP) synthetic data as a robust "Privacy by Design" framework to resolve this conflict, ensuring output privacy while satisfying stringent regulatory obligations. We examine two distinct generative paradigms: Direct Tabular Synthesis, which reconstructs high-fidelity joint distributions from raw data, and DP-Seeded Agent-Based Modeling (ABM), which uses DP-protected aggregates to parameterize complex, stateful simulations. While tabular synthesis excels at reflecting static historical correlations for QA testing and business analytics, the DP-Seeded ABM offers a forward-looking "counterfactual laboratory" capable of modeling dynamic market behaviors and black swan events. By decoupling individual identities from data utility, these methodologies eliminate traditional data-clearing bottlenecks, enabling seamless cross-institutional research and compliant decision-making in an evolving regulatory landscape.
Authors: Nikola Zubi\'c, Qian Li, Yuyi Wang, Davide Scaramuzza
Abstract: We study the expressive power and limitations of multi-layer state-space models (SSMs). First, we show that multi-layer SSMs face fundamental limitations in compositional tasks, revealing an inherent gap between SSMs and streaming models. Then, we examine the role of chain-of-thought (CoT), showing that offline CoT does not fundamentally increase the expressiveness, while online CoT can substantially increase its power. Indeed, with online CoT, multi-layer SSMs become equivalent in power to streaming algorithms. Finally, we investigate the tradeoff between width and precision, showing that these resources are not interchangeable in the base model, but admit a clean equivalence once online CoT is allowed. Overall, our results offer a unified perspective on how depth, finite precision, and CoT shape the power and limits of SSMs.
Authors: Rongyao Wang, Veronica Liesaputra, Zhiyi Huang
Abstract: News recommender systems are devised to alleviate the information overload, attracting more and more researchers' attention in recent years. The lack of a dedicated learner-oriented news recommendation toolkit hinders the advancement of research in news recommendation. We propose a PyTorch-based news recommendation toolkit called NewsTorch, developed to support learners in acquiring both conceptual understanding and practical experience. This toolkit provides a modular, decoupled, and extensible framework with a learner-friendly GUI platform that supports dataset downloading and preprocessing. It also enables training, validation, and testing of state-of-the-art neural news recommendation models with standardized evaluation metrics, ensuring fair comparison and reproducible experiments. Our open-source toolkit is released on Github: https://github.com/whonor/NewsTorch.
Authors: Hugo O'Connor
Abstract: Agent communication languages (ACLs) enable heterogeneous agents to share knowledge and coordinate across diverse domains. This diversity demands extensibility, but expressive extension mechanisms can push the input language beyond the complexity classes where full validation is tractable. We present CBCL (Common Business Communication Language), an agent communication language that constrains all messages, including runtime language extensions, to the deterministic context-free language (DCFL) class. CBCL allows agents to define, transmit, and adopt domain-specific "dialect" extensions as first-class messages; three safety invariants (R1--R3), machine-checked in Lean 4 and enforced in a Rust reference implementation, prevent unbounded expansion, applying declared resource limits, and preserving core vocabulary. We formalize the language and its safety properties in Lean 4, implement a reference parser and dialect engine in Rust with property-based and differential tests, and extract a verified parser binary. Our results demonstrate that homoiconic protocol design, where extension definitions share the same representation as ordinary messages, can be made provably safe. As autonomous agents increasingly extend their own communication capabilities, formally bounding what they can express to each other is a precondition for oversight.
Authors: Honglin Guo, Rihao Chang, He Jiao, Weizhi Nie, Zhongheng Zhang, Yuehao Shen
Abstract: Accurate prediction of future risk and disease progression in sepsis is clinically important for early warning and timely intervention in intensive care. However, short-window sepsis prediction remains challenging, because shorter observation windows provide limited historical evidence, whereas longer prediction horizons reduce the number of patient trajectories with valid future supervision. To address this problem, we propose CSRA, a Controlled Spectral Residual Augmentation framework for short-window multi-system ICU time series. CSRA first groups variables by clinical systems and extracts system-level and global representations. It then performs input-adaptive residual perturbation in the spectral domain to generate structured and clinically plausible trajectory variations. To improve augmentation stability and controllability, CSRA is trained end-to-end with the downstream predictor under a unified objective, together with anchor consistency loss and controller regularization. Experiments on a MIMIC-IV sepsis cohort across multiple downstream models show that CSRA is consistently competitive and often superior, reducing regression error by 10.2\% in MSE and 3.7\% in MAE over the non-augmentation baseline, while also yielding consistent gains on classification. CSRA further maintains more favorable performance under shorter observation windows, longer prediction horizons, and smaller training data scales, while also remaining effective on an external clinical dataset~(ZiGongICUinfection), indicating stronger robustness and generalizability in clinically constrained settings.
Authors: Sazzadul Islam, Tasnim Tabassum, Hao Zheng
Abstract: Generating synthesizable Verilog for large, hierarchical hardware designs remains a significant challenge for large language models (LLMs), which struggle to replicate the structured reasoning that human experts employ when translating complex specifications into RTL. When tasked with producing hierarchical Verilog, LLMs frequently lose context across modules, hallucinate interfaces, fabricate inter-module wiring, and fail to maintain structural coherence - failures that intensify as design complexity grows and specifications involve informal prose, figures, and tables that resist direct operationalization. To address these challenges, we present VeriGraphi, a framework that introduces a spec-anchored Knowledge Graph as the architectural substrate driving the RTL generation pipeline. VeriGraphi constructs a HDA, a structured knowledge graph that explicitly encodes module hierarchy, port-level interfaces, wiring semantics, and inter-module dependencies as first-class graph entities and relations. Built through iterative multi-agent analysis of the specification, this Knowledge Graph provides a deterministic, machine-checkable structural scaffold before code generation. Guided by the KG, a progressive coding module incrementally generates pseudo-code and synthesizable RTL while enforcing interface consistency and dependency correctness at each submodule stage. We evaluate VeriGraphi on a benchmark of three representative specification documents from the National Institute of Standards and Technology and their corresponding implementations, and we present a RV32I processor as a detailed case study to illustrate the full pipeline. The results demonstrate that VeriGraphi enables reliable hierarchical RTL generation with minimal human intervention for RISC-V, marking a significant milestone for LLM-generated hardware design while maintaining strong functional correctness.
Authors: Xia Qi, Peishan Cong, Yichen Yao, Ziyi Wang, Yaoqin Ye, Yuexin Ma
Abstract: Video object insertion is a critical task for dynamically inserting new objects into existing environments. Previous video generation methods focus primarily on synthesizing entire scenes while struggling with ensuring consistent object appearance, spatial alignment, and temporal coherence when inserting objects into existing videos. In this paper, we propose a novel solution for Video Object Insertion, which integrates multi-view object priors to address the common challenges of appearance inconsistency and occlusion handling in dynamic environments. By lifting 2D reference images into multi-view representations and leveraging a dual-path view-consistent conditioning mechanism, our framework ensures stable identity guidance and robust integration across diverse viewpoints. A quality-aware weighting mechanism is also employed to adaptively handle noisy or imperfect inputs. Additionally, we introduce an Integration-Aware Consistency Module that guarantees spatial realism, effectively resolving occlusion and boundary artifacts while maintaining temporal continuity across frames. Experimental results show that our solution significantly improves the quality of video object insertion, providing stable and realistic integration.
Authors: Yiqun Sun, Pengfei Wei, Lawrence B. Hsieh
Abstract: Retrieval-Augmented Generation (RAG) grounds LLM responses in external evidence but treats the model as a passive consumer of search results: it never sees how the corpus is organized or what it has not yet retrieved, limiting its ability to backtrack or combine scattered evidence. We present Corpus2Skill, which distills a document corpus into a hierarchical skill directory offline and lets an LLM agent navigate it at serve time. The compilation pipeline iteratively clusters documents, generates LLM-written summaries at each level, and materializes the result as a tree of navigable skill files. At serve time, the agent receives a bird's-eye view of the corpus, drills into topic branches via progressively finer summaries, and retrieves full documents by ID. Because the hierarchy is explicitly visible, the agent can reason about where to look, backtrack from unproductive paths, and combine evidence across branches. On WixQA, an enterprise customer-support benchmark for RAG, Corpus2Skill outperforms dense retrieval, RAPTOR, and agentic RAG baselines across all quality metrics.
Authors: Cheng Lu, Mengxin Wang, Dennis J. Zhang, Heng Zhang
Abstract: Data-driven operations management often relies on parameters estimated from costly human-generated labels. Recent advances in large language models (LLMs) and other AI systems offer inexpensive auxiliary data, but introduce a new challenge: AI outputs are not direct observations of the target outcomes, but could involve high-dimensional representations with complex and unknown relationships to human labels. Conventional methods leverage AI predictions as direct proxies for true labels, which can be inefficient or unreliable when this relationship is weak or misspecified. We propose Generative Augmented Inference (GAI), a general framework that incorporates AI-generated outputs as informative features for estimating models of human-labeled outcomes. GAI uses an orthogonal moment construction that enables consistent estimation and valid inference with flexible, nonparametric relationship between LLM-generated outputs and human labels. We establish asymptotic normality and show a "safe default" property: relative to human-data-only estimators, GAI weakly improves estimation efficiency under arbitrary auxiliary signals and yields strict gains whenever the auxiliary information is predictive. Empirically, GAI outperforms benchmarks across diverse settings. In conjoint analysis with weak auxiliary signals, GAI reduces estimation error by about 50% and lowers human labeling requirements by over 75%. In retail pricing, where all methods access the same auxiliary inputs, GAI consistently outperforms alternative estimators, highlighting the value of its construction rather than differences in information. In health insurance choice, it cuts labeling requirements by over 90% while maintaining decision accuracy. Across applications, GAI improves confidence interval coverage without inflating width. Overall, GAI provides a principled and scalable approach to integrating AI-generated information.
Authors: Xiping Li, Aier Yang, Jianghong Ma, Kangzhe Liu, Shanshan Feng, Haijun Zhang, Yi Zhao
Abstract: The rapid expansion of gaming industry requires advanced recommender systems tailored to its dynamic landscape. Existing Graph Neural Network (GNN)-based methods primarily prioritize accuracy over diversity, overlooking their inherent trade-off. To address this, we previously proposed CPGRec, a balance-oriented gaming recommender system. However, CPGRec fails to account for critical disparities in player-game interactions, which carry varying significance in reflecting players' personal preferences and may exacerbate over-smoothness issues inherent in GNN-based models. Moreover, existing approaches underutilize the reasoning capabilities and extensive knowledge of large language models (LLMs) in addressing these limitations. To bridge this gap, we propose two new modules. First, Preference-informed Edge Reweighting (PER) module assigns signed edge weights to qualitatively distinguish significant player interests and disinterests while then quantitatively measuring preference strength to mitigate over-smoothing in graph convolutions. Second, Preference-informed Representation Generation (PRG) module leverages LLMs to generate contextualized descriptions of games and players by reasoning personal preferences from comparing global and personal interests, thereby refining representations of players and games. Experiments on \textcolor{black}{two Steam datasets} demonstrate CPGRec+'s superior accuracy and diversity over state-of-the-art models. The code is accessible at https://github.com/HsipingLi/CPGRec-Plus.
Authors: Shreesha G. Bhat, Tony Hong, Michael Noguera, Ramnatthan Alagappan, Aishwarya Ganesan
Abstract: In modern data-streaming systems, alongside traditional programs, a new type of entity has emerged that can interact with streaming data: AI agents. Unlike traditional programs, AI agents use LLM reasoning to accomplish high-level tasks specified in natural language over streaming data. Unfortunately, current streaming systems cannot fully support agents: they lack the fundamental mechanisms to avoid the performance interference caused by agentic tasks and to safely handle agentic writes. We argue that the shared log, the core abstraction underlying streaming data, must support creating forks of itself, and that such a forkable shared log serves as a great substrate for agents acting on streaming data. We propose AgileLog, a new shared log abstraction that provides novel forking primitives for agentic use cases. We design Bolt, an implementation of the AgileLog abstraction, that uses novel techniques to make forks cheap, and provide logical and performance isolation.
Authors: Yitong Shou, Manhao Guan
Abstract: While Large Language Models (LLMs) demonstrate increasingly sophisticated affective capabilities, the internal mechanisms by which they process complex emotions remain unclear. Existing interpretability approaches often treat models as black boxes or focus on coarse-grained basic emotions, leaving the cognitive structure of more complex affective states underexplored. To bridge this gap, we propose a Cognitive Reverse-Engineering framework based on Representation Engineering (RepE) to analyze social-comparison jealousy. By combining appraisal theory with subspace orthogonalization, regression-based weighting, and bidirectional causal steering, we isolate and quantify two psychological antecedents of jealousy, Superiority of Comparison Person and Domain Self-Definitional Relevance, and examine their causal effects on model judgments. Experiments on eight LLMs from the Llama, Qwen, and Gemma families suggest that models natively encode jealousy as a structured linear combination of these constituent factors. Their internal representations are broadly consistent with the human psychological construct, treating Superiority as the foundational trigger and Relevance as the ultimate intensity multiplier. Our framework also demonstrates that toxic emotional states can be mechanically detected and surgically suppressed, suggesting a possible route toward representational monitoring and intervention for AI safety in multi-agent environments.
Authors: Yian Wang, Yuen Chen, Agam Goyal, Hari Sundaram
Abstract: Large language models (LLMs) frequently generate toxic content, posing significant risks for safe deployment. Current mitigation strategies often degrade generation quality or require costly human annotation. We propose CAUSALDETOX, a framework that identifies and intervenes on the specific attention heads causally responsible for toxic generation. Using the Probability of Necessity and Sufficiency (PNS), we isolate a minimal set of heads that are necessary and sufficient for toxicity. We utilize these components via two complementary strategies: (1) Local Inference-Time Intervention, which constructs dynamic, input-specific steering vectors for context-aware detoxification, and (2) PNS-Guided Fine-Tuning, which permanently unlearns toxic representations. We also introduce PARATOX, a novel benchmark of aligned toxic/non-toxic sentence pairs enabling controlled counterfactual evaluation. Experiments on ToxiGen, ImplicitHate, and ParaDetox show that CAUSALDETOX achieves up to 5.34% greater toxicity reduction compared to baselines while preserving linguistic fluency, and offers a 7x speedup in head selection.
Authors: Meng Chen, Kun Wang, Li Lu, Jiaheng Zhang, Tianwei Zhang
Abstract: Modern Large audio-language models (LALMs) power intelligent voice interactions by tightly integrating audio and text. This integration, however, expands the attack surface beyond text and introduces vulnerabilities in the continuous, high-dimensional audio channel. While prior work studied audio jailbreaks, the security risks of malicious audio injection and downstream behavior manipulation remain underexamined. In this work, we reveal a previously overlooked threat, auditory prompt injection, under realistic constraints of audio data-only access and strong perceptual stealth. To systematically analyze this threat, we propose \textit{AudioHijack}, a general framework that generates context-agnostic and imperceptible adversarial audio to hijack LALMs. \textit{AudioHijack} employs sampling-based gradient estimation for end-to-end optimization across diverse models, bypassing non-differentiable audio tokenization. Through attention supervision and multi-context training, it steers model attention toward adversarial audio and generalizes to unseen user contexts. We also design a convolutional blending method that modulates perturbations into natural reverberation, making them highly imperceptible to users. Extensive experiments on 13 state-of-the-art LALMs show consistent hijacking across 6 misbehavior categories, achieving average success rates of 79\%-96\% on unseen user contexts with high acoustic fidelity. Real-world studies demonstrate that commercial voice agents from Mistral AI and Microsoft Azure can be induced to execute unauthorized actions on behalf of users. These findings expose critical vulnerabilities in LALMs and highlight the urgent need for dedicated defense.
Authors: Xiangrui Xiong, Hang Liang, Baiyang Chen, Zifei Pan, Yanli Lee
Abstract: Learning Path Recommendation (LPR) is critical for personalized education, yet current methods often fail to account for historical interaction uncertainty (e.g., lucky guesses or accidental slips) and lack adaptability to diverse learning goals. We propose U-GLAD (Uncertainty-aware Generative Learning Path Recommendation with Cognition-Adaptive Diffusion). To address representation bias, the framework models cognitive states as probability distributions, capturing the learner's underlying true state via a Gaussian LSTM. To ensure highly personalized recommendation, a goal-oriented concept encoder utilizes multi-head attention and objective-specific transformations to dynamically align concept semantics with individual learning goals, generating uniquely tailored embeddings. Unlike traditional discriminative ranking approaches, our model employs a generative diffusion model to predict the latent representation of the next optimal concept. Extensive evaluations on three public datasets demonstrate that U-GLAD significantly outperforms representative baselines. Further analyses confirm its superior capability in perceiving interaction uncertainty and providing stable, goal-driven recommendation paths.
Authors: Sumit Mukherjee, Juan Shu, Nairwita Mazumder, Tate Kernell, Celena Wheeler, Shannon Hastings, Chris Sidey-Gibbons
Abstract: Clinical value set authoring -- the task of identifying all codes in a standardized vocabulary that define a clinical concept -- is a recurring bottleneck in clinical quality measurement and phenotyping. A natural approach is to prompt a large language model (LLM) to generate the required codes directly, but structured clinical vocabularies are large, version-controlled, and not reliably memorized during pretraining. We propose Retrieval-Augmented Set Completion (RASC): retrieve the $K$ most similar existing value sets from a curated corpus to form a candidate pool, then apply a classifier to each candidate code. Theoretically, retrieve-and-select can reduce statistical complexity by shrinking the effective output space from the full vocabulary to a much smaller retrieved candidate pool. We demonstrate the utility of RASC on 11,803 publicly available VSAC value sets, constructing the first large-scale benchmark for this task. A cross-encoder fine-tuned on SAPBert achieves AUROC~0.852 and value-set-level F1~0.298, outperforming a simpler three-layer Multilayer Perceptron (AUROC~0.799, F1~0.250) and both reduce the number of irrelevant candidates per true positive from 12.3 (retrieval-only) to approximately 3.2 and 4.4 respectively. Zero-shot GPT-4o achieves value-set-level F1~0.105, with 48.6\% of returned codes absent from VSAC entirely. This performance gap widens with increasing value set size, consistent with RASC's theoretical advantage. We observe similar performance gains across two other classifier model types, namely a cross-encoder initialized from pre-trained SAPBert and a LightGBM model, demonstrating that RASC's benefits extend beyond a single model class. The code to download and create the benchmark dataset, as well as the model training code is available at: \href{https://github.com/mukhes3/RASC}{https://github.com/mukhes3/RASC}.
URLs: https://github.com/mukhes3/RASC, https://github.com/mukhes3/RASC
Authors: Sanidhya Vijayvargiya, Vijay Viswanathan, Graham Neubig
Abstract: Humans often specify tasks incompletely, so assistants must know when and how to ask clarifying questions. However, effective clarification remains challenging in software engineering tasks as not all missing information is equally valuable, and questions must target information users can realistically provide. We study clarification in real software engineering tasks by quantifying which types of information most affect task success and which questions elicit useful responses from simulated users. Using Shapley attribution and distributional comparisons, we identify two key properties of effective clarification: task relevance (which information predicts success) and user answerability (what users can realistically provide). We operationalize these properties as multi-stage reinforcement learning rewards to train CLARITI, an 8B-parameter clarification module, that matches GPT-5's resolution rate on underspecified issues while generating 41% fewer questions. Our results suggest that grounding reward design in empirical analysis of information impact and user answerability improves clarification efficiency.
Authors: Yuseon Choi, Jingu Lee, Jungjun Oh, Sunjoo Whang, Byeongcheol Kim, Minsung Kim, Hoi-Jun Yoo, Sangjin Kim
Abstract: Mixture-of-Experts (MoE) models have become the dominant architecture for large-scale language models, yet on-premises serving remains fundamentally memory-bound as batching turns sparse per-token compute into dense memory activation. Memory-centric architectures (PIM, NMP) improve bandwidth but leave compute underutilized under MoE's low arithmetic intensity at high batch sizes. Speculative decoding (SD) trades idle compute for fewer target invocations, yet verification must load experts even for rejected tokens, severely limiting its benefit in MoE especially at low batch sizes. We propose ELMoE-3D, a hybrid-bonding (HB)-based HW-SW co-designed framework that unifies cache-based acceleration and speculative decoding to offer overall speedup across batch sizes. We identify two intrinsic elasticity axes of MoE-expert and bit-and jointly scale them to construct Elastic Self-Speculative Decoding (Elastic-SD), which serves as both an expert cache and a strongly aligned self-draft model accelerated by high HB bandwidth. Our LSB-augmented bit-sliced architecture exploits inherent redundancy in bit-slice representations to natively support bit-nested execution. On our 3D-stacked hardware, ELMoE-3D achieves an average $6.6\times$ speedup and $4.4\times$ energy efficiency gain over naive MoE serving on xPU across batch sizes 1-16, and delivers $2.2\times$ speedup and $1.4\times$ energy efficiency gain over the best-performing prior accelerator baseline.
Authors: Geonhui Jang, Dongyoon Han, YoungJoon Yoo
Abstract: Effective code generation requires both model capability and a problem representation that carefully structures how models reason and plan. Existing approaches augment reasoning steps or inject specific structure into how models think, but leave scattered problem conditions unchanged. Inspired by the way humans organize fragmented information into coherent explanations, we propose StoryCoder, a narrative reformulation framework that transforms code generation questions into coherent natural language narratives, providing richer contextual structure than simple rephrasings. Each narrative consists of three components: a task overview, constraints, and example test cases, guided by the selected algorithm and genre. Experiments across 11 models on HumanEval, LiveCodeBench, and CodeForces demonstrate consistent improvements, with an average gain of 18.7% in zero-shot pass@10. Beyond accuracy, our analyses reveal that narrative reformulation guides models toward correct algorithmic strategies, reduces implementation errors, and induces a more modular code structure. The analyses further show that these benefits depend on narrative coherence and genre alignment, suggesting that structured problem representation is important for code generation regardless of model scale or architecture. Our code is available at https://github.com/gu-ni/StoryCoder.
Authors: Cuong Hoang, Le-Minh Nguyen
Abstract: The proliferation of financial misinformation poses a severe threat to market stability and investor trust, misleading market behavior and creating critical information asymmetry. Detecting such misleading narratives is inherently challenging, particularly in real-world scenarios where external evidence or supplementary references for cross-verification are strictly unavailable. This paper presents our winning methodology for the "Reference-Free Financial Misinformation Detection" shared task. Built upon the recently proposed RFC-BENCH framework (Jiang et al. 2026), this task challenges models to determine the veracity of financial claims by relying solely on internal semantic understanding and contextual consistency, rather than external fact-checking. To address this formidable evaluation setup, we propose a comprehensive framework that capitalizes on the reasoning capabilities of state-of-the-art Large Language Models (LLMs). Our approach systematically integrates in-context learning, specifically zero-shot and few-shot prompting strategies, with Parameter-Efficient Fine-Tuning (PEFT) via Low-Rank Adaptation (LoRA) to optimally align the models with the subtle linguistic cues of financial manipulation. Our proposed system demonstrated superior efficacy, successfully securing the first-place ranking on both official leaderboards. Specifically, we achieved an accuracy of 95.4% on the public test set and 96.3% on the private test set, highlighting the robustness of our method and contributing to the acceleration of context-aware misinformation detection in financial Natural Language Processing. Our models (14B and 32B) are available at https://huggingface.co/KaiNKaiho.
Authors: Anusree M, Akhila Henry, Pramod P Nair
Abstract: Convolutional neural networks (CNNs) often exhibit poor generalisation in limited training data scenarios due to overfitting and insufficient feature diversity. In this work, a simple and effective chaos-based feature transformation is proposed to enhance CNN performance without increasing model complexity. The method applies nonlinear transformations using logistic, skew tent, and sine maps to normalised feature vectors before the classification layer, thereby reshaping the feature space and improving class separability. The approach is evaluated on greyscale datasets (MNIST and Fashion-MNIST) and an RGB dataset (CIFAR-10) using CNN architectures of varying depth under limited data conditions. The results show consistent improvement over the standalone (SA) CNN across all datasets. Notably, a maximum performance gain of 5.43% is achieved on MNIST using the skew tent map with a 3-layer CNN at 40 samples per class. A higher gain of 9.11% is observed on Fashion-MNIST using the sine map with a 3-layer CNN at 50 samples per class. Additionally, a strong gain of 7.47% is obtained on CIFAR-10 using the skew tent map at 200 samples per class. The consistent improvements across different chaotic maps indicate that the performance gain is driven by the shared nonlinear and dynamical properties of chaotic systems. The proposed method is computationally efficient, requires no additional trainable parameters, and can be easily integrated into existing CNN architectures, making it a practical solution for data-scarce image classification tasks.
Authors: Inseok Jeon, Minhyeok Lee, Seunghoon Lee, Minseok Kang, Suhwan Cho, Sangyoun Lee
Abstract: Video outpainting aims to expand the visible content of a video beyond the original frame boundaries while preserving spatial fidelity and temporal coherence across frames. Existing methods primarily rely on large-scale generative models, such as diffusion models. However, generationbased approaches suffer from implicit temporal modeling and limited spatial context. These limitations lead to intraframe and inter-frame inconsistencies, which become particularly pronounced in dynamic scenes and large outpainting scenarios. To overcome these challenges, we propose Seen-to-Scene, a novel framework that unifies propagationbased and generation-based paradigms for video outpainting. Specifically, Seen-to-Scene leverages flow-based propagation with a flow completion network pre-trained for video inpainting, which is fine-tuned in an end-to-end manner to bridge the domain gap and reconstruct coherent motion fields. To further improve the efficiency and reliability of propagation, we introduce a reference-guided latent propagation that effectively propagates source content across frames. Extensive experiments demonstrate that our method achieves superior temporal coherence and visual realism with efficient inference, surpassing even prior state-of-the-art methods that require input-specific adaptation.
Authors: Jianhao Su, Zhanwei Wu, ShengTing Huang, Weidong Feng
Abstract: Edge AI model deployment is a multi-stage engineering process involving model conversion, operator compatibility handling, quantization calibration, runtime integration, and accuracy validation. In practice, this workflow is long, failure-prone, and heavily dependent on deployment expertise, particularly when targeting hardware-specific inference runtimes. This technical report presents AIPC (AI Porting Conversion), an AI agent-driven approach for constrained automation of AI model deployment. AIPC decomposes deployment into standardized, verifiable stages and injects deployment-domain knowledge into agent execution through Agent Skills, helper scripts, and a stage-wise validation loop. This design reduces both the expertise barrier and the engineering time required for hardware deployment. Using Qualcomm AI Runtime (QAIRT) as the primary scenario, this report examines automated deployment across representative vision, multimodal, and speech models. In the cases covered here, AIPC can complete deployment from PyTorch to runnable QNN/SNPE inference within 7-20 minutes for structurally regular vision models, with indicative API costs roughly in the range of USD 0.7-10. For more complex models involving less-supported operators, dynamic shapes, or autoregressive decoding structures, fully automated deployment may still require further advances, but AIPC already provides practical support for execution, failure localization, and bounded repair.
Authors: Sarmad Sohail, Ghufran Haider
Abstract: Large language models are increasingly used as natural-language interfaces to enterprise software, but their direct use as system operators remains unsafe. Model errors can propagate into unauthorized actions, malformed requests, cross-workspace execution, and other costly failures. We argue this is primarily an execution architecture problem. We present a bounded-autonomy architecture in which language models may interpret intent and propose actions, but all executable behavior is constrained by typed action contracts, permission-aware capability exposure, scoped context, validation before side effects, consumer-side execution boundaries, and optional human approval. The enterprise application remains the source of truth for business logic and authorization, while the orchestration engine operates over an explicit published actions manifest. We evaluate the architecture in a deployed multi-tenant enterprise application across three conditions: manual operation, unconstrained AI with safety layers disabled, and full bounded autonomy. Across 25 scenario trials spanning seven failure families, the bounded-autonomy system completed 23 of 25 tasks with zero unsafe executions, while the unconstrained configuration completed only 17 of 25. Two wrong-entity mutations escaped all consumer-contributed layers; only disambiguation and confirmation mechanisms intercept this class. Both AI conditions delivered 13-18x speedup over manual operation. Critically, removing safety layers made the system less useful: structured validation feedback guided the model to correct outcomes in fewer turns, while the unconstrained system hallucinated success. Several safety properties are structurally enforced by code and intercepted all targeted violations regardless of model output. The result is a practical, deployed architecture for making imperfect language models operationally useful in enterprise systems.
Authors: Jiaqi Zhu, Shaofeng Cai, Jie Chen, Fang Deng, Beng Chin Ooi, Wenqiao Zhang
Abstract: Online anomaly detection (OAD) plays a pivotal role in real-time analytics and decision-making for evolving data streams. However, existing methods often rely on costly retraining and rigid decision boundaries, limiting their ability to adapt both effectively and efficiently to concept drift in dynamic environments. To address these challenges, we propose DyMETER, a dynamic concept adaptation framework for OAD that unifies on-the-fly parameter shifting and dynamic thresholding within a single online paradigm. DyMETER first learns a static detector on historical data to capture recurring central concepts, and then transitions to a dynamic mode to adapt to new concepts as drift occurs. Specifically, DyMETER employs a novel dynamic concept adaptation mechanism that leverages a hypernetwork to generate instance-aware parameter shifts for the static detector, thereby enabling efficient and effective adaptation without retraining or fine-tuning. To achieve robust and interpretable adaptation, DyMETER introduces a lightweight evolution controller to estimate instance-level concept uncertainty for adaptive updates. Further, DyMETER employs a dynamic threshold optimization module to adaptively recalibrates the decision boundary by maintaining a candidate window of uncertain samples, which ensures continuous alignment with evolving concepts. Extensive experiments demonstrate that DyMETER significantly outperforms existing OAD approaches across a wide spectrum of application scenarios.
Authors: Midan Shim, Seokju Hwang, Kaehyun Um, Kyong-Ho Lee
Abstract: Large language models still struggle with faithfulness and hallucinations despite their remarkable reasoning abilities. In Knowledge Graph Question Answering (KGQA), semantic parsing-based approaches address the limitations by understanding constraints in a user's question and converting them into a logical form to execute on a knowledge graph. However, existing KGQA benchmarks and methods are biased toward positive and calculation constraints. Negative constraints are neglected, although they frequently appear in real-world questions. In this paper, we introduce a new task, NEgative-conSTrained (NEST) KGQA, where each question contains at least one negative constraint, and a corresponding dataset, NestKGQA. We also design PyLF, a Python-formatted logical form, since existing logical forms are hardly suitable to express negation clearly while maintaining readability. Furthermore, NEST questions naturally contain multiple constraints. To mitigate their semantic complexity, we present a novel framework named CUCKOO, specialized to multiple-constrained questions and ensuring semantic executability. CUCKOO first generates a constraint-aware logical form draft and performs schema-guided semantic matching. It then selectively applies self-directed refinement only when executing improper logical forms yields an empty result, reducing cost while improving robustness. Experimental results demonstrate that CUCKOO consistently outperforms baselines on both conventional KGQA and NEST-KGQA benchmarks under few-shot settings.
Authors: Ali Bagheri Nejad, Mahdi Aliyari-Shoorehdeli, Abolfazl Hasanzadeh
Abstract: Preventing machine failure is inherently superior to reactive remediation, particularly for critical assets like gas turbines, where early fault detection (FD) is a cornerstone of industrial sustainability. However, modern deep learning-based FD models often face a significant trade-off between architectural complexity and real-time operational constraints, often hindered by a lack of temporal context within restricted vibration signal windows. To address these challenges, this study proposes a Temporal Cross-Modal Knowledge-Distillation Transfer-Learning (TCMKDTL) framework. The framework employs a "privileged" teacher model trained on expansive temporal windows incorporating both past and future signal context to distill latent feature-based knowledge into a compact student model. To mitigate issues of data scarcity and domain shift, the framework leverages robust pre-training on benchmark datasets (such as CWRU) followed by adaptation to target industrial data. Extensive evaluation using experimental and industrial gas turbine (MGT-40) datasets demonstrates that TCMKDTL achieves superior feature separability and diagnostic accuracy compared to conventional pre-trained architectures. Ultimately, this approach enables high-performance, unsupervised anomaly detection suitable for deployment on resource-constrained industrial hardware.
Authors: Haileab Yagersew
Abstract: Retail theft costs the global economy over \$100 billion annually, yet existing AI-based detection systems require expensive custom model training on proprietary datasets and charge \$200-500/month per store. We present Paza, a zero-shot retail theft detection framework that achieves practical concealment detection without training any model. Our approach orchestrates multiple existing models in a layered pipeline - cheap object detection and pose estimation running continuously, with an expensive vision-language model (VLM) invoked only when behavioral pre-filters trigger. A multi-signal suspicion pre-filter (requiring dwell time plus at least one behavioral signal) reduces VLM invocations by 240x compared to per-frame analysis, bounding calls to <=10/minute and enabling a single GPU to serve 10-20 stores. The architecture is model-agnostic: the VLM component accepts any OpenAI-compatible endpoint, enabling operators to swap between models such as Gemma 4, Qwen3.5-Omni, GPT-4o, or future releases without code changes - ensuring the system improves as the VLM landscape evolves. We evaluate the VLM component on the DCSASS synthesized shoplifting dataset (169 clips, controlled environment), achieving 89.5% precision and 92.8% specificity at 59.3% recall zero-shot - where the recall gap is attributable to sparse frame sampling in offline evaluation rather than VLM reasoning failures, as precision and specificity are the operationally critical metrics determining false alarm rates. We present a detailed cost model showing viability at \$50-100/month per store (3-10x cheaper than commercial alternatives), and introduce a privacy-preserving design that obfuscates faces in the detection pipeline. The source code is available at https://github.com/xHaileab/Paza-AI.
Authors: Emil Benedykciuk, Marcin Denkowski, Grzegorz M. W\'ojcik
Abstract: Purpose: Adaptive skip modules can improve medical image segmentation, but searching for them is computationally costly. Implantable Adaptive Cells (IACs) are compact NAS modules inserted into U-Net skip connections, reducing the search space compared with full-network NAS. However, the original IAC framework still requires a 200-epoch differentiable search for each backbone and dataset. Methods: We analyzed the temporal behavior of operations and edges within IAC cells during differentiable search on public medical image segmentation benchmarks. We found that operations selected in the final discrete cell typically emerge among the strongest candidates early in training, and their architecture parameters stabilize well before the final epoch. Based on this, we propose a Jensen--Shannon-divergence-based stability criterion that tracks per-edge operation-importance distributions and progressively prunes low-importance operations during search. The accelerated framework is called IAC-LTH. Results: Across four public benchmarks (ACDC, BraTS, KiTS, AMOS), several 2-D U-Net backbones, and a 2-D nnU-Net pipeline, IAC-LTH discovers IAC cells whose patient-level segmentation performance matches and sometimes slightly exceeds that of cells found by the original full-length search, while reducing wall-clock NAS cost by 3.7x to 16x across datasets and backbones. These results are consistent across architectures, benchmarks, and both non-augmented and augmented training settings, while preserving the gains of IAC-equipped U-Nets over strong attention-based and dense-skip baselines. Conclusion: Competitive IAC architectures can be identified from early-stabilizing operations without running the full search, making adaptive skip-module design more practical for medical image segmentation under realistic computational constraints.
Authors: Liesbeth Allein, Nataly Pineda-Casta\~neda, Andrea Rocci, Marie-Francine Moens
Abstract: Understanding climate change requires reasoning over complex causal networks. Yet, existing causal discovery datasets predominantly capture explicit, direct causal relations. We introduce ClimateCause, a manually expert-annotated dataset of higher-order causal structures from science-for-policy climate reports, including implicit and nested causality. Cause-effect expressions are normalized and disentangled into individual causal relations to facilitate graph construction, with unique annotations for cause-effect correlation, relation type, and spatiotemporal context. We further demonstrate ClimateCause's value for quantifying readability based on the semantic complexity of causal graphs underlying a statement. Finally, large language model benchmarking on correlation inference and causal chain reasoning highlights the latter as a key challenge.
Authors: Yifan Le
Abstract: Constrained decoding has been widely adopted for structured generation with large language models (LLMs), ensuring that outputs satisfy predefined formats such as JSON and XML. However, existing approaches largely treat schemas as purely structural constraints and overlook the possibility that their linguistic formulation may affect model behavior. In this work, we study how instruction placement influences model performance in structured generation and show that merely changing the wording of schema keys, without modifying the prompt or model parameters, can significantly alter model performance under constrained decoding. Based on this observation, we propose to reinterpret structured generation as a multi-channel instruction problem, where instructions can be conveyed explicitly through prompts and implicitly through schema keys during decoding. To the best of our knowledge, this is the first work to systematically study how schema key formulation acts as an implicit instruction channel and affects model performance under constrained decoding. Experiments on multiple mathematical reasoning benchmarks show that different model families exhibit distinct sensitivities to these instruction channels: Qwen models consistently benefit from schema-level instructions, while LLaMA models rely more heavily on prompt-level guidance. We further observe non-additive interaction effects between instruction channels, showing that combining multiple channels does not always lead to further improvement. These findings suggest that schema design not only determines output structure, but also carries instruction signals, offering a new perspective on structured generation in LLMs.
Authors: Meng-Xun Li, Wen-Hui Deng, Zhi-Xing Wu, Chun-Xiao Jin, Jia-Min Wu, Yue Han, James Kit Hon Tsoi, Gui-Song Xia, Cui Huang
Abstract: Vision-Language Models (VLMs) have demonstrated significant potential in medical image analysis, yet their application in intraoral photography remains largely underexplored due to the lack of fine-grained, annotated datasets and comprehensive benchmarks. To address this, we present MetaDent, a comprehensive resource that includes (1) a novel and large-scale dentistry image dataset collected from clinical, public, and web sources; (2) a semi-structured annotation framework designed to capture the hierarchical and clinically nuanced nature of dental photography; and (3) comprehensive benchmark suites for evaluating state-of-the-art VLMs on clinical image understanding. Our labeling approach combines a high-level image summary with point-by-point, free-text descriptions of abnormalities. This method enables rich, scalable, and task-agnostic representations. We curated 60,669 dental images from diverse sources and annotated a representative subset of 2,588 images using this meta-labeling scheme. Leveraging Large Language Models (LLMs), we derive standardized benchmarks: approximately 15K Visual Question Answering (VQA) pairs and an 18-class multi-label classification dataset, which we validated with human review and error analysis to justify that the LLM-driven transition reliably preserves fidelity and semantic accuracy. We then evaluate state-of-the-art VLMs across VQA, classification, and image captioning tasks. Quantitative results reveal that even the most advanced models struggle with a fine-grained understanding of intraoral scenes, achieving moderate accuracy and producing inconsistent or incomplete descriptions in image captioning. We publicly release our dataset, annotations, and tools to foster reproducible research and accelerate the development of vision-language systems for dental applications.
Authors: Michal T\"opfer, Franti\v{s}ek Pl\'a\v{s}il, Tom\'a\v{s} Bure\v{s}, Petr Hn\v{e}tynka
Abstract: Vibe coding inherently assumes iterative refinement of LLM-generated code through feedback loops. While effective for conventional software tasks, its reliability in runtime-adaptive systems is unclear -- especially when generated code is not manually inspected. This paper studies feedback-based automated verification of LLM-generated adaptation managers in Collective Adaptive Systems (CAS). We focus on the key challenges of verification in the loop: how to detect failures of generated code at runtime and how to report them precisely enough for an LLM to fix them. We combine the adaptation loop with a vibe-coding feedback loop where correctness is checked against (i) generic architectural constraints and (ii) functional constraints formalized in Functional Constraints Logic (FCL), a novel first-order temporal logic over potentially finite traces. Conducting the Dragon Hunt CAS case study, we show that fine-grained constraint violations provide actionable feedback that typically yields a valid adaptation manager within a few iterations, while simple coarse metric-based feedback often stalls. Our findings suggest that feedback precision is the dominant factor for reliable vibe coding in systems designed by domain experts with no programming skills, thereby obviating the need for human code inspection.
Authors: Yanyan Zou, Junbo Qi, Lunsong Huang, Yu Li, Kewei Xu, Jiabao Gao, Binglei Zhao, Xuanhua Yang, Sulong Xu, Shengjie Li
Abstract: Generative Retrieval (GR) offers a promising paradigm for recommendation through next-token prediction (NTP). However, scaling it to large-scale industrial systems introduces three challenges: (i) within a single request, the identical model inputs may produce inconsistent outputs due to the pagination request mechanism; (ii) the prohibitive cost of encoding long user behavior sequences with multi-token item representations based on semantic IDs, and (iii) aligning the generative policy with nuanced user preference signals. We present GenRec, a preference-oriented generative framework deployed on the JD App that addresses above challenges within a single decoder-only architecture. For training objective, we propose Page-wise NTP task, which supervises over an entire interaction page rather than each interacted item individually, providing denser gradient signal and resolving the one-to-many ambiguity of point-wise training. On the prefilling side, an asymmetric linear Token Merger compresses multi-token Semantic IDs in the prompt while preserving full-resolution decoding, reducing input length by ~2X with negligible accuracy loss. To further align outputs with user satisfaction, we introduce GRPO-SR, a reinforcement learning method that pairs Group Relative Policy Optimization with NLL regularization for training stability, and employs Hybrid Rewards combining a dense reward model with a relevance gate to mitigate reward hacking. In month-long online A/B tests serving production traffic, GenRec achieves 9.5% improvement in click count and 8.7% in transaction count over the existing pipeline.
Authors: Murat Furkan Mansur, Tufan Kumbasar
Abstract: Nonlinear system identification must balance physical interpretability with model flexibility. Classical methods yield structured, control-relevant models but rely on rigid parametric forms that often miss complex nonlinearities, whereas Neural ODEs are expressive yet largely black-box. Physics-Informed Neural Networks (PINNs) sit between these extremes, but inverse PINNs typically assume a known governing equation with fixed coefficients, leading to identifiability failures when the true dynamics are unknown or state-dependent. We propose \textbf{SOLIS}, which models unknown dynamics via a \emph{state-conditioned second-order surrogate model} and recasts identification as learning a Quasi-Linear Parameter-Varying (Quasi-LPV) representation, recovering interpretable natural frequency, damping, and gain without presupposing a global equation. SOLIS decouples trajectory reconstruction from parameter estimation and stabilizes training with a cyclic curriculum and \textbf{Local Physics Hints} windowed ridge-regression anchors that mitigate optimization collapse. Experiments on benchmarks show accurate parameter-manifold recovery and coherent physical rollouts from sparse data, including regimes where standard inverse methods fail.
Authors: Zihong Zhang, Zuchao Li, Lefei Zhang, Ping Wang, Hai Zhao
Abstract: Autoregressive decoding in Large Language Models (LLMs) generates one token per step, causing high inference latency. Speculative decoding (SD) mitigates this through a guess-and-verify strategy, but existing training-free variants face trade-offs: retrieval-based drafts break when no exact match exists, while logits-based drafts lack structural guidance. We propose $\textbf{RACER}$ ($\textbf{R}$etrieval-$\textbf{A}$ugmented $\textbf{C}$ont$\textbf{e}$xtual $\textbf{R}$apid Speculative Decoding), a lightweight and training-free method that integrates retrieved exact patterns with logit-driven future cues. This unification supplies both reliable anchors and flexible extrapolation, yielding richer speculative drafts. Experiments on Spec-Bench, HumanEval, and MGSM-ZH demonstrate that RACER consistently accelerates inference, achieving more than $2\times$ speedup over autoregressive decoding, and outperforms prior training-free methods, offering a scalable, plug-and-play solution for efficient LLM decoding. Our source code is available at $\href{https://github.com/hkr04/RACER}{https://github.com/hkr04/RACER}$.
URLs: https://github.com/hkr04/RACER, https://github.com/hkr04/RACER
Authors: Danae S\'anchez Villegas, Samuel Lewis-Lim, Nikolaos Aletras, Desmond Elliott
Abstract: Recent advances in vision language models (VLMs) offer reasoning capabilities, yet how these unfold and integrate visual and textual information remains unclear. We analyze reasoning dynamics in 18 VLMs covering instruction-tuned and reasoning-trained models from two different model families. We track confidence over Chain-of-Thought (CoT), measure the corrective effect of reasoning, and evaluate the contribution of intermediate reasoning steps. We find that models are prone to answer inertia, in which early commitments to a prediction are reinforced, rather than revised during reasoning steps. While reasoning-trained models show stronger corrective behavior, their gains depend on modality conditions, from text-dominant to vision-only settings. Using controlled interventions with misleading textual cues, we show that models are consistently influenced by these cues even when visual evidence is sufficient, and assess whether this influence is recoverable from CoT. Although this influence can appear in the CoT, its detectability varies across models and depends on what is being monitored. Reasoning-trained models are more likely to explicitly refer to the cues, but their longer and fluent CoTs can still appear visually grounded while actually following textual cues, obscuring modality reliance. In contrast, instruction-tuned models refer to the cues less explicitly, but their shorter traces reveal inconsistencies with the visual input. Taken together, these findings indicate that CoT provides only a partial view of how different modalities drive VLM decisions, with important implications for the transparency and safety of multimodal systems.
Authors: Amy Rouillard, Sitwala Mundiab, Linda Camarab, Michael Cameron Gramaniec, Ziyaad Dangorc, Ismail Kallad, Shabir A. Madhic, Kajal Morarc, Marlvin T. Ncubec, Haroon Saloojeee, Bruce A. Bassett
Abstract: Evaluating medical AI systems using expert clinician panels is costly and slow, motivating the use of large language models (LLMs) as alternative adjudicators. Here, we evaluate an LLM jury composed of three frontier AI models scoring 3333 diagnoses on 300 real-world middle-income country (MIC) hospital cases. Model performance was benchmarked against expert clinician panel and independent human re-scoring panel evaluations. Both LLM and clinician-generated diagnoses are scored across four dimensions: diagnosis, differential diagnosis, clinical reasoning and negative treatment risk. For each of these, we assess scoring difference, inter-rater agreement, scoring stability, severe safety errors and the effect of post-hoc calibration. We find that: (i) the uncalibrated LLM jury scores are systematically lower than clinician panels scores; (ii) the LLM Jury preserves ordinal agreement and exhibits better concordance with the primary expert panels than the human expert re-score panels do; (iii) the probability of severe errors is lower in \lj models compared to the human expert re-score panels; (iv) the LLM Jury shows excellent agreement with primary expert panels' rankings. We find that the LLM jury combined with AI model diagnoses can be used to identify ward diagnoses at high risk of error, enabling targeted expert review and improved panel efficiency; (v) LLM jury models show no self-preference bias. They did not score diagnoses generated by their own underlying model or models from the same vendor more (or less) favourably than those generated by other models. Finally, we demonstrate that LLM jury calibration using isotonic regression improves alignment with human expert panel evaluations. Together, these results provide compelling evidence that a calibrated, multi-model LLM jury can serve as a trustworthy and reliable proxy for expert clinician evaluation in medical AI benchmarking.
Authors: Ziwu Sun, Zhen Gao, Jiyong Zhang, Jiaheng Li
Abstract: We propose a new perspective on policy optimization: rather than reweighting all samples by their importance ratios, an optimizer should select which samples are trustworthy enough to drive a policy update. Building on this view, we introduce Rejection-Gated Policy Optimization (RGPO), which replaces the importance sampling ratio r_theta = pi_theta / pi_old with a smooth, differentiable acceptance gate alpha_theta(s, a) = g(r_theta(s, a)) in the range [0, 1]. Unlike prior work that applies rejection sampling as a data-level heuristic before training, RGPO elevates rejection to an optimization principle: the gate participates directly in gradient computation and is implicitly updated alongside the policy. RGPO provides a unified framework: the policy gradients of TRPO, PPO, and REINFORCE all correspond to specific choices of the effective gradient weight w(r) = g'(r) * r. We prove that RGPO guarantees finite, bounded gradient variance even when importance sampling ratios are heavy-tailed (where IS variance diverges). We further show that RGPO incurs only a bounded, controllable bias and provides an approximate monotonic policy improvement guarantee analogous to TRPO. RGPO matches PPO in computational cost, requires no second-order optimization, and extends naturally to RLHF-style preference alignment. In online preference fine-tuning of Qwen2.5-1.5B-Instruct on Anthropic HH-RLHF (n = 3 seeds), RGPO uses a dual-ratio gate that anchors learning to both the previous policy and the reference model, achieving a Pareto-dominant outcome: the highest reward among online RL methods (+14.8% vs. PPO-RLHF) and the lowest KL divergence to the reference model (-16.0% vs. PPO-RLHF, -53.1% vs. GRPO).
Authors: Dongsheng Wang, Jinsen Zhang, Dawei Su, Hui Huang
Abstract: Recently, sparse autoencoders (SAEs) have emerged as a promising technique for interpreting activations in foundation models by disentangling features into a sparse set of concepts. However, identifying the optimal level of sparsity for each neuron remains challenging in practice: excessive sparsity can lead to poor reconstruction, whereas insufficient sparsity may harm interpretability. While existing activation functions such as ReLU and TopK provide certain sparsity guarantees, they typically require additional sparsity regularization or cherry-picked hyperparameters. We show in this paper that dynamically sparse attention mechanisms using sparsemax can bridge this trade-off, due to their ability to determine the activation numbers in a data-dependent manner. Specifically, we first explore a new class of SAEs based on the cross-attention architecture with the latent features as queries and the learnable dictionary as the key and value matrices. To encourage sparse pattern learning, we employ a sparsemax-based attention strategy that automatically infers a sparse set of elements according to the complexity of each neuron, resulting in a more flexible and general activation function. Through comprehensive evaluation and visualization, we show that our approach successfully achieves lower reconstruction loss while producing high-quality concepts, particularly in top-n classification tasks.
Authors: Shen Fan, Miko{\l}aj Kida, Przemyslaw Musialski
Abstract: Many CAD learning pipelines discretize Boundary Representations (B-Reps) into triangle meshes, discarding analytic surface structure and topological adjacency and thereby weakening consistent instance-level analysis. We present STEP-Parts, a deterministic CAD-to-supervision toolchain that extracts geometric instance partitions directly from raw STEP B-Reps and transfers them to tessellated carriers through retained source-face correspondence, yielding instance labels and metadata for downstream learning and evaluation. The construction merges adjacent B-Rep faces only when they share the same analytic primitive type and satisfy a near-tangent continuity criterion. On ABC, same-primitive dihedral angles are strongly bimodal, yielding a threshold-insensitive low-angle regime for part extraction. Because the partition is defined on intrinsic B-Rep topology rather than on a particular triangulation, the resulting boundaries remain stable under changes in tessellation. Applied to the DeepCAD subset of ABC, the pipeline processes approximately 180{,}000 models in under six hours on a consumer CPU. We release code and precomputed labels, and show that STEP-Parts serves both as a tessellation-robust geometric reference and as a useful supervision source in two downstream probes: an implicit reconstruction--segmentation network and a dataset-level point-based backbone.
Authors: Gabriele Mattioli, Evelyn Turri, Sara Sarto, Lorenzo Baraldi, Marcella Cornia, Lorenzo Baraldi, Rita Cucchiara
Abstract: Tool learning with foundation models aims to endow AI systems with the ability to invoke external resources -- such as APIs, computational utilities, and specialized models -- to solve complex tasks beyond the reach of standalone language generation. While recent advances in Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) have expanded their reasoning and perception capabilities, existing tool-use methods are predominantly limited to text-only inputs and closed-world settings. Consequently, they struggle to interpret multimodal user instructions and cannot generalize to tools unseen during training. In this work, we introduce RaTA-Tool, a novel framework for open-world multimodal tool selection. Rather than learning direct mappings from user queries to fixed tool identifiers, our approach enables an MLLM to convert a multimodal query into a structured task description and subsequently retrieve the most appropriate tool by matching this representation against semantically rich, machine-readable tool descriptions. This retrieval-based formulation naturally supports extensibility to new tools without retraining. To further improve alignment between task descriptions and tool selection, we incorporate a preference-based optimization stage using Direct Preference Optimization (DPO). To support research in this setting, we also introduce the first dataset for open-world multimodal tool use, featuring standardized tool descriptions derived from Hugging Face model cards. Extensive experiments demonstrate that our approach significantly improves tool-selection performance, particularly in open-world, multimodal scenarios.
Authors: Maksim Pershin, Ivan Golovanov, Pavel Baltabaev, Natalia Trankova
Abstract: Contextual bandit algorithms suffer from high regret during cold-start, when the learner has insufficient data to distinguish good arms from bad. We propose augmenting Disjoint LinUCB with LLM pseudo-observations: after each round, a large language model predicts counterfactual rewards for the unplayed arms, and these predictions are injected into the learner as weighted pseudo-observations. The injection weight is controlled by a calibration-gated decay schedule that tracks the LLM's prediction accuracy on played arms via an exponential moving average; high calibration error suppresses the LLM's influence, while accurate predictions receive higher weight during the critical early rounds. We evaluate on two contextual bandit environments - UCI Mushroom (2-arm, asymmetric rewards) and MIND-small (5-arm news recommendation) - and find that when equipped with a task-specific prompt, LLM pseudo-observations reduce cumulative regret by 19% on MIND relative to pure LinUCB. However, generic counterfactual prompt framing increases regret on both environments, demonstrating that prompt design is the dominant factor, more important than the choice of decay schedule or calibration gating parameters. We analyze the failure modes of calibration gating on domains with small prediction errors and provide a theoretical motivation for the bias-variance trade-off governing pseudo-observation weight.
Authors: Jun Wang, Shuo Tan, Zelong Sun, Tiancheng Gu, Yongle Zhao, Ziyong Feng, Kaicheng Yang, Cewu Lu
Abstract: Retrieval-Augmented Generation (RAG) extends Large Vision-Language Models (LVLMs) with external visual knowledge. However, existing visual RAG systems typically rely on generic retrieval signals that overlook the fine-grained visual semantics essential for complex reasoning. To address this limitation, we propose UniDoc-RL, a unified reinforcement learning framework in which an LVLM agent jointly performs retrieval, reranking, active visual perception, and reasoning. UniDoc-RL formulates visual information acquisition as a sequential decision-making problem with a hierarchical action space. Specifically, it progressively refines visual evidence from coarse-grained document retrieval to fine-grained image selection and active region cropping, allowing the model to suppress irrelevant content and attend to information-dense regions. For effective end-to-end training, we introduce a dense multi-reward scheme that provides task-aware supervision for each action. Based on Group Relative Policy Optimization (GRPO), UniDoc-RL aligns agent behavior with multiple objectives without relying on a separate value network. To support this training paradigm, we curate a comprehensive dataset of high-quality reasoning trajectories with fine-grained action annotations. Experiments on three benchmarks demonstrate that UniDoc-RL consistently surpasses state-of-the-art baselines, yielding up to 17.7% gains over prior RL-based methods.
Authors: Yomna Elsayed, Cecily Jones
Abstract: As companies enter the race for agentic AI adoption, fears surface around agentic autonomy and its subsequent risks. These fears compound as companies scale their agentic AI adoption with low-code applications, without a comparable scaling in their governance processes and expertise resulting in a phenomenon known as "Agent Sprawl". While shadow AI tools can help with agentic discovery and identification, few observability tools offer insights into the agents' configuration and settings or the decision-making process during agent-to-agent communication and orchestration. This paper explores AI governance professionals' concerns in enterprise settings, while offering design-time and runtime explainability techniques as suggested by AI governance experts for addressing those fears. Finally, we provide a preliminary prototype of an Agentic AI Card that can help companies feel at ease deploying agents at scale.
Authors: \'Eric Jacopin
Abstract: When do transformers commit to a decision, and what prevents them from correcting it? We introduce \textbf{prolepsis}: a transformer commits early, task-specific attention heads sustain the commitment, and no layer corrects it. Replicating \citeauthor{lindsey2025biology}'s (\citeyear{lindsey2025biology}) planning-site finding on open models (Gemma~2 2B, Llama~3.2 1B), we ask five questions. (Q1)~Planning is invisible to six residual-stream methods; CLTs are necessary. (Q2)~The planning-site spike replicates with identical geometry. (Q3)~Specific attention heads route the decision to the output, filling a gap flagged as invisible to attribution graphs. (Q4)~Search requires ${\leq}16$ layers; commitment requires more. (Q5)~Factual recall shows the same motif at a different network depth, with zero overlap between recurring planning heads and the factual top-10. Prolepsis is architectural: the template is shared, the routing substrates differ. All experiments run on a single consumer GPU (16\,GB VRAM).
Authors: Haochun Tang, Yuliang Yan, Jiahua Lu, Huaxiao Liu, Enyan Dai
Abstract: Cost-aware routing dynamically dispatches user queries to models of varying capability to balance performance and inference cost. However, the routing strategy introduces a new security concern that adversaries may manipulate the router to consistently select expensive high-capability models. Existing routing attacks depend on either white-box access or heuristic prompts, rendering them ineffective in real-world black-box scenarios. In this work, we propose R$^2$A, which aims to mislead black-box LLM routers to expensive models via adversarial suffix optimization. Specifically, R$^2$A deploys a hybrid ensemble surrogate router to mimic the black-box router. A suffix optimization algorithm is further adapted for the ensemble-based surrogate. Extensive experiments on multiple open-source and commercial routing systems demonstrate that {R$^2$A} significantly increases the routing rate to expensive models on queries of different distributions. Code and examples: https://github.com/thcxiker/R2A-Attack.
Authors: Khalid Adnan Alsayed
Abstract: The evaluation of fairness in machine learning systems has become a central concern in high-stakes applications, including biometric recognition, healthcare decision-making, and automated risk assessment. Existing approaches typically rely on a small number of fairness metrics to assess model behaviour across group partitions, implicitly assuming that these metrics provide consistent and reliable conclusions. However, different fairness metrics capture distinct statistical properties of model performance and may therefore produce conflicting assessments when applied to the same system. In this work, we investigate the consistency of fairness evaluation by conducting a systematic multi-metric analysis of demographic bias in machine learning models. Using face recognition as a controlled experimental setting, we evaluate model performance across multiple group partitions under a range of commonly used fairness metrics, including error-rate disparities and performance-based measures. Our results demonstrate that fairness assessments can vary significantly depending on the choice of metrics, leading to contradictory conclusions regarding model bias. To quantify this phenomenon, we introduce the Fairness Disagreement Index (FDI), a measure designed to capture the degree of inconsistency across fairness metrics. We further show that disagreement remains high across thresholds and model configurations. These findings highlight a critical limitation in current fairness evaluation practices and suggest that single-metric reporting is insufficient for reliable bias assessment.
Authors: Chase McDonald, Cleotilde Gonzalez
Abstract: The increasing integration of artificial intelligence (AI) in everyday life brings with it new challenges and questions for regarding how humans interact with autonomous agents. Multi-agent experiments, where humans and AI act together, can offer important opportunities to study social decision making, but there is a lack of accessible tooling available to researchers to run such experiments. We introduce two tools designed to reduce these barriers. The first, CoGrid, is a multi-agent grid-based simulation library with dual NumPy and JAX backends. The second, Multi-User Gymnasium (MUG), translates such simulation environments directly into interactive web-based experiments. MUG supports interactions with arbitrary numbers of humans and AI, utilizing either server-authoritative or peer-to-peer networking with rollback netcode to account for latency. Together, these tools can enable researchers to deploy studies of human-AI interaction, facilitating inquiry into core questions of psychology, cognition, and decision making and their relationship to human-AI interaction. Both tools are open source and available to the broader research community. Documentation and source code is available at {cogrid, multi-user-gymnasium}.readthedocs.io. This paper details the functionality of these tools and presents several case studies to illustrate their utility in human-AI multi-agent experimentation.
Authors: Francesco Diana, Chuan Xu, Andr\'e Nusser, Giovanni Neglia
Abstract: Gradient inversion attacks threaten client privacy in federated learning by reconstructing training samples from clients' shared gradients. Gradients aggregate contributions from multiple records and existing attacks may fail to disentangle them, yielding incorrect reconstructions with no intrinsic way to certify success. In vision and language, attackers may fall back on human inspection to judge reconstruction plausibility, but this is far less feasible for numerical tabular records, fueling the impression that tabular data is less vulnerable. We challenge this perception by proposing a verifiable gradient inversion attack (VGIA) that provides an explicit certificate of correctness for reconstructed samples. Our method adopts a geometric view of ReLU leakage: the activation boundary of a fully connected layer defines a hyperplane in input space. VGIA introduces an algebraic, subspace-based verification test that detects when a hyperplane-delimited region contains exactly one record. Once isolation is certified, VGIA recovers the corresponding feature vector analytically and reconstructs the target via a lightweight optimization step. Experiments on tabular benchmarks with large batch sizes demonstrate exact record and target recovery in regimes where existing state-of-the-art attacks either fail or cannot assess reconstruction fidelity. Compared to prior geometric approaches, VGIA allocates hyperplane queries more effectively, yielding faster reconstructions with fewer attack rounds.
Authors: Hibatallah Meliani, Khadija Slimani, Samira Khoulji
Abstract: To navigate a space, the brain makes an internal representation of the environment using different cells such as place cells, grid cells, head direction cells, border cells, and speed cells. All these cells, along with sensory inputs, enable an organism to explore the space around it. Inspired by these biological principles, we developed NEATNC, a Neuro-Evolution of Augmenting Topology guided Navigation Cells. The goal of the paper is to improve NEAT algorithm performance in path planning in dynamic environments using spatial cognitive cells. This approach uses navigation cells as inputs and evolves recurrent neural networks, representing the hippocampus part of the brain. The performance of the proposed algorithm is evaluated in different static and dynamic scenarios. This study highlights NEAT's adaptability to complex and different environments, showcasing the utility of biological theories. This suggests that our approach is well-suited for real-time dynamic path planning for robotics and games.
Authors: Cunxi Yu, Haoxing Ren
Abstract: This paper introduces the first \emph{self-evolving} logic synthesis framework, which leverages Large Language Model (LLM) agents to autonomously improve the source code of \textsc{ABC}, the widely adopted logic synthesis system. Our framework operates on the \emph{entire integrated ABC codebase}, and the output repository preserves its single-binary execution model and command interface. In the initial evolution cycle, we bootstrap the system using existing prior open-source synthesis components, covering flow tuning, logic minimization, and technology mapping, but without manually injecting new heuristics. On top of this foundation, a team of LLM-based agents iteratively rewrites and evolves specific sub-components of ABC following our ``programming guidance`` prompts under a unified correctness and QoR-driven evaluation loop. Each evolution cycle proposes code modifications, compiles the integrated binary, validates correctness, and evaluates quality-of-results (QoR) on \emph{multi-suite benchmarks including ISCAS~85/89/99, VTR, EPFL, and IWLS~2005}. Through continuous feedback, the system discovers optimizations beyond human-designed heuristics, effectively \emph{learning new synthesis strategies} that enhance QoR. We detail the architecture of this self-improving system, its integration with \textsc{ABC}, and results demonstrating that the framework can autonomously and progressively improve EDA tool at full million-line scale.
Authors: Haozhi Fan, Jinhao Duan, Kaidi Xu
Abstract: Despite the rapid advancement of Large Language Models (LLMs), uncertainty quantification in LLM generation is a persistent challenge. Although recent approaches have achieved strong performance by restricting LLMs to produce short or constrained answer sets, many real-world applications require long-form and free-form text generation. A key difficulty in this setting is that LLMs often produce responses that are semantically coherent yet factually inaccurate, while the underlying semantics are multifaceted and the linguistic structure is complex. To tackle this challenge, this paper introduces Interrogative Uncertainty Quantification (IUQ), a novel framework that leverages inter-sample consistency and intra-sample faithfulness to quantify the uncertainty in long-form LLM outputs. By utilizing an interrogate-then-respond paradigm, our method provides reliable measures of claim-level uncertainty and the model's faithfulness. Experimental results across diverse model families and model sizes demonstrate the superior performance of IUQ over two widely used long-form generation datasets. The code is available at https://github.com/louisfanhz/IUQ.
Authors: Minh-Phuc Truong, Khai Nguyen
Abstract: We propose a novel amortized optimization method for predicting optimal transport (OT) plans across multiple pairs of measures by leveraging Kantorovich potentials derived from sliced OT. We introduce two amortization strategies: regression-based amortization (RA-OT) and objective-based amortization (OA-OT). In RA-OT, we formulate a functional regression model that treats Kantorovich potentials from the original OT problem as responses and those obtained from sliced OT as predictors, and estimate these models via least-squares methods. In OA-OT, we estimate the parameters of the functional model by optimizing the Kantorovich dual objective. In both approaches, the predicted OT plan is subsequently recovered from the estimated potentials. As amortized OT methods, both RA-OT and OA-OT enable efficient solutions to repeated OT problems across different measure pairs by reusing information learned from prior instances to rapidly approximate new solutions. Moreover, by exploiting the structure provided by sliced OT, the proposed models are more parsimonious, independent of specific structures of the measures, such as the number of atoms in the discrete case, while achieving high accuracy. We demonstrate the effectiveness of our approaches on tasks including MNIST digit transport, color transfer, supply-demand transportation on spherical data, and mini-batch OT conditional flow matching.
Authors: Duan Zhou
Abstract: This work simulates the developmental process of cortical neurogenesis, initiating from a single stem cell and governed by gene regulatory rules derived from mouse single-cell transcriptomic data. The developmental process spontaneously generates a heterogeneous population of 5,000 cells, yet yields only 85 mature neurons - merely 1.7% of the total population. These 85 neurons form a densely interconnected core of 200,400 synapses, corresponding to an average degree of 4,715 per neuron. At iteration zero, this minimal circuit performs at chance level on MNIST. However, after a single epoch of standard training, accuracy surges to over 90% - a gain exceeding 80 percentage points - with typical runs falling in the 89-94% range depending on developmental stochasticity. The identical circuit, without any architectural modification or data augmentation, achieves 40.53% on CIFAR-10 after one epoch. These findings demonstrate that developmental rules sculpt a domain-general topological substrate exceptionally amenable to rapid learning, suggesting that biological developmental processes inherently encode powerful structural priors for efficient computation.
Authors: Lukas Helff, Quentin Delfosse, David Steinmann, Ruben H\"arle, Hikaru Shindo, Patrick Schramowski, Wolfgang Stammer, Kristian Kersting, Felix Friedrich
Abstract: As reinforcement Learning with Verifiable Rewards (RLVR) has become the dominant paradigm for scaling reasoning capabilities in LLMs, a new failure mode emerges: LLMs gaming verifiers. We study this phenomenon on inductive reasoning tasks, where models must induce and output logical rules. We find that RLVR-trained models systematically abandon rule induction. Instead of learning generalizable patterns (e.g., ``trains carrying red cars go east''), they enumerate instance-level labels, producing outputs that pass verifiers without capturing the relational patterns required by the task. We show that this behavior is not a failure of understanding but a form of reward hacking: imperfect verifiers that check only extensional correctness admit false positives. To detect such shortcuts, we introduce Isomorphic Perturbation Testing (IPT), which evaluates a single model output under both extensional and isomorphic verification, where the latter enforces invariance under logically isomorphic tasks. While genuine rule induction remains invariant, shortcut strategies fail. We find that shortcut behavior is specific to RLVR-trained reasoning models (e.g., GPT-5, Olmo3) and absent in non-RLVR models (e.g., GPT-4o, GPT-4.5, Ministral). Moreover, shortcut prevalence increases with task complexity and inference-time compute. In controlled training experiments, extensional verification directly induces shortcut strategies, while isomorphic verification eliminates them. These results show that RLVR can incentivize reward hacking not only through overt manipulation but also by exploiting what the verifier fails to enforce.
Authors: Zihao Xu, John Harvill, Ziwei Fan, Yizhou Sun, Hao Ding, Hao Wang
Abstract: Large Language Models (LLMs) incur significant computational and memory costs when processing long prompts, as full self-attention scales quadratically with input length. Token compression aims to address this challenge by reducing the number of tokens representing inputs. However, existing prompt-compression approaches primarily operate in token space and overlook inefficiencies in the latent embedding space. In this paper, we propose K-Token Merging, a latent-space compression framework that merges each contiguous block of K token embeddings into a single embedding via a lightweight encoder. The compressed sequence is processed by a LoRA-adapted LLM, while generation remains in the original vocabulary. Experiments on structural reasoning (Textualized Tree), sentiment classification (Amazon Reviews), and code editing (CommitPackFT) show that K-Token Merging lies on the Pareto frontier of performance vs. compression, achieving up to 75% input length reduction with minimal performance degradation.
Authors: Arman Hatami, Romina Aalishah, Ilya E. Monosov
Abstract: Machine unlearning aims to remove targeted knowledge from a trained model without the cost of retraining from scratch. In class unlearning, however, reducing accuracy on forget classes does not necessarily imply true forgetting: forgotten information can remain encoded in internal representations, and apparent forgetting may arise from classifier-head suppression rather than representational removal. We show that existing class-unlearning methods often exhibit weak or negative selectivity, preserve forget-class structure in deep representations, or rely heavily on final-layer bias shifts. We then introduce DAMP (Depth-Aware Modulation by Projection), a one-shot, closed-form weight-surgery method that removes forget-specific directions from a pretrained network without gradient-based optimization. At each stage, DAMP computes class prototypes in the input space of the next learnable operator, extracts forget directions as residuals relative to retain-class prototypes, and applies a projection-based update to reduce downstream sensitivity to those directions. To preserve utility, DAMP uses a parameter-free depth-aware scaling rule derived from probe separability, applying smaller edits in early layers and larger edits in deeper layers. The method naturally extends to multi-class forgetting through low-rank subspace removal. Across MNIST, CIFAR-10, CIFAR-100, and Tiny ImageNet, and across convolutional and transformer architectures, DAMP more closely resembles the retraining gold standard than some of the prior methods, improving selective forgetting while better preserving retain-class performance and reducing residual forget-class structure in deep layers.
Authors: Yoo-Min Jung, Leekyung Kim
Abstract: Despite recent advances in state space models (SSMs) such as Mamba across various sequence domains, research on their standalone capacity for time series classification (TSC) has remained limited. We propose MambaSL, a framework that minimally redesigns the selective SSM and projection layers of a single-layer Mamba, guided by four TSC-specific hypotheses. To address benchmarking limitations -- restricted configurations, partial University of East Anglia (UEA) dataset coverage, and insufficiently reproducible setups -- we re-evaluate 20 strong baselines across all 30 UEA datasets under a unified protocol. As a result, MambaSL achieves state-of-the-art performance with statistically significant average improvements, while ensuring reproducibility via public checkpoints for all evaluated models. Together with visualizations, these results demonstrate the potential of Mamba-based architectures as a TSC backbone.
Authors: Marcel Wagenl\"ander, Otto White, Britannio Jarrett, Pedro Silvestre, Yanda Tao, Guo Li, Huanzhou Zhu, Ll\'uis Vilanova, Peter Pietzuch
Abstract: Agentic workflows carry out complex tasks by orchestrating multiple large language models (LLMs) and tools. Serving such workflows at a target throughput with low latency is challenging because they can be defined using arbitrary agentic frameworks and exhibit unpredictable execution times: execution may branch, fan-out, or recur in data-dependent ways. Since LLMs in workflows often outnumber available GPUs, their execution also leads to GPU oversubscription. We describe Scepsy, a new agentic serving system that efficiently schedules arbitrary multi-LLM agentic workflows onto a GPU cluster. Scepsy exploits the insight that, while agentic workflows have unpredictable end-to-end latencies, the shares of each LLM's total execution times are comparatively stable across executions. Scepsy decides on GPU allocations based on these aggregate shares: first, it profiles the LLMs under different parallelism degrees. It then uses these statistics to construct an Aggregate LLM Pipeline, which is a lightweight latency/throughput predictor for allocations. To find a GPU allocation that minimizes latency while achieving a target throughput, Scepsy uses the Aggregate LLM Pipeline to explore a search space over fractional GPU shares, tensor parallelism degrees, and replica counts. It uses a hierarchical heuristic to place the best allocation onto the GPU cluster, minimizing fragmentation, while respecting network topology constraints. Our evaluation on realistic agentic workflows shows that Scepsy achieves up to 2.4x higher throughput and 27x lower latency compared to systems that optimize LLMs independently or rely on user-specified allocations.
Authors: Huawei Ji, Yuanhao Sun, Yuan Jin, Cheng Deng, Jiaxin Ding, Luoyi Fu, Xinbing Wang
Abstract: Visual token pruning methods effectively mitigate the quadratic computational growth caused by processing high-resolution images and video frames in vision-language models (VLMs). However, existing approaches rely on predefined pruning configurations without determining whether they achieve computation-performance optimality. In this work, we introduce , a novel framework that formulates visual token pruning as a Pareto configuration optimization problem to automatically identify optimal configurations. Our approach employs continuous relaxation and straight-through estimators to enable gradient-based search, solved via the Augmented Lagrangian method. Extensive experiments across 8 visual benchmarks demonstrate that effectively approximates the empirical Pareto frontier obtained through grid search and generalizes well across various pruning methods and VLM architectures. Furthermore, through learnable kernel functions, we investigate layer-wise pruning patterns and reveal that multi-step progressive pruning captures VLMs' hierarchical compression structure, achieving superior accuracy-efficiency trade-offs compared to single-layer approaches.
Authors: Carlos S. Sep\'ulveda, Gonzalo A. Ruz
Abstract: Coverage path planning on irregular hexagonal grids is relevant to maritime surveillance, search and rescue and environmental monitoring, yet classical methods are often compared on small ad hoc examples or on rectangular grids. This paper presents a reproducible benchmark of deterministic single-vehicle coverage path planning heuristics on irregular hexagonal graphs derived from synthetic but maritime-motivated areas of interest. The benchmark contains 10,000 Hamiltonian-feasible instances spanning compact, elongated, and irregular morphologies, 17 heuristics from seven families, and a common evaluation protocol covering Hamiltonian success, complete-coverage success, revisits, path length, heading changes, and CPU latency. Across the released dataset, heuristics with explicit shortest-path reconnection solve the relaxed coverage task reliably but almost never produce zero-revisit tours. Exact Depth-First Search confirms that every released instance is Hamiltonian-feasible. The strongest classical Hamiltonian baseline is a Warnsdorff variant that uses an index-based tie-break together with a terminal-inclusive residual-degree policy, reaching 79.0% Hamiltonian success. The dominant design choice is not tie-breaking alone, but how the residual degree is defined when the endpoint is reserved until the final move. This shows that underreported implementation details can materially affect performance on sparse geometric graphs with bottlenecks. The benchmark is intended as a controlled testbed for heuristic analysis rather than as a claim of operational optimality at fleet scale.
Authors: Oz Levy, Ilya Dikman, Natan Levy, Michael Winokur
Abstract: Artificial Intelligence is increasingly introduced into systems engineering activities, particularly within requirements engineering, where quality assessment and validation remain heavily dependent on expert judgment. While recent AI tools demonstrate promising capabilities in analyzing and generating requirements, their role within formal systems engineering processes-and their alignment with established INCOSE criteria-remains insufficiently understood. This paper investigates the extent to which AI-based tools can support systems engineers in evaluating requirement quality, without replacing professional expertise. The research adopts a structured systems engineering methodology to compare AI-assisted requirement evaluation with human expert assessment. A controlled study was conducted in which system requirements were evaluated against established INCOSE ``good requirement'' criteria by both experienced systems engineers and an AI-based assessment tool. The evaluation focused on consistency, completeness, clarity, and testability, examining not only accuracy but also the decision logic underlying each assessment. Results indicate that AI tools can provide consistent and rapid preliminary assessments, particularly for syntactic and structural quality attributes. However, expert judgment remains essential for contextual interpretation, ambiguity resolution, and trade-off reasoning. Rather than positioning AI as a replacement for systems engineers, the findings support its role as a decision-support mechanism within the RE lifecycle. From a systems engineering perspective, this study contributes empirical evidence on how AI can be integrated into RE workflows while preserving traceability, accountability, and engineering consistency.
Authors: Federico Pierucci, Matteo Prandi, Marcantonio Bracale Syrnikov, Marcello Galisai, Piercosma Bisconti
Abstract: This paper advances a methodological proposal for safety research in agentic AI. As systems acquire planning, memory, tool use, persistent identity, and sustained interaction, safety can no longer be analysed primarily at the level of the isolated model. Population-level risks arise from structured interaction among agents, through processes of communication, observation, and mutual influence that shape collective behaviour over time. As the object of analysis shifts, a methodological gap emerges. Approaches focused either on single agents or on aggregate outcomes do not identify the interaction-level mechanisms that generate collective risks or the design variables that control them. A framework is required that links local interaction structure to population-level dynamics in a causally explicit way, allowing both explanation and intervention. We introduce two linked concepts. Agentic microphysics defines the level of analysis: local interaction dynamics where one agent's output becomes another's input under specific protocol conditions. Generative safety defines the methodology: growing phenomena and elicit risks from micro-level conditions to identify sufficient mechanisms, detect thresholds, and design effective interventions.
Authors: Asher Labovich
Abstract: Looped transformers promise test-time compute scaling by spending more iterations on harder problems, but it remains unclear which architectural choices let them extrapolate to harder problems at test time rather than memorize training-specific solutions. We introduce a fixed-point based framework for analyzing looped architectures along three axes of stability -- reachability, input-dependence, and geometry -- and use it to characterize when fixed-point iteration yields meaningful predictions. Theoretically, we prove that looped networks without recall have countable fixed points and cannot achieve strong input-dependence at any spectral regime, while recall combined with outer normalization reliably produces a regime in which fixed points are simultaneously reachable, locally smooth in the input, and supported by stable backpropagation. Empirically, we train single-layer looped transformers on chess, sudoku, and prefix-sums and find that downstream performance tracks the framework's predictions across tasks and architectural configurations. We additionally introduce internal recall, a novel recall placement variant, and show that it becomes competitive with -- and on sudoku, substantially better than -- standard recall placement once outer normalization is applied.
Authors: Emanuel Tewolde, Xiao Zhang, David Guzman Piedrahita, Vincent Conitzer, Zhijing Jin
Abstract: It is increasingly important that LLM agents interact effectively and safely with other goal-pursuing agents, yet, recent works report the opposite trend: LLMs with stronger reasoning capabilities behave _less_ cooperatively in mixed-motive games such as the prisoner's dilemma and public goods settings. Indeed, our experiments show that recent models -- with or without reasoning enabled -- consistently defect in single-shot social dilemmas. To tackle this safety concern, we present the first comparative study of game-theoretic mechanisms that are designed to enable cooperative outcomes between rational agents _in equilibrium_. Across four social dilemmas testing distinct components of robust cooperation, we evaluate the following mechanisms: (1) repeating the game for many rounds, (2) reputation systems, (3) third-party mediators to delegate decision making to, and (4) contract agreements for outcome-conditional payments between players. Among our findings, we establish that contracting and mediation are most effective in achieving cooperative outcomes between capable LLM models, and that repetition-induced cooperation deteriorates drastically when co-players vary. Moreover, we demonstrate that these cooperation mechanisms become _more effective_ under evolutionary pressures to maximize individual payoffs.
Authors: Tianhao Fu, Austin Wang, Charles Chen, Roby Aldave-Garza, Yucheng Chen
Abstract: Reliable uncertainty estimation is critical for medical image segmentation, where automated contours feed downstream quantification and clinical decision support. Many strong uncertainty methods require repeated inference, while efficient single-forward-pass alternatives often provide weaker failure ranking or rely on restrictive feature-space assumptions. We present $\textbf{SegWithU}$, a post-hoc framework that augments a frozen pretrained segmentation backbone with a lightweight uncertainty head. SegWithU taps intermediate backbone features and models uncertainty as perturbation energy in a compact probe space using rank-1 posterior probes. It produces two voxel-wise uncertainty maps: a calibration-oriented map for probability tempering and a ranking-oriented map for error detection and selective prediction. Across ACDC, BraTS2024, and LiTS, SegWithU is the strongest and most consistent single-forward-pass baseline, achieving AUROC/AURC of $0.9838/2.4885$, $0.9946/0.2660$, and $0.9925/0.8193$, respectively, while preserving segmentation quality. These results suggest that perturbation-based uncertainty modeling is an effective and practical route to reliability-aware medical segmentation. Source code is available at https://github.com/ProjectNeura/SegWithU.
Authors: Mengdi Wu, Xiaoyu Jiang, Oded Padon, Zhihao Jia
Abstract: This paper presents Prism, the first symbolic superoptimizer for tensor programs. The key idea is sGraph, a symbolic, hierarchical representation that compactly encodes large classes of tensor programs by symbolically representing some execution parameters. Prism organizes optimization as a two-level search: it constructs symbolic graphs that represent families of programs, and then instantiates them into concrete implementations. This formulation enables structured pruning of provably suboptimal regions of the search space using symbolic reasoning over operator semantics, algebraic identities, and hardware constraints. We develop techniques for efficient symbolic graph generation, equivalence verification via e-graph rewriting, and parameter instantiation through auto-tuning. Together, these components allow Prism to bridge the rigor of exhaustive search with the scalability required for modern ML workloads. Evaluation on five commonly used LLM workloads shows that Prism achieves up to $2.2\times$ speedup over best superoptimizers and $4.9\times$ over best compiler-based approaches, while reducing end-to-end optimization time by up to $3.4\times$.
Authors: Madhav Agarwal, Sotirios A. Tsaftaris, Laura Sevilla-Lara, Steven McDonagh
Abstract: Understanding emotions is a fundamental ability for intelligent systems to be able to interact with humans. Vision-language models (VLMs) have made tremendous progress in the last few years for many visual tasks, potentially offering a promising solution for understanding emotions. However, it is surprising that even the most sophisticated contemporary VLMs struggle to recognize human emotions or to outperform even specialized vision-only classifiers. In this paper we ask the question "Why do VLMs struggle to recognize human emotions?", and observe that the inherently continuous and dynamic task of facial expression recognition (DFER) exposes two critical VLM vulnerabilities. First, emotion datasets are naturally long-tailed, and the web-scale data used to pre-train VLMs exacerbates this head-class bias, causing them to systematically collapse rare, under-represented emotions into common categories. We propose alternative sampling strategies that prevent favoring common concepts. Second, temporal information is critical for understanding emotions. However, VLMs are unable to represent temporal information over dense frame sequences, as they are limited by context size and the number of tokens that can fit in memory, which poses a clear challenge for emotion recognition. We demonstrate that the sparse temporal sampling strategy used in VLMs is inherently misaligned with the fleeting nature of micro-expressions (0.25-0.5 seconds), which are often the most critical affective signal. As a diagnostic probe, we propose a multi-stage context enrichment strategy that utilizes the information from "in-between" frames by first converting them into natural language summaries. This enriched textual context is provided as input to the VLM alongside sparse keyframes, preventing attentional dilution from excessive visual data while preserving the emotional trajectory.
Authors: Fabrizio Genilotti, Arianna Stropeni, Gionata Grotto, Francesco Borsatti, Manuel Barusco, Davide Dalle Pezze, Gian Antonio Susto
Abstract: The reliability of a machine vision system for autonomous driving depends heavily on its training data distribution. When a vehicle encounters significantly different conditions, such as atypical obstacles, its perceptual capabilities can degrade substantially. Unlike many domains where errors carry limited consequences, failures in autonomous driving translate directly into physical risk for passengers, pedestrians, and other road users. To address this challenge, we explore Visual Anomaly Detection (VAD) as a solution. VAD enables the identification of anomalous objects not present during training, allowing the system to alert the driver when an unfamiliar situation is detected. Crucially, VAD models produce pixel-level anomaly maps that can guide driver attention to specific regions of concern without requiring any prior assumptions about the nature or form of the hazard. We benchmark eight state-of-the-art VAD methods on AnoVox, the largest synthetic dataset for anomaly detection in autonomous driving. In particular, we evaluate performance across four backbone architectures spanning from large networks to lightweight ones such as MobileNet and DeiT-Tiny. Our results demonstrate that VAD transfers effectively to road scenes. Notably, Tiny-Dinomaly achieves the best accuracy-efficiency trade-off for edge deployment, matching full-scale localization performance at a fraction of the memory cost. This study represents a concrete step toward safer, more responsible deployment of autonomous vehicles, ultimately improving protection for passengers, pedestrians, and all road users.
Authors: Yan Li, Zezi Zeng, Yifan Yang, Yuqing Yang, Ning Liao, Weiwei Guo, Lili Qiu, Mingxi Cheng, Qi Dai, Zhendong Wang, Zhengyuan Yang, Xue Yang, Ji Li, Lijuan Wang, Chong Luo
Abstract: The rapid progress of Artificial Intelligence Generated Content (AIGC) tools enables images, videos, and visualizations to be created on demand for webpage design, offering a flexible and increasingly adopted paradigm for modern UI/UX. However, directly integrating such tools into automated webpage generation often leads to style inconsistency and poor global coherence, as elements are generated in isolation. We propose MM-WebAgent, a hierarchical agentic framework for multimodal webpage generation that coordinates AIGC-based element generation through hierarchical planning and iterative self-reflection. MM-WebAgent jointly optimizes global layout, local multimodal content, and their integration, producing coherent and visually consistent webpages. We further introduce a benchmark for multimodal webpage generation and a multi-level evaluation protocol for systematic assessment. Experiments demonstrate that MM-WebAgent outperforms code-generation and agent-based baselines, especially on multimodal element generation and integration. Code & Data: https://aka.ms/mm-webagent.
Authors: Maximilian J. Kramer, Paul Boes, Jens Eisert
Abstract: The Boolean Satisfiability problem (SAT), as the prototypical $\mathsf{NP}$-complete problem, is crucial in both theoretical computer science and practical applications. To address this problem, stochastic local search (SLS) algorithms, which iteratively and randomly update candidate assignments, present an important and theoretically well-studied class of solvers. Recent theoretical advancements have identified conditions under which SLS solvers efficiently solve SAT instances, provided they have access to suitable ``oracles'', i.e., instance-specific distribution samples. We propose leveraging machine learning models, particularly graph neural networks (GNN), as oracles to enhance the performance of SLS solvers. Our approach, evaluated on random and pseudo-industrial SAT instances, demonstrates a significant performance improvement regarding step counts and solved instances. Our work bridges theoretical results and practical applications, highlighting the potential of purpose-trained SAT solvers with performance guarantees.
Authors: Sagar Pathak, Bidhya Shrestha
Abstract: This research project presents the implementation of a Deep Q-Learning Network (DQN) for a self-driving car on a 2-dimensional (2D) custom track, with the objective of enhancing the DQN network's performance. It encompasses the development of a custom driving environment using Pygame on a track surrounding the University of Memphis map, as well as the design and implementation of the DQN model. The algorithm utilizes data from 7 sensors installed in the car, which measure the distance between the car and the track. These sensors are positioned in front of the vehicle, spaced 20 degrees apart, enabling them to sense a wide area ahead. We successfully implemented the DQN and also a modified version of the DQN with a priority-based action selection mechanism, which we refer to as modified DQN. The model was trained over 1000 episodes, and the average reward received by the agent was found to be around 40, which is approximately 60% higher than the original DQN and around 50% higher than the vanilla neural network.
Authors: Amy Rafferty, Rishi Ramaesh, Ajitha Rajan
Abstract: Deep learning models have shown promise in lung pathology detection from chest X-rays, but widespread clinical adoption remains limited due to opaque model decision-making. In prior work, we introduced ClinicXAI, a human-centric, expert-guided concept bottleneck model (CBM) designed for interpretable lung cancer diagnosis. We now extend that approach and present XpertXAI, a generalizable expert-driven model that preserves human-interpretable clinical concepts while scaling to detect multiple lung pathologies. Using a high-performing InceptionV3-based classifier and a public dataset of chest X-rays with radiology reports, we compare XpertXAI against leading post-hoc explainability methods and an unsupervised CBM, XCBs. We assess explanations through comparison with expert radiologist annotations and medical ground truth. Although XpertXAI is trained for multiple pathologies, our expert validation focuses on lung cancer. We find that existing techniques frequently fail to produce clinically meaningful explanations, omitting key diagnostic features and disagreeing with radiologist judgments. XpertXAI not only outperforms these baselines in predictive accuracy but also delivers concept-level explanations that better align with expert reasoning. While our focus remains on explainability in lung cancer detection, this work illustrates how human-centric model design can be effectively extended to broader diagnostic contexts - offering a scalable path toward clinically meaningful explainable AI in medical diagnostics.
Authors: Sitong Fang, Wenjing Cao, Jiahao Li, Xuyao Wang, Juntao Dai, Chi-Min Chan, Sirui Han, Yike Guo, Yaodong Yang, Jiaming Ji
Abstract: Reasoning models have attracted increasing attention for their ability to tackle complex tasks, embodying the System II (slow thinking) paradigm in contrast to System I (fast, intuitive responses). Yet a key question remains: Does slower reasoning necessarily lead to more truthful answers? Our findings suggest otherwise. We conduct the first systematic study of the inverse scaling law in slow-thinking paradigms for multimodal reasoning. We find that when confronted with incomplete or misleading visual inputs, slow-thinking models are more prone to fabricating plausible yet false details to justify untruthful reasoning. To analyze this behavior, we construct a 5,000-sample hierarchical prompt dataset annotated by 50 human participants. The prompts progressively increase in complexity, revealing a consistent pattern: slower reasoning models tend to follow depth-first search (DFS) thinking, persistently exploring flawed premises, while faster chat models favor breadth-first search (BFS) inference, showing greater caution under uncertainty. These findings reveal a critical vulnerability of reasoning models: while effective in structured domains such as math, their DFS-style reasoning becomes fragile when confronted with ambiguous, multimodal inputs.
Authors: Baochang Ren, Shuofei Qiao, Da Zheng, Huajun Chen, Ningyu Zhang
Abstract: Large Language Models (LLMs), particularly slow-thinking models, often exhibit severe hallucination, outputting incorrect content due to an inability to accurately recognize knowledge boundaries during reasoning. While Reinforcement Learning (RL) can enhance complex reasoning abilities, its outcome-oriented reward mechanism often lacks factual supervision over the thinking process, further exacerbating the hallucination problem. To address the high hallucination in slow-thinking models, we propose Knowledge-enhanced RL, KnowRL. KnowRL guides models to perform fact-based slow thinking by integrating a factuality reward, based on knowledge verification, into the RL training process, helping them recognize their knowledge boundaries. KnowRL guides models to perform fact-based slow thinking by integrating a factuality reward, based on knowledge verification, into the RL training process, helping them recognize their knowledge boundaries. This targeted factual input during RL training enables the model to learn and internalize fact-based reasoning strategies. By directly rewarding adherence to facts within the reasoning steps, KnowRL fosters a more reliable thinking process. Experimental results on three hallucination evaluation datasets and two reasoning evaluation datasets demonstrate that KnowRL effectively mitigates hallucinations in slow-thinking models while maintaining their original strong reasoning capabilities. Our code is available at https://github.com/zjunlp/KnowRL.
Authors: Zijian Zhao, Sen Li
Abstract: Order dispatch is a critical task in ride-sharing systems with Autonomous Vehicles (AVs), directly influencing efficiency and profits. Recently, Multi-Agent Reinforcement Learning (MARL) has emerged as a promising solution to this problem by decomposing the large state and action spaces among individual agents, effectively addressing the Curse of Dimensionality (CoD) in transportation market, which is caused by the substantial number of vehicles, passengers, and orders. However, conventional MARL-based approaches heavily rely on accurate estimation of the value function, which becomes problematic in large-scale, highly uncertain environments. To address this issue, we propose two novel methods that bypass value function estimation, leveraging the homogeneous property of AV fleets. First, we draw an analogy between AV fleets and groups in Group Relative Policy Optimization (GRPO), adapting it to the order dispatch task. By replacing the Proximal Policy Optimization (PPO) baseline with the group average reward-to-go, GRPO eliminates critic estimation errors and reduces training bias. Inspired by this baseline replacement, we further propose One-Step Policy Optimization (OSPO), demonstrating that the optimal policy can be trained using only one-step group rewards under a homogeneous fleet. Experiments on a real-world ride-hailing dataset show that both GRPO and OSPO achieve promising performance across all scenarios, efficiently optimizing pickup times and the number of served orders using simple Multilayer Perceptron (MLP) networks. Furthermore, OSPO outperforms GRPO in all scenarios, attributed to its elimination of bias caused by the bounded time horizon of GRPO. Our code, trained models, and processed data are provided at https://github.com/RS2002/OSPO .
Authors: Zihan Zheng, Tianle Cui, Taoran Wang, Fengtao Wang, Jiahui Pan, Lewei He, Qianglong Chen
Abstract: Despite significant advances in LLM-driven GUI agents, the field remains constrained by the challenge of reconciling high-fidelity realism with verifiable evaluation accuracy. To address this, we introduce NaturalGAIA, a verifiable evaluation dataset grounded in real-world human GUI interaction intents. By decoupling logical causal pathways from linguistic narratives, it rigorously simulates natural human intent, characterized by cognitive non-linearity and contextual dependencies. Furthermore, we propose LightManus-Jarvis, a hierarchical collaborative framework where LightManus manages dynamic topological planning and context evolution, while Jarvis~ensures execution precision via hybrid visual-structural perception. Experiments demonstrate that our approach achieves a Weighted Pathway Success Rate of 45.6%, significantly outperforming the state-of-the-art baseline (21.1%), while reducing token consumption by 75% and execution time by 76%. These results validate the efficacy of the macro-planning and micro-execution paradigm in handling complex naturalized tasks. Our code is publicly available at: https://github.com/KeLes-Coding/NatureGAIA.
Authors: Wenquan Ma, Jiayan Nan, Wenlong Wu, Yize Chen
Abstract: Memory systems for LLM agents struggle to determine what information deserves retention. Existing approaches rely on predefined heuristics such as importance scores, emotional tags, or factual templates, encoding designer intuition rather than learning from the data itself. Inspired by cognitive ideas, we propose NEMORI, an adaptive memory distillation framework that casts the assessment of experience's future utility as a matter of predictability. Specifically, NEMORI comprises two cascading modules: Episodic Memory Integration transforms raw interactions into coherent narratives, and Semantic Knowledge Distillation extracts insights via prediction error. Centering on distillation, the framework remains agnostic to downstream management. Extensive experiments confirm that NEMORI achieves strong performance, efficiency, and storage reduction. Our work suggests that observing the intrinsic properties of interaction sequences offers a viable, data-driven alternative to heuristic-based memory design. Code: https://github.com/nemori-ai/nemori.
Authors: Ruiying Ma, Chieh-Jan Mike Liang, Yanjie Gao, Francis Y. Yan
Abstract: Designing system algorithms remains challenging, where the discontinuous nature of the solution space often forces system engineers to rely on generic heuristics at the expense of performance. We study whether LLMs can practically drive algorithm generation, and find that they are biased towards well-known generic designs, rather than making the creative leaps needed to navigate the discontinuous solution space. To address this limitation, we introduce MetaMuse, a framework for creative ideation built on three self-reflection principles: (1) quantifying solution diversity and usefulness in measurable performance space, rather than abstract idea space, (2) steering ideation through external stimuli, rather than internal randomness, and (3) constructing executable solutions using waypoint reasoning, rather than free-form chain-of-thought. Considering two critical online problems at a global cloud provider, extensive evaluations show that MetaMuse can generate high-performing solutions: it reduces cache misses by up to 35.76% in cache replacement and reduces bin usage by up to 30.93% in online bin packing.
Authors: Ziying Zhang, Yaqing Wang, Quanming Yao
Abstract: Meta reasoning behaviors work as a skeleton to guide large language model (LLM) reasoning, thus help to improve reasoning performance. However, prior researches implement meta reasoning skeleton with manually designed structure, limiting ability to adapt to query-specific requirement and capture intricate logical dependency among reasoning steps. To deal with the challenges, we represent meta reasoning skeleton with directed acyclic graph (DAG) to unify skeletons proposed in prior works and model intricate logical dependency. Then we propose AutoMR, a framework that searches for query-aware meta reasoning skeleton automatically inspired by automated machine learning (AutoML). Specifically, we construct search space based on DAG representation of skeleton and then formulate the search problem. We design a dynamic skeleton sampling algorithm by expanding meta reasoning skeleton along with reasoning context at inference time. This algorithm can derive any meta reasoning skeleton in search space efficiently and adapt skeleton to evolving base reasoning context, thus enable efficient query-aware skeleton search. We conduct experiments on extensive benchmark datasets. Experimental results show that AutoMR achieves better reasoning performance than previous works broadly.
Authors: Can Xie, Ruotong Pan, Xiangyu Wu, Yunfei Zhang, Jiayi Fu, Tingting Gao, Guorui Zhou
Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has shown significant promise for enhancing the reasoning capabilities of large language models (LLMs). However, prevailing algorithms like GRPO broadcast a uniform advantage signal across all tokens in a sequence. This coarse-grained approach overlooks the pivotal role of uncertain, high-stakes decisions during reasoning, leading to inefficient exploration and the well-documented problem of entropy collapse. To address this, we introduce UnCertainty-aware Advantage Shaping (UCAS), a model-free method that refines credit assignment by leveraging the model's internal uncertainty signals. UCAS operates in two stages: it first modulates the response-level advantage using a logit-space self-confidence proxy, and then applies an asymmetric token-level penalty based on raw logit certainty. This dual mechanism encourages exploration of high-uncertainty paths that yield correct answers while penalizing overconfident yet erroneous reasoning, effectively balancing the exploration-exploitation trade-off. Extensive experiments on five mathematical reasoning benchmarks show that UCAS significantly outperforms strong RLVR baselines across multiple model scales, including 1.5B and 7B. Our analysis confirms that UCAS not only achieves higher rewards but also promotes greater reasoning diversity and successfully mitigates entropy collapse. Code is available at https://github.com/xvolcano02/UCAS.
Authors: Rikard Rosenbacke, Carl Rosenbacke, Victor Rosenbacke, Martin McKee
Abstract: As large language models (LLMs) become integrated into everyday and high-stakes decision-making, they inherit the ambiguity and biases of human language. While they produce fluent and coherent outputs, they rely on statistical pattern prediction rather than grounded reasoning, creating a risk of outputs that are plausible but incorrect. This paper argues that these failures are not only technical but cognitive. LLMs reproduce associative patterns similar to intuitive human reasoning, amplifying systematic misinterpretations when combined with human users. To analyse this, we introduce the Rose-Frame, a cognitive-epistemological framework for diagnosing breakdowns in human-AI interaction. The framework identifies three recurrent traps: (i) map vs territory, distinguishing representations from reality; (ii) intuition vs reason, separating fast associative judgments from reflective reasoning; and (iii) conflict vs confirmation, examining whether ideas are critically tested or mutually reinforced. These mechanisms can compound into epistemic drift when human and model reasoning interact. We show how these failures emerge in practice and propose human-side interventions, including interpretive cues, reflective prompts, and structured disagreement, to stabilise reasoning. Rather than modifying models, the framework focuses on governing interaction. The central claim is that fluency can create an illusion of understanding. Aligning AI therefore requires not only technical improvements but structures that enable reflective and falsifiable human oversight.
Authors: Wenhao Wang, Peizhi Niu, Zhao Xu, Zhaoyu Chen, Jian Du, Yaxin Du, Xianghe Pang, Keduan Huang, Yanfeng Wang, Qiang Yan, Siheng Chen
Abstract: Large Language Models (LLMs) increasingly rely on external tools to perform complex, realistic tasks, yet their ability to utilize the rapidly expanding Model Contextual Protocol (MCP) ecosystem remains limited. Existing MCP research covers few servers, depends on costly manual curation, and lacks training support, hindering progress toward real-world deployment. To overcome these limitations, we introduce MCP-Flow, an automated web-agent-driven pipeline for large-scale server discovery, data synthesis, and model training. MCP-Flow collects and filters data from 1166 servers and 11536 tools, producing 68733 high-quality instruction-function call pairs and 6439 trajectories, far exceeding prior work in scale and diversity. Extensive experiments demonstrate MCP-Flow's effectiveness in driving superior MCP tool selection, function-call generation, and enhanced agentic task performance. MCP-Flow thus provides a scalable foundation for advancing LLM agents' proficiency in real-world MCP environments. MCP-Flow is publicly available at https://github.com/wwh0411/MCP-Flow.
Authors: Ali Taheri, Alireza Taban, Sadegh Soudjani, Ashutosh Trivedi
Abstract: Safety verification of dynamical systems via barrier certificates is essential for ensuring correctness in autonomous applications. Synthesizing these certificates involves discovering mathematical functions with current methods suffering from poor scalability, dependence on carefully designed templates, and exhaustive or incremental function-space searches. They also demand substantial manual expertise--selecting templates, solvers, and hyperparameters, and designing sampling strategies--requiring both theoretical and practical knowledge traditionally shared through linguistic reasoning rather than formalized methods. This motivates a key question: can such expert reasoning be captured and operationalized by language models? We address this by introducing an LLM-based agentic framework for barrier certificate synthesis. The framework uses natural language reasoning to propose, refine, and validate candidate certificates, integrating LLM-driven template discovery with SMT-based verification, and supporting barrier-controller co-synthesis to ensure consistency between safety certificates and controllers. To evaluate this capability, we introduce BarrierBench, a benchmark of 100 dynamical systems spanning linear, nonlinear, discrete-time, and continuous-time settings. Our experiments assess not only the effectiveness of LLM-guided barrier synthesis but also the utility of retrieval-augmented generation and agentic coordination strategies in improving its reliability and performance. Across these tasks, the framework achieves more than 90% success in generating valid certificates. By releasing BarrierBench and the accompanying toolchain, we aim to establish a community testbed for advancing the integration of language-based reasoning with formal verification in dynamical systems. The benchmark is publicly available at https://hycodev.com/dataset/barrierbench
Authors: Tuan-Anh Le, Anh Mai Vu, David Yang, Akash Awasthi, Hien Van Nguyen
Abstract: IMACT-CXR is an interactive multi-agent conversational tutor that helps trainees interpret chest X-rays by unifying spatial annotation, gaze analysis, knowledge retrieval, and image-grounded reasoning in a single AutoGen-based workflow. The tutor simultaneously ingests learner bounding boxes, gaze samples, and free-text observations. Specialized agents evaluate localization quality, generate Socratic coaching, retrieve PubMed evidence, suggest similar cases from REFLACX, and trigger NV-Reason-CXR-3B for vision-language reasoning when mastery remains low or the learner explicitly asks. Bayesian Knowledge Tracing (BKT) maintains skill-specific mastery estimates that drive both knowledge reinforcement and case similarity retrieval. A lung-lobe segmentation module derived from a TensorFlow U-Net enables anatomically aware gaze feedback, and safety prompts prevent premature disclosure of ground-truth labels. We describe the system architecture, implementation highlights, and integration with the REFLACX dataset for real DICOM cases. IMACT-CXR demonstrates responsive tutoring flows with bounded latency, precise control over answer leakage, and extensibility toward live residency deployment. Preliminary evaluation shows improved localization and diagnostic reasoning compared to baselines.
Authors: Xuyuan Liu, Shengyu Chen, Xinshuai Dong, Yanchi Liu, Xujiang Zhao, Haoyu Wang, Yujun Yan, Haifeng Chen, Zhengzhang Chen
Abstract: Large language models (LLMs) often produce incorrect or outdated content after being employed. Efficient and accurate knowledge updates without costly retraining are a major challenge. This problem is particularly challenging in lifelong settings, where complex, unstructured knowledge must coexist without interference. We introduce RILKE (Representation Intervention for Lifelong KnowledgE Control), a robust and scalable method that treats knowledge control as interventions within the model's representation space. Leveraging representation-space expressiveness, we identify two key properties enabling RILKE to achieve fine-grained control over complex, unstructured knowledge while maintaining general utility with frozen base weights. During training, RILKE learns paraphrase-robust and edit-localized modules that limit each update to a low-dimensional subspace to minimize cross-edit interference. At inference, a query-adaptive router selects the appropriate module to guide the model's generation. Across LLaMA and Qwen models, RILKE scales effectively to large-scale benchmarks, demonstrating high edit success and strong paraphrase generalization while preserving general utility with modest memory overhead. These results show RILKE is an effective and scalable solution for lifelong knowledge control in LLMs.
Authors: Austin Spizzirri
Abstract: Static content-based AI value alignment is insufficient for robust alignment under capability scaling, distributional shift, and increasing autonomy. This holds for any approach that treats alignment as optimizing toward a fixed formal value-object, whether reward function, utility function, constitutional principles, or learned preference representation. Three philosophical results create compounding difficulties: Hume's is-ought gap (behavioral data underdetermines normative content), Berlin's value pluralism (human values resist consistent formalization), and the extended frame problem (any value encoding will misfit future contexts that advanced AI creates). RLHF, Constitutional AI, inverse reinforcement learning, and cooperative assistance games each instantiate this specification trap, and their failure modes reflect structural vulnerabilities, not merely engineering limitations that better data or algorithms will straightforwardly resolve. Known workarounds for individual components face mutually reinforcing difficulties when the specification is closed: the moment it ceases to update from the process it governs. Drawing on compatibilist philosophy, the paper argues that behavioral compliance under training conditions does not guarantee robust alignment under novel conditions, and that this gap grows with system capability. For value-laden autonomous systems, known closed approaches face structural vulnerabilities that worsen with capability. The constructive burden shifts to open, developmentally responsive approaches, though whether such approaches can be achieved remains an empirical question.
Authors: Haoyu Dong, Pengkun Zhang, Yan Gao, Xuanyu Dong, Yilin Cheng, Mingzhe Lu, Zikun Zhu, Adina Yakefu, Shuxin Zheng
Abstract: We introduce FinWorkBench (a.k.a. Finch) for evaluating AI agents on real-world, enterprise-grade finance and accounting workflows that interleave data entry, structuring, formatting, web search, cross-file retrieval, calculation, modeling, validation, translation, visualization, and reporting. Finch is sourced from authentic enterprise workspaces from Enron (15,000 files and 500,000 emails) and other financial institutions, covering the period 2000--2025 and preserving the in-the-wild messiness of multimodal artifacts such as tables and charts across diverse domains including budgeting, trading, asset management, and operational management. We propose a workflow construction process that combines LLM-assisted mining of workflows from authentic enterprise environments with expert annotation: (1) LLM-assisted, expert-verified derivation of workflows from real-world email threads and spreadsheet version histories, and (2) meticulous annotation requiring over 700 hours of expert effort. This yields 172 composite workflows with 384 tasks, involving 1,710 spreadsheets with 27 million cells, along with PDFs and other artifacts, capturing the intrinsically messy, long-horizon, knowledge-intensive, and collaborative nature of real-world enterprise work. We conduct both human and automated evaluations of frontier AI systems, including GPT-5.1, Claude Sonnet 4.5, Claude Opus 4.5, Gemini 3 Pro, Grok 4, and Qwen 3 Max. Under human evaluation, GPT-5.1 Pro spends an average of 16.8 minutes per workflow yet passes only 38.4% of workflows. Comprehensive case studies further surface the challenges that real-world enterprise workflows pose for AI agents.
Authors: Dongming Jiang, Yi Li, Guanpeng Li, Bingzhe Li
Abstract: Memory-Augmented Generation (MAG) extends Large Language Models with external memory to support long-context reasoning, but existing approaches largely rely on semantic similarity over monolithic memory stores, entangling temporal, causal, and entity information. This design limits interpretability and alignment between query intent and retrieved evidence, leading to suboptimal reasoning accuracy. In this paper, we propose MAGMA, a multi-graph agentic memory architecture that represents each memory item across orthogonal semantic, temporal, causal, and entity graphs. MAGMA formulates retrieval as policy-guided traversal over these relational views, enabling query-adaptive selection and structured context construction. By decoupling memory representation from retrieval logic, MAGMA provides transparent reasoning paths and fine-grained control over retrieval. Experiments on LoCoMo and LongMemEval demonstrate that MAGMA consistently outperforms state-of-the-art agentic memory systems in long-horizon reasoning tasks.
Authors: Qirui Mi, Zhijian Ma, Mengyue Yang, Haoxuan Li, Yisen Wang, Haifeng Zhang, Jun Wang
Abstract: LLM-driven agents demonstrate strong performance in sequential decision-making but often rely on on-the-fly reasoning, re-deriving solutions even in recurring scenarios. This insufficient experience reuse leads to computational redundancy and execution instability. To bridge this gap, we propose Skill-Pro, a framework that enables agents to autonomously learn reusable procedural skills from interaction experiences without parameter updates. By formalizing a Skill-MDP, Skill-Pro transforms passive episodic narratives into executable Skills defined by activation, execution, and termination conditions to ensure executability. To achieve reliable reusability without capability degradation, we introduce Non-Parametric PPO, which leverages semantic gradients for high-quality candidate generation and a PPO Gate for robust Skill verification. Through score-based maintenance, Skill-Pro sustains compact, high-quality procedural memory. Experimental results across in-domain, cross-task, and cross-agent scenarios demonstrate that Skill-Pro achieves superior reuse rates and significant performance gains with extreme memory compression. Visualized evolutionary trajectories and Skill distributions further reveal how Skill-Pro transparently accumulates, refines, and reuses procedural knowledge to facilitate long-term autonomy.
Authors: Siyuan Li, Yunjia Wu, Yiyong Xiao, Pingyang Huang, Peize Li, Ruitong Liu, Yan Wen, Te Sun
Abstract: Temporal knowledge graph (TKG) forecasting requires predicting future facts by jointly modeling structural dependencies within each snapshot and temporal evolution across snapshots. However, most existing methods are stateless: they recompute entity representations at each timestamp from a limited query window, leading to episodic amnesia and rapid decay of long-term dependencies. To address this limitation, we propose Entity State Tuning (EST), an encoder-agnostic framework that endows TKG forecasters with persistent and continuously evolving entity states. EST maintains a global state buffer and progressively aligns structural evidence with sequential signals via a closed-loop design. Specifically, a topology-aware state perceiver first injects entity-state priors into structural encoding. Then, a unified temporal context module aggregates the state-enhanced events with a pluggable sequence backbone. Subsequently, a dual-track evolution mechanism writes the updated context back to the global entity state memory, balancing plasticity against stability. Experiments on multiple benchmarks show that EST consistently improves diverse backbones and achieves state-of-the-art performance, highlighting the importance of state persistence for long-horizon TKG forecasting.
Authors: Tan Bui-Thanh
Abstract: Can artificial intelligence truly contribute to creative mathematical research, or does it merely automate routine calculations while introducing risks of error? We provide empirical evidence through a detailed case study: the discovery of novel error representations and bounds for Hermite quadrature rules via systematic human-AI collaboration. Working with multiple AI assistants, we extended results beyond what manual work achieved, formulating and proving several theorems with AI assistance. The collaboration revealed both remarkable capabilities and critical limitations. AI excelled at algebraic manipulation, systematic proof exploration, literature synthesis, and LaTeX preparation. However, every step required rigorous human verification, mathematical intuition for problem formulation, and strategic direction. We document the complete research workflow with unusual transparency, revealing patterns in successful human-AI mathematical collaboration and identifying failure modes researchers must anticipate. Our experience suggests that, when used with appropriate skepticism and verification protocols, AI tools can meaningfully accelerate mathematical discovery while demanding careful human oversight and deep domain expertise.
Authors: Drew Prinster, Clara Fannjiang, Ji Won Park, Kyunghyun Cho, Anqi Liu, Suchi Saria, Samuel Stanton
Abstract: An agent must try new behaviors to explore and improve. In high-stakes environments, an agent that violates safety constraints may cause harm and must be taken offline, curtailing any future interaction. Imitating old behavior is safe, but excessive conservatism discourages exploration. How much behavior change is too much? We show how to use any safe reference policy as a probabilistic regulator for any optimized but untested policy. Conformal calibration on data from the safe policy determines how aggressively the new policy can act, while provably enforcing the user's declared risk tolerance. Unlike conservative optimization methods, we do not assume the user has identified the correct model class nor tuned any hyperparameters. Unlike previous conformal methods, our theory provides finite-sample guarantees even for non-monotonic bounded loss functions. Our experiments on applications ranging from natural language question answering to biomolecular engineering show that safe exploration is not only possible from the first moment of deployment, but can also improve performance.
Authors: Jiangyu Chen
Abstract: Automated design of chemical formulations is a cornerstone of materials science, yet it requires navigating a high-dimensional combinatorial space involving discrete compositional choices and continuous geometric constraints. Existing Large Language Model (LLM) agents face significant challenges in this setting, including context window limitations during long-horizon reasoning and path-dependent exploration that may lead to mode collapse. To address these issues, we introduce AI4S-SDS, a closed-loop neuro-symbolic framework that integrates multi-agent collaboration with a tailored Monte Carlo Tree Search (MCTS) engine. We propose a Sparse State Storage mechanism with Dynamic Path Reconstruction, which decouples reasoning history from context length and enables arbitrarily deep exploration under fixed token budgets. To reduce local convergence and improve coverage, we implement a Global--Local Search Strategy: a memory-driven planning module adaptively reconfigures the search root based on historical feedback, while a Sibling-Aware Expansion mechanism promotes orthogonal exploration at the node level. Furthermore, we bridge symbolic reasoning and physical feasibility through a Differentiable Physics Engine, employing a hybrid normalized loss with sparsity-inducing regularization to optimize continuous mixing ratios under thermodynamic constraints. Empirical results show that AI4S-SDS achieves full validity under the adopted HSP-based physical constraints and substantially improves exploration diversity compared to baseline agents. In preliminary lithography experiments, the framework identifies a novel photoresist developer formulation that demonstrates competitive or superior performance relative to a commercial benchmark, highlighting the potential of diversity-driven neuro-symbolic search for scientific discovery.
Authors: Alvin Rajkomar, Pavan Sudarshan, Angela Lai, Lily Peng
Abstract: Background: Clinical trials rely on transparent inclusion criteria to ensure generalizability. In contrast, benchmarks validating health-related large language models (LLMs) rarely characterize the "patient" or "query" populations they contain. Without defined composition, aggregate performance metrics may misrepresent model readiness for clinical use. Methods: We analyzed 18,707 consumer health queries across six public benchmarks using LLMs as automated coding instruments to apply a standardized 16-field taxonomy profiling context, topic, and intent. Results: We identified a structural "validity gap." While benchmarks have evolved from static retrieval to interactive dialogue, clinical composition remains misaligned with real-world needs. Although 42% of the corpus referenced objective data, this was polarized toward wellness-focused wearable signals (17.7%); complex diagnostic inputs remained rare, including laboratory values (5.2%), imaging (3.8%), and raw medical records (0.6%). Safety-critical scenarios were effectively absent: suicide/self-harm queries comprised <0.7% of the corpus and chronic disease management only 5.5%. Benchmarks also neglected vulnerable populations (pediatrics/older adults <11%) and global health needs. Conclusions: Evaluation benchmarks remain misaligned with real-world clinical needs, lacking raw clinical artifacts, adequate representation of vulnerable populations, and longitudinal chronic care scenarios. The field must adopt standardized query profiling--analogous to clinical trial reporting--to align evaluation with the full complexity of clinical practice.
Authors: Richard Servajean, Philippe Servajean
Abstract: A robust decision-making process must take into account uncertainty, especially when the choice involves inherent risks. Because artificial Intelligence (AI) systems are increasingly integrated into decision-making workflows, managing uncertainty relies more and more on the metacognitive capabilities of these systems; i.e, their ability to assess the reliability of and regulate their own decisions. Hence, it is crucial to employ robust methods to measure the metacognitive abilities of AI. This paper is primarily a methodological contribution arguing for the adoption of the meta-d' framework as the gold standard for assessing the metacognitive sensitivity of AIs--the ability to generate confidence ratings that distinguish correct from incorrect responses. Moreover, we propose to leverage signal detection theory (SDT) to measure the ability of AIs to spontaneously regulate their decisions based on uncertainty and risk. To demonstrate the practical utility of these psychophysical frameworks, we conduct two series of experiments on three large language models (LLMs)--GPT-5, DeepSeek-V3.2-Exp, and Mistral-Medium-2508. In the first experiments, LLMs performed a primary judgment followed by a confidence rating. In the second, LLMs only performed the primary judgment, while we manipulated the risk associated with either response. On the one hand, applying the meta-d' framework allows us to conduct comparisons along three axes: comparing an LLM to optimality, comparing different LLMs on a given task, and comparing the same LLM across different tasks. On the other hand, SDT allows us to assess whether LLMs become more conservative when risk is high.
Authors: Hyunji Nam, Dorottya Demszky
Abstract: LLMs are increasingly used for high-stakes decision-making, yet their sensitivity to spurious contextual information can introduce harmful biases. This is a critical concern when models are deployed for tasks like evaluating teachers' instructional quality, where biased assessment can affect teachers' professional development and career trajectories. We investigate model robustness to spurious social contexts using the largest publicly available dataset of U.S. classroom transcripts (NCTE) paired with expert rubric scores. Evaluating seven frontier and open-weight models across seven categories of spurious contexts -- including teacher experience, education level, demographic identity, and sycophancy-inducing framings -- we find that irrelevant contextual information can shift model predictions by up to 1.48 points on a 7-point scale, with larger models sometimes exhibiting greater sensitivity despite higher predictive accuracy. Mitigations using prompts and standard direct preference optimization (DPO) prove largely insufficient. We propose **Debiasing-DPO**,, a self-supervised training method that pairs neutral reasoning generated from the query alone, with the model's biased reasoning generated with both the query and additional spurious context. We further combine this objective with supervised fine-tuning on ground-truth labels to prevent losses in predictive accuracy. Applied to Llama 3B \& 8B and Qwen 3B \& 7B Instruct models, Debiasing-DPO reduces bias by 84\% and improves predictive accuracy by 52\% on average. Our findings from the educational case study highlight that robustness to spurious context is not a natural byproduct of model scaling and that our proposed method can yield substantial gains in both accuracy and robustness for prompt-based prediction tasks.
Authors: Albert Sadowski, Jaros{\l}aw A. Chudziak
Abstract: AI agents operating over extended time horizons accumulate experiences that serve multiple concurrent goals, and must often maintain conflicting interpretations of the same events. A concession during a client negotiation encodes as a ``trust-building investment'' for one strategic goal and a ``contractual liability'' for another. Current memory architectures assume a single correct encoding, or at best support multiple views over unified storage. We propose Rashomon Memory: an architecture where parallel goal-conditioned agents encode experiences according to their priorities and negotiate at query time through argumentation. Each perspective maintains its own ontology and knowledge graph. At retrieval, perspectives propose interpretations, critique each other's proposals using asymmetric domain knowledge, and Dung's argumentation semantics determines which proposals survive. The resulting attack graph is itself an explanation: it records which interpretation was selected, which alternatives were considered, and on what grounds they were rejected. We present a proof-of-concept showing that retrieval modes (selection, composition, conflict surfacing) emerge from attack graph topology, and that the conflict surfacing mode, where the system reports genuine disagreement rather than forcing resolution, lets decision-makers see the underlying interpretive conflict directly.
Authors: Myeongsoo Kim, Joe Hsu, Dingmin Wang, Shweta Garg, Varun Kumar, Murali Krishna Ramanathan
Abstract: LLM-based code agents treat repositories as unstructured text, applying edits through brittle string matching that frequently fails due to formatting drift or ambiguous patterns. We propose reframing the codebase as a structured action space where agents operate on named AST entities rather than text spans. Our framework, CODESTRUCT, provides readCode for retrieving complete syntactic units and editCode for applying syntax-validated transformations to semantic program elements. Evaluated on SWE-Bench Verified across six LLMs, CODESTRUCT improves Pass@1 accuracy by 1.2-5.0% while reducing token consumption by 12-38% for most models. Models that frequently fail to produce valid patches under text-based interfaces benefit most: GPT-5-nano improves by 20.8% as empty-patch failures drop from 46.6% to 7.2%. On CodeAssistBench, we observe consistent accuracy gains (+0.8-4.4%) with cost reductions up to 33%. Our results show that structure-aware interfaces offer a more reliable foundation for code agents.
Authors: Shantam Srivastava, Mahesh Bhosale, David Doermann, Mingchen Gao
Abstract: Interpreting chest X-rays is inherently challenging due to the overlap between anatomical structures and the subtle presentation of many clinically significant pathologies, making accurate diagnosis time-consuming even for experienced radiologists. Recent radiology-focused foundation models, such as LLaVA-Rad and Maira-2, have positioned multi-modal large language models (MLLMs) at the forefront of automated radiology report generation (RRG). However, despite these advances, current foundation models generate reports in a single forward pass. This decoding strategy diminishes attention to visual tokens and increases reliance on language priors as generation proceeds, which in turn introduces spurious pathology co-occurrences in the generated reports. To mitigate these limitations, we propose Category-Wise Contrastive Decoding (CWCD), a novel and modular framework designed to enhance structured radiology report generation (SRRG). Our approach introduces category-specific parameterization and generates category-wise reports by contrasting normal X-rays with masked X-rays using category-specific visual prompts. Experimental results demonstrate that CWCD consistently outperforms baseline methods across both clinical efficacy and natural language generation metrics. An ablation study further elucidates the contribution of each architectural component to overall performance.
Authors: Chen Huang, Zitan Jiang, Changyi Zou, Wenqiang Lei, See-Kiong Ng
Abstract: Customer service chatbots are increasingly expected to serve not merely as reactive support tools for users, but as strategic interfaces for harvesting high-value information and business intelligence. In response, we make three main contributions. 1) We introduce and define a novel task of Proactive Information Probing, which optimizes when to probe users for pre-specified target information while minimizing conversation turns and user friction. 2) We propose PROCHATIP, a proactive chatbot framework featuring a specialized conversation strategy module trained to master the delicate timing of probes. 3) Experiments demonstrate that PROCHATIP significantly outperforms baselines, exhibiting superior capability in both information probing and service quality. We believe that our work effectively redefines the commercial utility of chatbots, positioning them as scalable, cost-effective engines for proactive business intelligence. Our code is available at https://github.com/SCUNLP/PROCHATIP.
Authors: Charafeddine Mouzouni
Abstract: We introduce Context Kubernetes, an architecture for orchestrating enterprise knowledge in agentic AI systems, with a prototype implementation and eight experiments. The core observation is that delivering the right knowledge, to the right agent, with the right permissions, at the right freshness -- across an entire organization -- is structurally analogous to the container orchestration problem Kubernetes solved a decade ago. We formalize six core abstractions, a YAML-based declarative manifest for knowledge-architecture-as-code, a reconciliation loop, and a three-tier agent permission model where agent authority is always a strict subset of human authority. On synthetic seed data, we compare four governance baselines of increasing strength: ungoverned RAG, ACL-filtered retrieval, RBAC-aware routing, and the full architecture. Each layer contributes a different capability: ACL filtering eliminates cross-domain leaks, intent routing reduces noise by 19 percentage points, and only the three-tier model blocks all five tested attack scenarios -- the one attack RBAC misses is an agent sending confidential pricing via email, which RBAC cannot distinguish from ordinary email. TLA+ model-checking verifies safety properties across 4.6 million reachable states with zero violations. A survey of four major platforms (Microsoft, Salesforce, AWS, Google) documents that none architecturally isolates agent approval channels. We identify four properties that make context orchestration harder than container orchestration, and argue these make the solution more valuable.
Authors: Georgianna "Blue" Lin, Rencong Jiang, No\'emie Elhadad, Xuhai "Orson" Xu
Abstract: Although artificial intelligence (AI) agents are increasingly proposed to support potentially longitudinal health tasks, such as symptom management, behavior change, and patient support, most current implementations fall short of facilitating user intent and fostering accountability. This contrasts with prior work on supporting longitudinal needs, where follow-up, coherent reasoning, and sustained alignment with individuals' goals are critical for both effectiveness and safety. In this paper, we draw on established clinical and personal health informatics frameworks to define what it would mean to orchestrate longitudinal health interactions with AI agents. We propose a multi-layer framework and corresponding agent architecture that operationalizes adaptation, coherence, continuity, and agency across repeated interactions. Through representative use cases, we demonstrate how longitudinal agents can maintain meaningful engagement, adapt to evolving goals, and support safe, personalized decision-making over time. Our findings underscore both the promise and the complexity of designing systems capable of supporting health trajectories beyond isolated interactions, and we offer guidance for future research and development in multi-session, user-centered health AI.
Authors: Weikang Zhang, Zimo Zhu, Zhichuan Yang, Chen Huang, Wenqiang Lei, See-Kiong Ng
Abstract: Simulating Standardized Patients with cognitive impairment offers a scalable and ethical solution for clinical training. However, existing methods rely on discrete prompt engineering and fail to capture the heterogeneity of deficits across varying domains and severity levels. To address this limitation, we propose StsPatient for the fine-grained simulation of cognitively impaired patients. We innovatively capture domain-specific features by extracting steering vectors from contrastive pairs of instructions and responses. Furthermore, we introduce a Stochastic Token Modulation (STM) mechanism to regulate the intervention probability. STM enables precise control over impairment severity while mitigating the instability of conventional vector methods. Comprehensive experiments demonstrate that StsPatient significantly outperforms baselines in both clinical authenticity and severity controllability.
Authors: Lei Lin, Jizhao Zhu, Yong Liu, Donghong Sun, Hongbo He, Yihua Du
Abstract: This paper addresses two limitations of large language models (LLMs) in solving complex problems: (1) their reasoning processes exhibit Bayesian-like stochastic generation, where each token is sampled from a context-dependent probability distribution, leading to inherently random decision trajectories rather than deterministic planning; (2) the reasoning and decision-making mechanisms are statically decoupled, meaning dynamically retrieved domain knowledge fails to dynamically adjust the underlying reasoning strategy. These dual deficiencies result in initial decisions lacking strategic anchoring and reasoning chains often failing to converge on correct solutions, as stochastic generation lacks mechanisms for trajectory correction or knowledge-guided optimization during sequential reasoning. To resolve these issues, we propose a problem-solving method integrated into the LLM's generation process to guide reasoning. This method, compatible with numerous LLMs and featuring reusable solutions, is grounded in a novel Heuristic-Classification-of-Thoughts prompting schema (HCoT). HCoT synergizes the LLM's reasoning ability with a structured problem space via a heuristic classification model that controls the reasoning process and provides reusable abstract solutions. Evaluated on two complex inductive reasoning tasks with ill-defined search spaces, HCoT outperforms existing approaches (e.g., Tree-of-Thoughts and Chain-of-Thoughts prompting) in performance. On the well-structured 24 Game task, HCoT demonstrates significantly higher token efficiency compared to the state-of-the-art Tree-of-Thoughts-Breadth-First-Search. In terms of both accuracy and token usage, HCoT achieves a Pareto frontier balance, offering a strong trade-off between performance and computational cost.
Authors: Jintao Xue, Xiao Li, Nianmin Zhang
Abstract: Human-robot collaborative manufacturing, a core aspect of Industry 5.0, emphasizes ergonomics to enhance worker well-being. This paper addresses the dynamic human-robot task planning and allocation (HRTPA) problem, which involves determining when to perform tasks and who should execute them to maximize efficiency while ensuring workers' physical fatigue remains within safe limits. The inclusion of fatigue constraints, combined with production dynamics, significantly increases the complexity of the HRTPA problem. Traditional fatigue-recovery models in HRTPA often rely on static, predefined hyperparameters. However, in practice, human fatigue sensitivity varies daily due to factors such as changed work conditions and insufficient sleep. To better capture this uncertainty, we treat fatigue-related parameters as inaccurate and estimate them online based on observed fatigue progression during production. To address these challenges, we propose PF-CD3Q, a safe reinforcement learning (safe RL) approach that integrates the particle filter with constrained dueling double deep Q-learning for real-time fatigue-predictive HRTPA. Specifically, we first develop PF-based estimators to track human fatigue and update fatigue model parameters in real-time. These estimators are then integrated into CD3Q by making task-level fatigue predictions during decision-making and excluding tasks that exceed fatigue limits, thereby constraining the action space and formulating the problem as a constrained Markov decision process (CMDP).
Authors: Jintao Xue, Xiao Li, Nianmin Zhang
Abstract: In advanced manufacturing systems, humans and robots collaborate to conduct the production process. Effective task planning and allocation (TPA) is crucial for achieving high production efficiency, yet it remains challenging in complex and dynamic manufacturing environments. The dynamic nature of humans and robots, particularly the need to consider spatial information (e.g., humans' real-time position and the distance they need to move to complete a task), substantially complicates TPA. To address the above challenges, we decompose production tasks into manageable subtasks. We then implement a real-time hierarchical human-robot TPA algorithm, including a high-level agent for task planning and a low-level agent for task allocation. For the high-level agent, we propose an efficient buffer-based deep Q-learning method (EBQ), which reduces training time and enhances performance in production problems with long-term and sparse reward challenges. For the low-level agent, a path planning-based spatially aware method (SAP) is designed to allocate tasks to the appropriate human-robot resources, thereby achieving the corresponding sequential subtasks. We conducted experiments on a complex real-time production process in a 3D simulator. The results demonstrate that our proposed EBQ&SAP method effectively addresses human-robot TPA problems in complex and dynamic production processes.
Authors: Serdar Kadioglu, Karthik Uppuluri, Akash Singirikonda
Abstract: There is growing interest in leveraging large language models (LLMs) for text-to-model translation and optimization tasks. This paper aims to advance this line of research by introducing \textsc{Text2Model} and \textsc{Text2Zinc}. \textsc{Text2Model} is a suite of copilots based on several LLM strategies with varying complexity, along with an online leaderboard. \textsc{Text2Zinc} is a cross-domain dataset for capturing optimization and satisfaction problems specified in natural language, along with an interactive editor with built-in AI assistant. While there is an emerging literature on using LLMs for translating combinatorial problems into formal models, our work is the first attempt to integrate \textit{both} satisfaction and optimization problems within a \textit{unified architecture} and \textit{dataset}. Moreover, our approach is \textit{solver-agnostic} unlike existing work that focuses on translation to a solver-specific model. To achieve this, we leverage \textsc{MiniZinc}'s solver-and-paradigm-agnostic modeling capabilities to formulate combinatorial problems. We conduct comprehensive experiments to compare execution and solution accuracy across several single- and multi-call strategies, including; zero-shot prompting, chain-of-thought reasoning, intermediate representations via knowledge-graphs, grammar-based syntax encoding, and agentic approaches that decompose the model into sequential sub-tasks. Our copilot strategies are competitive, and in parts improve, recent research in this domain. Our findings indicate that while LLMs are promising they are not yet a push-button technology for combinatorial modeling. We contribute \textsc{Text2Model} copilots and leaderboard, and \textsc{Text2Zinc} and interactive editor to open-source to support closing this performance gap.
Authors: Stefan Kambiz Behfar, Jon Crowcroft
Abstract: Web3 systems expose a fundamentally different security landscape from centralized platforms, characterized by composability, pseudonymous identities, decentralized governance, and rapidly evolving attack strategies that span social, application, and protocol layers. Existing security mechanisms, such as static smart contract analysis, blacklist-based phishing detection, and network-level mitigation, operate in isolation and assume fixed threat models, limiting their effectiveness against adaptive, cross-layer adversaries. This position paper argues that securing Web3 requires a shift from static, tool-centric defenses to learning-driven security primitives capable of continuous reasoning, adaptation, and actuation. We introduce AI-powered smart certificates as a new security abstraction: programmable, continuously updated trust artifacts that integrate on-chain verifiability with off-chain machine learning signals derived from user behavior, transaction dynamics, and social context. Unlike traditional certificates or audits, these certificates maintain state, learn under distribution shift, and support automated policy enforcement and revocation in response to evolving threats. We argue that existing paradigms, formal verification, threat modeling, and isolated anomaly detection, are structurally limited in capturing the non-stationary and socio-technical nature of Web3 attacks. We outline an architecture in which AI-powered smart certificates serve as cross-layer sentinels that coordinate heterogeneous security signals in real time, and position smart certificates as a research direction, raising questions around learning under partial observability, adversarial adaptation, and trustworthy ML deployment in decentralized systems.
Authors: Martin Kuo, Jianyi Zhang, Dongting Li, Yiran Chen
Abstract: Pretraining language models is still a challenge for many researchers due to its substantial computational costs. As such, there is growing interest in developing more affordable pretraining methods. One notable advancement in this area is the Cramming technique (Geiping and Goldstein, 2022), which enables the pretraining of BERT-style language models using just one GPU in a single day. Building on this innovative approach, we introduce the Dependency Agreement Cramming (DA-Cramming), an efficient framework that integrates information about dependency agreements into the pretraining process. Unlike existing methods that leverage similar semantic information during finetuning, our approach represents a pioneering effort focusing on enhancing the foundational language understanding with semantic information during pretraining. We meticulously design a dual-stage pretraining work flow with four dedicated submodels to capture representative dependency agreements at the chunk level, effectively transforming these agreements into embeddings to benefit the pretraining. Extensive empirical results demonstrate that our method significantly outperforms previous methods across various tasks.
Authors: Bo Shu, Yiting Zhang, Saisai Hu, Dong Shu
Abstract: This report investigates the history and impact of Generative Models and Connected and Automated Vehicles (CAVs), two groundbreaking forces pushing progress in technology and transportation. By focusing on the application of generative models within the context of CAVs, the study aims to unravel how this integration could enhance predictive modeling, simulation accuracy, and decision-making processes in autonomous vehicles. This thesis discusses the benefits and challenges of integrating generative models and CAV technology in transportation. It aims to highlight the progress made, the remaining obstacles, and the potential for advancements in safety and innovation.
Authors: Bongsoo Yi, Rongjie Lai, Yao Li
Abstract: Adversarial training has proven effective in improving the robustness of deep neural networks against adversarial attacks. However, this enhanced robustness often comes at the cost of a substantial drop in accuracy on clean data. In this paper, we address this limitation by introducing Tangent Direction Guided Adversarial Training (TART), a novel method that enhances clean accuracy by exploiting the geometry of the data manifold. We argue that adversarial examples with large components in the normal direction can overly distort the decision boundary and degrade clean accuracy. TART addresses this issue by estimating the tangent direction of adversarial examples and adaptively modulating the perturbation bound based on the norm of their tangential component. To the best of our knowledge, TART is the first adversarial defense framework that explicitly incorporates the concept of tangent space and direction into adversarial training. Extensive experiments on both synthetic and benchmark datasets demonstrate that TART consistently improves clean accuracy while maintaining robustness against adversarial attacks.
Authors: Jente Vandersanden, Sascha Holl, Xingchang Huang, Gurprit Singh
Abstract: Classical diffusion models typically rely on isotropic Gaussian noise, treating all regions uniformly and overlooking structural information important for high-quality generation. We introduce an edge-preserving diffusion process that generalizes isotropic models via a hybrid noise scheme with an edge-aware scheduler that smoothly transitions from edge-preserving to isotropic noise. This enables the model to capture fine structural details while generally maintaining global performance. We evaluate the impact of structure-aware noise in both diffusion and flow-matching frameworks, and show that existing isotropic models can be efficiently fine-tuned with edge-preserving noise, making our framework practical for adapting pre-trained systems. Beyond unconditional generation, our method particularly shows improvements in structure-guided tasks such as stroke-to-image synthesis, improving robustness and perceptual quality, as evidenced by consistent improvements across FID, KID, and CLIP-score.
Authors: Samiha Sharlin, Tyler R. Josephson
Abstract: Large Language Models (LLMs) are transformer-based machine learning models that have shown remarkable performance in tasks for which they were not explicitly trained. Here, we explore the potential of LLMs to perform symbolic regression -- a machine-learning method for finding simple and accurate equations from datasets. We prompt GPT-4 and GPT-4o models to suggest expressions from data, which are then optimized and evaluated using external Python tools. These results are fed back to the LLMs, which propose improved expressions while optimizing for complexity and loss. Using chain-of-thought prompting, we instruct the models to analyze data, prior expressions, and the scientific context (expressed in natural language) for each problem before generating new expressions. We evaluated the workflow in rediscovery of Langmuir and dual-site Langmuir's model for adsorption, along with Nikuradse's dataset on flow in rough pipes, which does not have a known target model equation. Both the GPT-4 and GPT-4o models successfully rediscovered equations, with better performance when using a scratchpad and considering scientific context. GPT-4o model demonstrated improved reasoning with data patterns, particularly evident in the dual-site Langmuir and Nikuradse dataset. We demonstrate how strategic prompting improves the model's performance and how the natural language interface simplifies integrating theory with data. We also applied symbolic mathematical constraints based on the background knowledge of data via prompts and found that LLMs generate meaningful equations more frequently. Although this approach does not outperform established SR programs where target equations are more complex, LLMs can nonetheless iterate toward improved solutions while following instructions and incorporating scientific context in natural language.
Authors: Yuwei Yin, Giuseppe Carenini
Abstract: Intent, a critical cognitive notion and mental state, is ubiquitous in human communication and problem-solving. Accurately understanding the underlying intent behind questions is imperative to reasoning towards correct answers. However, this significant concept has been largely disregarded in the rapid development of language models (LMs). To unleash the potential of intent and instill it into LMs, this paper introduces Intentional Analysis (IA), which explicitly invokes intent-aware analysis and reasoning during the problem-solving process. Comprehensive experiments across diverse benchmarks, model types, and configurations demonstrate the effectiveness, robustness, and generalizability of IA. Notably, IA consistently improves task performance even on SOTA proprietary models like GPT-5 and Claude-Opus-4.6. Moreover, IA not only outperforms Chain-of-Thought (CoT) across various experimental settings, but it can also synergistically work with CoT reasoning. Further qualitative analysis and case studies reveal that the benefits of IA stem from addressing several weaknesses in baseline methods, such as intent misunderstanding, hasty generalization, and mental laziness. Case studies also provide insights into the mechanisms underlying IA and clarify how it differs from CoT in mitigating these weaknesses. This study sheds light on a promising direction for the development of future LLMs with intentional analysis.
Authors: Ido Galil, Moshe Kimhi, Ran El-Yaniv
Abstract: Deep Neural Networks (DNNs) can be catastrophically disrupted by flipping only a handful of parameter bits. We introduce Deep Neural Lesion (DNL), a data-free and optimizationfree method that locates critical parameters, and an enhanced single-pass variant, 1P-DNL, that refines this selection with one forward and backward pass on random inputs. We show that this vulnerability spans multiple domains, including image classification, object detection, instance segmentation, and reasoning large language models. In image classification, flipping just two sign bits in ResNet-50 on ImageNet reduces accuracy by 99.8%. In object detection and instance segmentation, one or two sign flips in the backbone collapse COCO detection and mask AP for Mask R-CNN and YOLOv8-seg models. In language modeling, two sign flips into different experts reduce Qwen3-30B-A3B-Thinking from 78% to 0% accuracy. We also show that selectively protecting a small fraction of vulnerable sign bits provides a practical defense against such attacks.
Authors: Andrea Apicella, Salvatore Giugliano, Francesco Isgr\`o, Andrea Pollastro, Roberto Prevete
Abstract: The eXplainable Artificial Intelligence (XAI) research predominantly concentrates to provide explainations about AI model decisions, especially Deep Learning (DL) models. However, there is a growing interest in using XAI techniques to automatically improve the performance of the AI systems themselves. This paper proposes IMPACTX, a novel approach that leverages XAI as a fully automated attention mechanism, without requiring external knowledge or human feedback. Experimental results show that IMPACTX has improved performance respect to the standalone ML model by integrating an attention mechanism based an XAI method outputs during the model training. Furthermore, IMPACTX directly provides proper feature attribution maps for the model's decisions, without relying on external XAI methods during the inference process. Our proposal is evaluated using three widely recognized DL models (EfficientNet-B2, MobileNet, and LeNet-5) along with three standard image datasets: CIFAR-10, CIFAR-100, and STL-10. The results show that IMPACTX consistently improves the performance of all the inspected DL models across all evaluated datasets, and it directly provides appropriate explanations for its responses.
Authors: Hiba Arnaout, Noy Sternlicht, Tom Hope, Iryna Gurevych
Abstract: Understanding the impact of scientific publications is crucial for identifying breakthroughs and guiding future research. Traditional metrics based on citation counts often miss the nuanced ways a paper contributes to its field. In this work, we propose a new task: generating nuanced, expressive, and time-aware impact summaries that capture both praise (confirmation citations) and critique (correction citations) through the evolution of fine-grained citation intents. We introduce an evaluation framework tailored to this task, showing moderate to strong human correlation on subjective metrics such as insightfulness. Expert feedback from professors reveals a strong interest in these summaries and suggests future improvements. Data and code are made available.
Authors: Ido Guy, Daniel Haimovich, Fridolin Linder, Nastaran Okati, Lorenzo Perini, Niek Tax, Mark Tygert
Abstract: A suitable scalar metric can help measure multi-calibration, defined as follows. When the expected values of observed responses are equal to corresponding predicted probabilities, the probabilistic predictions are known as "perfectly calibrated." When the predicted probabilities are perfectly calibrated simultaneously across several subpopulations, the probabilistic predictions are known as "perfectly multi-calibrated." In practice, predicted probabilities are seldom perfectly multi-calibrated, so a statistic measuring the distance from perfect multi-calibration is informative. A recently proposed metric for calibration, based on the classical Kuiper statistic, is a natural basis for a new metric of multi-calibration and avoids well-known problems of metrics based on binning or kernel density estimation. The newly proposed metric weights the contributions of different subpopulations in proportion to their signal-to-noise ratios; data analyses' ablations demonstrate that the metric becomes noisy when omitting the signal-to-noise ratios from the metric. Numerical examples on benchmark data sets illustrate the new metric.
Authors: Yixian Xu, Shengjie Luo, Liwei Wang, Di He, Chang Liu
Abstract: Diffusion models have achieved remarkable success in generative modeling. Despite more stable training, the loss of diffusion models is not indicative of absolute data-fitting quality, since its optimal value is typically not zero but unknown, leading to confusion between large optimal loss and insufficient model capacity. In this work, we advocate the need to estimate the optimal loss value for diagnosing and improving diffusion models. We first derive the optimal loss in closed form under a unified formulation of diffusion models, and develop effective estimators for it, including a stochastic variant scalable to large datasets with proper control of variance and bias. With this tool, we unlock the inherent metric for diagnosing the training quality of mainstream diffusion model variants, and develop a more performant training schedule based on the optimal loss. Moreover, using models with 120M to 1.5B parameters, we find that the power law is better demonstrated after subtracting the optimal loss from the actual training loss, suggesting a more principled setting for investigating the scaling law for diffusion models.
Authors: Hongyi Pan, Ziliang Hong, Gorkem Durak, Ziyue Xu, Ulas Bagci
Abstract: Federated learning enables collaborative training of deep learning models across institutions without sharing sensitive patient data. However, its performance is often limited by small datasets and non-independent, identically distributed data, which can impair model generalization. In this work, we propose a generative model-based data augmentation framework for breast ultrasound classification. It leverages synthetic images generated by deep convolutional generative adversarial networks and a class-conditioned denoising diffusion probabilistic model. Experiments on three publicly available datasets (BUSI, BUS-BRA, and UDIAT) demonstrated that incorporating a suitable number of synthetic images improved average AUC from 0.9206 to 0.9362 for FedAvg and from 0.9429 to 0.9574 for FedProx. Furthermore, we noticed that excessive use of synthetic data reduced performance. This highlights the importance of balancing real and synthetic samples. Our results underscore the potential of generative model-based augmentation to enhance federated breast ultrasound image classification.
Authors: Fardin Saad, Pradeep K. Murukannaiah, Munindar P. Singh
Abstract: Successful human-agent teaming relies on an agent being able to understand instructions given by a (human) principal. In many cases, an instruction may be incomplete or ambiguous. In such cases, the agent must infer the unspoken intentions from their shared context, that is, it must exercise the principal's Theory of Mind (ToM) and infer the mental states of its principal. We consider the prospects of effective human-agent collaboration using large language models (LLMs). To assess ToM in a dynamic, goal-oriented, and collaborative environment, we introduce a novel task, Instruction Inference, in which an agent assists a principal in reaching a goal by interpreting incomplete or ambiguous instructions. We present Tomcat, an LLM-based agent, designed to exhibit ToM reasoning in interpreting and responding to the principal's instructions.We implemented two variants of Tomcat. One, dubbed Fs-CoT (Fs for few-shot, CoT for chain-of-thought), is based on a small number of examples demonstrating the requisite structured reasoning. One, dubbed CP (commonsense prompt), relies on commonsense knowledge and information about the problem. We realized both variants of Tomcat on three leading LLMs, namely, GPT-4o, DeepSeek-R1, and Gemma-3-27B. To evaluate the effectiveness of Tomcat, we conducted a study with 52 human participants in which we provided participants with the same information as the CP variant. We computed intent accuracy, action optimality, and planning optimality to measure the ToM capabilities of Tomcat and our study participants. We found that Tomcat with Fs-CoT, particularly with GPT-4o and DeepSeek-R1, achieves performance comparable to the human participants, underscoring its ToM potential for human-agent collaboration.
Authors: Yiyuan Yang, Zichuan Liu, Lei Song, Kai Ying, Zhiguang Wang, Tom Bamford, Svitlana Vyetrenko, Jiang Bian, Qingsong Wen
Abstract: Time series anomaly detection (TSAD) has traditionally focused on binary classification and often lacks the fine-grained categorization and explanatory reasoning required for transparent decision-making. To address these limitations, we propose Time-series Reasoning for Anomaly (Time-RA), a novel task that reformulates TSAD from a discriminative into a generative, reasoning-intensive paradigm. To facilitate this, we introduce RATs40K, the first real-world large-scale multimodal benchmark with ~40,000 samples across 10 domains, integrating raw time series, textual context, and visual plots with structured reasoning annotations. Extensive benchmarking shows that while supervised fine-tuning and visual representations boost diagnostic accuracy and reasoning consistency, performance varies across complex scenarios. Notably, fine-tuned models demonstrate strong "plug-and-play" transferability, outperforming traditional baselines on unseen real-world datasets. Our work establishes a foundation for interpretable, multimodal time series analysis. All code (https://github.com/yyysjz1997/Time-RA) and the RATs40K dataset (https://huggingface.co/datasets/Time-RA/RATs40K) are fully open-sourced to facilitate future research.
URLs: https://github.com/yyysjz1997/Time-RA), https://huggingface.co/datasets/Time-RA/RATs40K)
Authors: Xinwei Wu, Haojie Li, Hongyu Liu, Xinyu Ji, Ruohan Li, Yule Chen, Yigeng Zhang
Abstract: In this work, we study a critical research problem regarding the trustworthiness of large language models (LLMs): how LLMs behave when encountering ambiguous narrative text, with a particular focus on Chinese textual ambiguity. We created a benchmark dataset by collecting and generating ambiguous sentences with context and their corresponding disambiguated pairs, representing multiple possible interpretations. These annotated examples are systematically categorized into 3 main categories and 9 subcategories. Through experiments, we discovered significant fragility in LLMs when handling ambiguity, revealing behavior that differs substantially from humans. Specifically, LLMs cannot reliably distinguish ambiguous text from unambiguous text, show overconfidence in interpreting ambiguous text as having a single meaning rather than multiple meanings, and exhibit overthinking when attempting to understand the various possible meanings. Our findings highlight a fundamental limitation in current LLMs that has significant implications for their deployment in real-world applications where linguistic ambiguity is common, calling for improved approaches to handle uncertainty in language understanding. The dataset and code are publicly available at this GitHub repository: https://github.com/ictup/LLM-Chinese-Textual-Disambiguation.
URLs: https://github.com/ictup/LLM-Chinese-Textual-Disambiguation.
Authors: Dai Do, Manh Nguyen, Svetha Venkatesh, Hung Le
Abstract: Large language models (LLMs) have shown strong reasoning capabilities when fine-tuned with reinforcement learning (RL). However, such methods require extensive data and compute, making them impractical under many realistic training budgets. Many existing pipelines sample training examples uniformly across steps or epochs, ignoring differences in difficulty, redundancy, and learning value, which slows learning and wastes computation. We propose \textbf{SPaCe}, a self-paced learning framework that enables efficient learning based on the capability of the model being trained through optimizing which data to use and when. First, we apply \emph{cluster-based data reduction} to partition training data by semantics and difficulty, extracting a compact yet diverse subset that reduces redundancy. Then, a \textit{multi-armed bandit} treats data clusters as arms, allocating training samples based on the model's solve rates and learning progress. Experiments across multiple reasoning benchmarks show that SPaCe achieves comparable or better accuracy than state-of-the-art baselines while using up to \(100\times\) fewer samples. Ablation studies and analyses further highlight the importance of both data clustering and adaptive selection. Our results demonstrate that carefully curated, performance-driven training curricula can unlock strong reasoning abilities in LLMs with minimal resources.
Authors: Shaocong Wang, Tong Liu, Yihan Li, Ming Li, Kairui Wen, Pei Yang, Wenqi Ji, Minjing Yu, Yong-Jin Liu
Abstract: Recent advances in self-supervised learning for EEG representation have largely relied on masked reconstruction, where models are trained to recover randomly masked signal segments. While effective at modeling local dependencies, such objectives are inherently limited in capturing the global dynamics and long-range dependencies essential for characterizing neural activity. To address this limitation, we propose EEGDM, a novel self-supervised framework that leverages latent diffusion models to generate EEG signals as an objective. Unlike masked reconstruction, diffusion-based generation progressively denoises signals from noise to realism, compelling the model to capture holistic temporal patterns and cross-channel relationships. Specifically, EEGDM incorporates an EEG encoder that distills raw signals and their channel augmentations into a compact representation, acting as conditional information to guide the diffusion model for generating EEG signals. This design endows EEGDM with a compact latent space, which not only offers ample control over the generative process but also can be leveraged for downstream tasks. Experimental results show that EEGDM (1) reconstructs high-quality EEG signals, (2) learns robust representations, and (3) achieves competitive performance across diverse downstream tasks, thus exploring a new direction for self-supervised EEG representation learning.
Authors: Diego Di Carlo (RIKEN AIP), Shoichi Koyama (UTokyo), Nugraha Aditya Arie (RIKEN AIP), Fontaine Mathieu (S2A, IDS), Bando Yoshiaki (AIST), Yoshii Kazuyoshi (RIKEN AIP)
Abstract: This paper investigates continuous representations of steering vectors over frequency and microphone/source positions for augmented listening (e.g., spatial filtering and binaural rendering), enabling user-parameterized control of the reproduced sound field. Steering vectors have typically been used for representing the spatial response of a microphone array as a function of the look-up direction. The basic algebraic representation of these quantities assuming an idealized environment cannot deal with the scattering effect of the sound field. One may thus collect a discrete set of real steering vectors measured in dedicated facilities and super-resolve (i.e., upsample) them. Recently, physics-aware deep learning methods have been effectively used for this purpose. Such deterministic super-resolution, however, suffers from the overfitting problem due to the non-uniform uncertainty over the measurement space. To solve this problem, we integrate an expressive representation based on the neural field (NF) into the principled probabilistic framework based on the Gaussian process (GP). Specifically, we propose a physics-aware composite kernel that models the directional incoming waves and the subsequent scattering effect. Our comprehensive comparative experiment showed the effectiveness of the proposed method under data insufficiency conditions. In downstream tasks such as speech enhancement and binaural rendering using the simulated data of the SPEAR challenge, the oracle performances were attained with less than ten times fewer measurements.
Authors: Yubo Gao, Renbo Tu, Gennady Pekhimenko, Nandita Vijaykumar
Abstract: Differentially-Private SGD (DP-SGD) and its adaptive variant DP-Adam are powerful techniques to protect user privacy when using sensitive data to train neural networks. During training, converting model weights and activations into low-precision formats, i.e., quantization, can drastically reduce training times, energy consumption, and cost, and is thus a widely used technique. In this work, we demonstrate for the first time that quantization causes significantly higher accuracy degradation in DP training compared to regular SGD. We observe that this is caused by noise injection, which amplifies quantization variance, leading to disproportionately large accuracy degradation. To address this challenge, we present DPQuant, a dynamic quantization framework that adaptively selects a changing subset of layers to quantize at each epoch. Our method combines two key ideas that effectively reduce quantization variance: (i) probabilistic sampling that rotates which layers are quantized every epoch, and (ii) loss-aware layer prioritization, which uses a differentially private loss sensitivity estimator to identify layers that can be quantized with minimal impact on model quality. This estimator consumes a negligible fraction of the overall privacy budget, preserving DP guarantees. Empirical evaluations on ResNet18, ResNet50, and DenseNet121 across a range of datasets demonstrate that DPQuant consistently outperforms static quantization baselines, achieving near Pareto-optimal accuracy-compute trade-offs and up to $2.21\times$ theoretical throughput improvements on low-precision hardware, with less than 2% drop in validation accuracy. We further show that our framework extends to DP-Adam with similar gains.
Authors: Liting Gao, Yi Yuan, Yaru Chen, Yuelan Cheng, Zhenbo Li, Juan Wen, Shubin Zhang, Wenwu Wang
Abstract: Diffusion models have shown remarkable progress in text-to-audio generation. However, text-guided audio editing remains in its early stages. This task focuses on modifying the target content within an audio signal while preserving the rest, thus demanding precise localization and faithful editing according to the text prompt. Existing training-based and zero-shot methods that rely on full-caption or costly optimization often struggle with complex editing or lack practicality. In this work, we propose a novel end-to-end efficient rectified flow matching-based diffusion framework for audio editing, and construct a dataset featuring overlapping multi-event audio to support training and benchmarking in complex scenarios. Experiments show that our model achieves faithful semantic alignment without requiring auxiliary captions or masks, while maintaining competitive editing quality across metrics.
Authors: Ivan Ternovtsii, Yurii Bilak
Abstract: Mixture-of-Experts (MoE) models improve efficiency through sparse activation, but their learned gating functions provide limited insight into routing decisions. This work introduces the Semantic Resonance Architecture (SRA), which routes tokens to experts via cosine similarity between token representations and learnable semantic anchors, making every routing decision directly traceable to anchor-token similarity scores. We evaluate SRA on WikiText-103 across 17 configurations. In a controlled multi-seed comparison (3 seeds x 4 configurations, 256 experts, $D_{ff}=256$), cosine routing achieves competitive perplexity with standard linear routing ($12.57 \pm 0.03$ vs $12.45 \pm 0.03$ for $K=1 \to 4$; $12.52 \pm 0.02$ vs $12.57 \pm 0.02$ for $K=2 \to 4$). The training recipe -- not the routing function -- drives specialization quality, while cosine routing provides inherent inspectability. We introduce a bandpass routing loss -- a floor-and-ceiling corridor on expert utilization -- that reduces dead experts from 30-45% to 0-6% and transfers to both routing types. Routing-space evaluation shows cosine routing provides significantly better word-level subtoken coherence in deeper layers ($p < 0.001$), with 44-54% of expert specialization being syntactic rather than semantic. Extended analysis reveals cosine routing maintains more stable router saturation and tighter per-expert vocabulary distributions -- structural advantages from the bounded cosine similarity range. An inference-time $k$-sweep shows that $k=5$ yields a free 0.08-0.16 perplexity gain over $k=4$. Cross-dataset validation on OpenWebText confirms generalization: cosine routing achieves comparable perplexity (44.88 vs 45.44), the bandpass loss eliminates dead experts, and specialization patterns are preserved.
Authors: Armin Karamzade, Kyungmin Kim, JB Lanier, Davide Corsi, Roy Fox
Abstract: Delays frequently occur in real-world environments, yet standard reinforcement learning (RL) algorithms often assume instantaneous perception of the environment. We study random sensor delays in POMDPs, where observations may arrive out-of-sequence, a setting that has not been previously addressed in RL. We analyze the structure of such delays and demonstrate that naive approaches, such as stacking past observations, are insufficient for reliable performance. To address this, we propose a model-based filtering process that sequentially updates the belief state based on an incoming stream of observations. We then introduce a simple delay-aware framework that incorporates this idea into model-based RL, enabling agents to effectively handle random delays. Applying this framework to the Dreamer world-modeling scheme, our method consistently outperforms delay-aware baselines developed for MDPs and demonstrates robustness to delay distribution shifts during deployment. Additionally, we present experiments on simulated robotic tasks, comparing our method to common practical heuristics and emphasizing the importance of explicitly modeling observation delays.
Authors: Zijian Zhao, Dian Jin, Zijing Zhou
Abstract: Recently, Image-to-Music (I2M) generation has garnered significant attention, with potential applications in fields such as gaming, advertising, and multi-modal art creation. However, due to the ambiguous and subjective nature of I2M tasks, most end-to-end methods lack interpretability, leaving users puzzled about the generation results. Even methods based on emotion mapping face controversy, as emotion represents only a singular aspect of art. Additionally, most learning-based methods require substantial computational resources and large datasets for training, hindering accessibility for common users. To address these challenges, we propose the first Vision Language Model (VLM)-based I2M framework that offers high interpretability and low computational cost. Specifically, we utilize ABC notation to bridge the text and music modalities, enabling the VLM to generate music using natural language. We then apply multi-modal Retrieval-Augmented Generation (RAG) and self-refinement techniques to allow the VLM to produce high-quality music without external training. Furthermore, we leverage the generated motivations in text and the attention maps from the VLM to provide explanations for the generated results in both text and image modalities. To validate our method, we conduct both human studies and machine evaluations, where our method outperforms others in terms of music quality and music-image consistency, indicating promising results. Our code is available at https://github.com/RS2002/Image2Music .
Authors: Haonan Chen, Jiaming Xu, Hongyu Chen, Kaiwen Hong, Binghao Huang, Chaoqi Liu, Jiayuan Mao, Yunzhu Li, Yilun Du, Katherine Driggs-Campbell
Abstract: Effectively integrating diverse sensory modalities is crucial for robotic manipulation. However, the typical approach of feature concatenation is often suboptimal: dominant modalities such as vision can overwhelm sparse but critical signals like touch in contact-rich tasks, and monolithic architectures cannot flexibly incorporate new or missing modalities without retraining. Our method factorizes the policy into a set of diffusion models, each specialized for a single representation (e.g., vision or touch), and employs a router network that learns consensus weights to adaptively combine their contributions, enabling incremental of new representations. We evaluate our approach on simulated manipulation tasks in {RLBench}, as well as real-world tasks such as occluded object picking, in-hand spoon reorientation, and puzzle insertion, where it significantly outperforms feature-concatenation baselines on scenarios requiring multimodal reasoning. Our policy further demonstrates robustness to physical perturbations and sensor corruption. We further conduct perturbation-based importance analysis, which reveals adaptive shifts between modalities.
Authors: Eleonora Ristori, Luca Bindini, Paolo Frasconi
Abstract: Research on audio generation has progressively developed along both waveform-based and spectrogram-based directions, giving rise to diverse strategies for representing and generating audio. At the same time, advances in image synthesis have shown that autoregression across scales, rather than tokens, improves coherence and detail. Building on these ideas, we introduce MARS (Multi-channel AutoRegression on Spectrograms), which, to the best of our knowledge, is the first adaptation of next-scale autoregressive modeling to the spectrogram domain. MARS treats spectrograms as multi-channel images and employs channel multiplexing (CMX), a reshaping strategy that reduces spatial resolution without information loss. A shared tokenizer provides consistent discrete representations across scales, enabling a transformer-based autoregressor to refine spectrograms from coarse to fine resolutions efficiently. Experiments on a large-scale dataset demonstrate that MARS performs comparably or better than state-of-the-art baselines across multiple evaluation metrics, establishing an efficient and scalable paradigm for high-fidelity sound generation.
Authors: Anukriti Singh, Kasra Torshizi, Khuzema Habib, Kelin Yu, Ruohan Gao, Pratap Tokekar
Abstract: Vision-based robot learning often relies on dense image or point-cloud inputs, which are computationally heavy and entangle irrelevant background features. Existing keypoint-based approaches can focus on manipulation-centric features and be lightweight, but either depend on manual heuristics or task-coupled selection, limiting scalability and semantic understanding. To address this, we propose AFFORD2ACT, an affordance-guided framework that distills a minimal set of semantic 2D keypoints from a text prompt and a single image. AFFORD2ACT follows a three-stage pipeline: affordance filtering, category-level keypoint construction, and transformer-based policy learning with embedded gating to reason about the most relevant keypoints, yielding a compact 38-dimensional state policy that can be trained in 15 minutes, which performs well in real-time without proprioception or dense representations. Across diverse real-world manipulation tasks, AFFORD2ACT consistently improves data efficiency, achieving an 82% success rate on unseen objects, novel categories, backgrounds, and distractors.
Authors: Aleksi Huotala, Miikka Kuutila, Olli-Pekka Turtio, Simo Sipil\"a, Mika M\"antyl\"a
Abstract: Conducting systematic reviews is laborious. In the screening or study selection phase, the number of papers can be overwhelming. Recent research has demonstrated that large language models (LLMs) can perform title-abstract screening and support humans in the task. To this end, we developed AISysRev, an LLM-based screening tool implemented as a containerized web application. The tool accepts CSV files containing paper titles and abstracts. Users specify inclusion and exclusion criteria. Multiple different LLMs can be used, such as Gemini, Claude, Mistral or ChatGPT via OpenRouter. We also support locally hosted models and any model compatible with the OpenAI SDK. AISysRev implements both zero-shot and few-shot prompting, and also allows for manual screening through interfaces that display LLM results as guidance for human reviewers. LLM calls are parallelized, meaning screening speed is typically between 100 to 300 papers per minute, depending on the model and the host. To demonstrate the tool's use in practice, we conducted a qualitative trial study with 137 papers using the tool. Our findings indicate that papers can be classified into four categories: Easy Includes, Easy Excludes, Boundary Includes, and Boundary Excludes. The Boundary cases, where LLMs are prone to errors, highlight the need for human intervention. While LLMs do not replace human judgment in systematic reviews, they can reduce the burden of assessing large volumes of scientific literature. Video: https://www.youtube.com/watch?v=HeblemlgnAQ Tool: https://github.com/EvoTestOps/AISysRev
URLs: https://www.youtube.com/watch?v=HeblemlgnAQ, https://github.com/EvoTestOps/AISysRev
Authors: Shangqing Tu, Yaxuan Li, Yushi Bai, Lei Hou, Juanzi Li
Abstract: Parallel scaling has emerged as a powerful paradigm to enhance reasoning capabilities in large language models (LLMs) by generating multiple Chain-of-Thought (CoT) traces simultaneously. However, this approach introduces significant computational inefficiency due to inter-trace redundancy -- our analysis reveals that over 80% of parallel reasoning traces yield identical final answers, representing substantial wasted computation. To address this critical efficiency bottleneck, we propose DeepPrune, a novel framework that enables efficient parallel scaling through dynamic pruning. Our method features a specialized judge model trained with out-of-distribution data (AIME 2022, AIME 2023, and MATH 500) using oversampling techniques to accurately predict answer equivalence from partial reasoning traces, achieving 0.7072 AUROC on unseen reasoning models. Combined with an online greedy clustering algorithm that dynamically prunes redundant paths while preserving answer diversity. Comprehensive evaluations across three challenging benchmarks (AIME 2024, AIME 2025, and GPQA) and multiple reasoning models demonstrate that DeepPrune achieves remarkable token reduction of 65.73%--88.50% compared to conventional consensus sampling, while maintaining competitive accuracy within 3 percentage points. Our work establishes a new standard for efficient parallel reasoning, making high-performance reasoning more efficient. Our code and data are here: https://deepprune.github.io/.
Authors: Shinwoo Park, Hyejin Park, Hyeseon An, Yo-Sub Han
Abstract: As large language models (LLMs) continue to advance rapidly, reliable governance tools have become critical. Publicly verifiable watermarking is particularly essential for fostering a trustworthy AI ecosystem. A central challenge persists: balancing text quality against detection robustness. Recent studies have sought to navigate this trade-off by leveraging signals from model output distributions (e.g., token-level entropy); however, their reliance on these model-specific signals presents a significant barrier to public verification, as the detection process requires access to the logits of the underlying model. We introduce STELA, a novel framework that aligns watermark strength with the linguistic degrees of freedom inherent in language. STELA dynamically modulates the signal using part-of-speech (POS) n-gram-modeled linguistic indeterminacy, weakening it in grammatically constrained contexts to preserve quality and strengthening it in contexts with greater linguistic flexibility to enhance detectability. Our detector operates without access to any model logits, thus facilitating publicly verifiable detection. Through extensive experiments on typologically diverse languages-analytic English, isolating Chinese, and agglutinative Korean-we show that STELA surpasses prior methods in detection robustness. Our code is available at https://github.com/Shinwoo-Park/stela_watermark.
Authors: Jingyao Liu, Chen Huang, Zhizhao Guan, Wenqiang Lei, Yang Deng
Abstract: The rapid advancement in large language models (LLMs) has demonstrated significant potential in End-to-End Software Development (E2ESD). However, existing E2ESD benchmarks are limited by coarse-grained requirement specifications and unreliable evaluation protocols, hindering a true understanding of current framework capabilities. To address these limitations, we present E2EDev, a novel benchmark grounded in the principles of Behavior-Driven Development (BDD), which evaluates the capabilities of E2ESD frameworks by assessing whether the generated software meets user needs through mimicking real user interactions (Figure 1). E2EDev comprises (i) a fine-grained set of user requirements, (ii) multiple BDD test scenarios with corresponding Python step implementations for each requirement, and (iii) a fully automated testing pipeline built on the Behave framework. To ensure its quality while reducing the annotation effort, E2EDev leverages our proposed Human-in-the-Loop Multi-Agent Annotation Framework (HITL-MAA). By evaluating various E2ESD frameworks and LLM backbones with E2EDev, our analysis reveals a persistent struggle to effectively solve these tasks, underscoring the critical need for more effective and cost-efficient E2ESD solutions. Our codebase and benchmark are publicly available at https://github.com/SCUNLP/E2EDev.
Authors: Wenshuo Wang, Ziyou Jiang, Junjie Wang, Mingyang Li, Jie Huang, Yuekai Huang, Zhiyuan Chang, Feiyan Duan, Qing Wang
Abstract: Internet memes have emerged as a popular multimodal medium, yet they are increasingly weaponized to convey harmful opinions through subtle rhetorical devices like irony and metaphor. Existing detection approaches, including Multimodal Large Language Model (MLLM)-based techniques, struggle with these implicit expressions, leading to frequent misjudgments. This paper introduces PatMD, a novel approach that detects harmful memes by learning from and proactively mitigating these potential misjudgment risks. Our core idea is to move beyond superficial content-level matching and instead identify the underlying misjudgment risk patterns, proactively guiding the MLLMs to avoid known misjudgment pitfalls. We first construct a knowledge base where each meme is deconstructed into a misjudgment risk pattern explaining why it might be misjudged, either overlooking harmful undertones (false negative) or overinterpreting benign content (false positive). For a given target meme, PatMD retrieves relevant patterns and utilizes them to dynamically guide the MLLM's reasoning. Experiments on a benchmark of 6,626 memes across 5 harmful detection tasks show that PatMD outperforms state-of-the-art baselines, achieving an average of 8.30% improvement in F1-score and 7.71% improvement in accuracy, while exhibiting consistent robustness on unseen and adversarial memes.
Authors: Jiahao Tang, Henry Hengyuan Zhao, Lijian Wu, Yifei Tao, Dongxing Mao, Yang Wan, Jingru Tan, Min Zeng, Min Li, Alex Jinpeng Wang
Abstract: We introduce Chart2Code, a new benchmark for evaluating the chart understanding and code generation capabilities of large multimodal models (LMMs). Chart2Code is explicitly designed from a user-driven perspective, capturing diverse real-world scenarios and progressively increasing task difficulty. It consists of three levels: Level 1 (Chart Reproduction) reproduces charts from a reference figure and user query; Level 2 (Chart Editing) involves complex modifications such as changing chart types or adding elements; and Level 3 (Long-Table to Chart Generation) requires models to transform long, information-dense tables into faithful charts following user instructions. To our knowledge, this is the first hierarchical benchmark that reflects practical chart2code usage while systematically scaling task complexity. In total, Chart2Code contains 2,023 tasks across 22 chart types, paired with multi-level evaluation metrics that assess both code correctness and the visual fidelity of rendered charts. We benchmark 25 state-of-the-art (SoTA) LMMs, including both proprietary and the latest open-source models such as GPT-5, Qwen2.5-VL, InternVL3/3.5, MiMo-VL, and Seed-1.6-VL. Experimental results demonstrate that even the SoTA model GPT-5 averages only 0.57 on code-based evaluation and 0.22 on chart-quality assessment across the editing tasks, underscoring the difficulty of Chart2Code. We anticipate this benchmark will drive advances in multimodal reasoning and foster the development of more robust and general-purpose LMMs. Our code and data are available on Chart2Code.
Authors: Zirui Deng, Netanel Raviv
Abstract: Vector symbolic architectures (VSAs) are a family of information representation techniques which enable composition, i.e., creating complex information structures from atomic vectors via binding and superposition, and have recently found wide ranging applications in various neurosymbolic artificial intelligence (AI) systems and hardware systems. Recently, Raviv proposed the use of random linear codes in VSAs, suggesting that their subcode structure enables efficient unbinding, while preserving the quasi-orthogonality that is necessary for neural processing. Yet, random linear codes are difficult to decode under noise, which severely limits the resulting VSA's ability to support recovery, i.e., the retrieval of information objects and their attributes from a noisy compositional representation. In this work we bridge this gap by utilizing coding theoretic tools. First, we argue that the concatenation of Reed-Solomon and Hadamard codes is suitable for VSA, due to the mutual quasi-orthogonality of the resulting codewords (a folklore result). Second, we show that recovery of the resulting compositional representations can be done by solving a problem we call histogram recovery. In histogram recovery, a collection of $N$ histograms over a finite field is given as input, and one must find a collection of Reed-Solomon codewords of length $N$ whose entry-wise symbol frequencies obey those histograms. We present an optimal solution to the histogram recovery problem by using algorithms related to list-decoding, and analyze the resulting noise resilience. Our results give rise to a noise-resilient VSA with formal guarantees regarding efficient encoding, quasi-orthogonality, and recovery, without relying on any heuristics or training, and while operating at improved parameters relative to similar solutions such as the Hadamard code.
Authors: Zhuoyun Du, Runze Wang, Huiyu Bai, Zouying Cao, Xiaoyong Zhu, Yu Cheng, Bo Zheng, Wei Chen, Haochao Ying
Abstract: While natural language is the de facto communication medium for LLM-based agents, it presents a fundamental constraint. The process of downsampling rich, internal latent states into discrete tokens inherently limits the depth and nuance of information that can be transmitted, thereby hindering collaborative problem-solving. Inspired by telepathy, which bypasses symbolic language in communication, we propose Interlat (Inter-agent Latent Space Communication), a paradigm that leverages the continuous last hidden states of an LLM as a representation of its thought for direct communication (termed latent communication). An additional learned compression process further compresses latent communication via latent space reasoning. Experiments demonstrate that Interlat outperforms both fine-tuned chain-of-thought (CoT) prompting and single-agent baselines, even across heterogeneous models, promoting more exploratory behavior and enabling genuine utilization of latent information. Further compression not only substantially accelerates inference by up to 24 times but also maintains competitive performance through an efficient information-preserving mechanism. We position this work as a feasibility study of entirely latent space inter-agent communication, and our results highlight its potential, offering valuable insights for future research. Our code is available at https://github.com/XiaoDu-flying/Interlat.
Authors: Zhuo Li, Junjia Liu, Zhipeng Dong, Tao Teng, Quentin Rouxel, Darwin Caldwell, Fei Chen
Abstract: Vision-Language-Action (VLA) models have demonstrated significant potential in real-world robotic manipulation. However, pre-trained VLA policies still suffer from substantial performance degradation during downstream deployment. Although fine-tuning can mitigate this issue, its reliance on costly demonstration collection and intensive computation makes it impractical in real-world settings. In this work, we introduce VLA-Pilot, a plug-and-play inference-time policy steering method for zero-shot deployment of pre-trained VLA without any additional fine-tuning or data collection. We evaluate VLA-Pilot on six real-world downstream manipulation tasks across two distinct robotic embodiments, encompassing both in-distribution and out-of-distribution scenarios. Experimental results demonstrate that VLA-Pilot substantially boosts the success rates of off-the-shelf pre-trained VLA policies, enabling robust zero-shot generalization to diverse tasks and embodiments. Experimental videos and code are available at: https://rip4kobe.github.io/vla-pilot/.
Authors: Ahmad Mohammadshirazi, Pinaki Prasad Guha Neogi, Ser-Nam Lim, Rajiv Ramnath
Abstract: Document visual question answering requires models not only to answer questions correctly, but also to precisely localize answers within complex document layouts. While large vision-language models (VLMs) achieve strong spatial grounding, their inference cost and latency limit real-world deployment. Compact VLMs are more efficient, but they often suffer substantial localization degradation under standard fine-tuning or distillation. To address this gap, we propose DocVAL, a validated chain-of-thought (CoT) distillation framework that transfers explicit spatial reasoning from large teacher models to compact, deployable student VLMs. DocVAL combines (1) teacher-generated spatial CoT supervision, (2) a rule-based dual-mode validator that filters low-quality training signals and provides fine-grained, pixel-level corrective feedback, and (3) a validation-driven two-stage training procedure with iterative refinement. Text detection is used only as training-time scaffolding for supervision and validation, enabling the final student to operate as a pure VLM without OCR or detection at inference. Across multiple document understanding benchmarks, DocVAL yields consistent improvements of up to 6-7 ANLS points over comparable compact VLMs. We further introduce mean Average Precision (mAP) as a localization metric for document question answering and report strong spatial grounding performance under this new evaluation. We release 95K validator-verified CoT traces and show that high-quality, validated supervision is more effective than scaling unfiltered data, enabling efficient and trustworthy document grounding. Dataset and implementation: https://github.com/ahmad-shirazi/DocVAL
Authors: Garud Iyengar, Yu-Shiou Willy Lin, Kaizheng Wang
Abstract: As generative AI models are increasingly used to simulate real-world systems, quantifying the ``sim-to-real'' gap is critical. For each input setting of interest -- which we call a \emph{scenario}, such as a survey question or operating condition -- the real and simulated systems are associated with unobserved latent population parameters, and their discrepancy varies across scenarios. A fundamental challenge is that, for any given scenario, this discrepancy cannot be observed directly, since both systems are accessible only through finite samples, often of heterogeneous sizes across scenarios. Standard predictive inference methods are therefore ill-suited, as they quantify uncertainty in observable outputs rather than latent population parameters. To address this, we construct confidence sets for these latent parameters and use them to derive a robust proxy for the sim-to-real discrepancy. We then estimate the quantile function of this proxy to obtain a distribution-level risk profile of the simulator, which supports a broad range of statistical summaries, including statistical inference for the real output distribution in a new scenario, the calculation of risk measures like Conditional Value-at-Risk (CVaR), and principled comparisons across simulators. Our method is model-agnostic and handles general output spaces, such as categorical survey responses and continuous multi-dimensional data. We demonstrate the practical utility of this method by evaluating the alignment of four major LLMs with human populations on the WorldValueBench dataset.
Authors: Liangliang Chen, Weiyu Sun, Huiru Xie, Yongnuo Cai, Ying Zhang
Abstract: LLMs have demonstrated strong performance in data-rich domains such as programming, yet their reliability in engineering tasks remains limited. Circuit analysis--requiring multimodal understanding and precise mathematical reasoning--highlights these challenges. Although Gemini 2.5 Pro shows improved capabilities in diagram interpretation and analog-circuit reasoning, it still struggles to consistently produce correct solutions when given both textual problem descriptions and circuit diagrams. Meanwhile, engineering education demands scalable AI tools capable of generating accurate solutions for applications such as automated homework feedback. This paper presents an enhanced end-to-end circuit problem-solving framework built upon Gemini. We first conduct a systematic benchmark on undergraduate circuit problems and identify two key failure modes: 1) circuit-recognition hallucinations, particularly incorrect source polarity detection, and 2) reasoning-process hallucinations, such as incorrect current direction assumptions. To address recognition errors, we integrate a fine-tuned YOLO detector and OpenCV-based processing to isolate voltage and current sources, enabling Gemini to accurately re-identify source polarities from cropped images. To mitigate reasoning errors, we introduce an ngspice-driven verification loop, in which simulation discrepancies trigger iterative solution refinement with optional HITL feedback. Experimental results demonstrate that the proposed pipeline achieves 97.59% accuracy, substantially outperforming Gemini's baseline of 79.52%. Furthermore, on four variations of hand-drawn circuit diagrams, accuracy improves from 56.06%--71.21% to 93.94%--95.45% with statistically significant gains. These results highlight the robustness, scalability, and practical applicability of the proposed framework for engineering education and real-world circuit analysis tasks.
Authors: Joel Mire, Maria Antoniak, Steven R. Wilson, Zexin Ma, Achyutarama R. Ganti, Andrew Piper, Maarten Sap
Abstract: Reading stories evokes rich interpretive, affective, and evaluative responses, such as inferences about narrative intent or judgments about characters. Yet, computational models of reader response are limited, preventing nuanced analyses. To address this gap, we introduce SocialStoryFrames, a formalism for distilling plausible inferences about reader response, such as perceived author intent, explanatory and predictive reasoning, affective responses, and value judgments, using conversational context and a taxonomy grounded in narrative theory, linguistic pragmatics, and psychology. We develop two models, SSF-Generator and SSF-Classifier, validated through human surveys (N=382 participants) and expert annotations, respectively. We conduct pilot analyses to showcase the utility of the formalism for studying storytelling at scale. Specifically, applying our models to SSF-Corpus, a curated dataset of 6,140 social media stories from diverse contexts, we characterize the frequency and interdependence of storytelling intents, and we compare and contrast narrative practices (and their diversity) across communities. By linking fine-grained, context-sensitive modeling with a generic taxonomy of reader responses, SocialStoryFrames enable new research into storytelling in online communities.
Authors: Toshiaki Hori, Jonathan DeCastro, Deepak Gopinath, Avinash Balachandran, Guy Rosman
Abstract: We propose a new approach for solving planning problems with a hierarchical structure, fusing reinforcement learning and MPC planning. Our formulation tightly and elegantly couples the two planning paradigms. It leverages reinforcement learning actions to inform the MPPI sampler, and adaptively aggregates MPPI samples to inform the value estimation. The resulting adaptive process leverages further MPPI exploration where value estimates are uncertain, and improves training robustness and the overall resulting policies. This results in a robust planning approach that can handle complex planning problems and easily adapts to different applications, as demonstrated over several domains, including race driving, modified Acrobot, and Lunar Lander with added obstacles. Our results in these domains show better data efficiency and overall performance in terms of both rewards and task success, with up to a 72% increase in success rate compared to existing approaches, as well as accelerated convergence (x2.1) compared to non-adaptive sampling.
Authors: Raghuvir Duvvuri, Chandini Vysyaraju, Avi Goyal, Dmitry Ignatov, Radu Timofte
Abstract: Automated neural network architecture design remains a significant challenge in computer vision. Task diversity and computational constraints require both effective architectures and efficient search methods. Large Language Models (LLMs) present a promising alternative to computationally intensive Neural Architecture Search (NAS), but their application to architecture generation in computer vision has not been systematically studied, particularly regarding prompt engineering and validation strategies. Building on the task-agnostic NNGPT/LEMUR framework, this work introduces and validates two key contributions for computer vision. First, we present Few-Shot Architecture Prompting (FSAP), the first systematic study of the number of supporting examples (n = 1, 2, 3, 4, 5, 6) for LLM-based architecture generation. We find that using n = 3 examples best balances architectural diversity and context focus for vision tasks. Second, we introduce Whitespace-Normalized Hash Validation, a lightweight deduplication method (less than 1 ms) that provides a 100x speedup over AST parsing and prevents redundant training of duplicate computer vision architectures. In large-scale experiments across seven computer vision benchmarks (MNIST, CIFAR-10, CIFAR-100, CelebA, ImageNette, SVHN, Places365), we generated 1,900 unique architectures. We also introduce a dataset-balanced evaluation methodology to address the challenge of comparing architectures across heterogeneous vision tasks. These contributions provide actionable guidelines for LLM-based architecture search in computer vision and establish rigorous evaluation practices, making automated design more accessible to researchers with limited computational resources.
Authors: Hao Jiang, Zhi Yang, Annan Wang, Yichi Zhang, Weisi Lin
Abstract: Review ranking is pivotal in e-commerce for prioritizing diagnostic and authentic feedback from the deluge of user-generated content. While large language models have improved semantic assessment, existing ranking paradigms face a persistent trade-off in long-context settings. Pointwise scoring is efficient but often fails to account for list-level interactions, leading to miscalibrated top-$k$ rankings. Listwise approaches can leverage global context, yet they are computationally expensive and become unstable as candidate lists grow. To address this, we propose Residual Listwise Preference Optimization (RLPO), which formulates ranking as listwise representation-level residual correction over a strong pointwise LLM scorer. RLPO first produces calibrated pointwise scores and item representations, then applies a lightweight encoder over the representations to predict listwise score residuals, avoiding full token-level listwise processing. We also introduce a large-scale benchmark for long-context review ranking with human verification. Experiments show RLPO improves NDCG@k over strong pointwise and listwise baselines and remains robust as list length increases.
Authors: Rei Taniguchi, Yuyang Dong, Makoto Onizuka, Chuan Xiao
Abstract: Due to the prevalence of large language models (LLMs), key-value (KV) cache reduction for LLM inference has received remarkable attention. Among numerous works that have been proposed in recent years, layer-wise token pruning approaches, which select a subset of tokens at particular layers to retain in KV cache and prune others, are one of the most popular schemes. They primarily adopt a set of pre-defined layers, at which tokens are selected. Such design is inflexible in the sense that the accuracy significantly varies across tasks and deteriorates in harder tasks such as KV retrieval. In this paper, we propose ASL, a training-free method that adaptively chooses the selection layer for KV cache reduction, exploiting the variance of token ranks ordered by attention score. The proposed method balances the performance across different tasks while meeting the user-specified KV budget requirement. ASL operates during the prefilling stage and can be jointly used with existing KV cache reduction methods such as SnapKV to optimize the decoding stage. By evaluations on the InfiniteBench, RULER, and NIAH benchmarks, we show that ASL, equipped with one-shot token selection, adaptively trades inference speed for accuracy, outperforming state-of-the-art layer-wise token pruning methods in difficult tasks.
Authors: Kun Liang, Clive Bai, Xin Xu, Chenming Tang, Sanwoo Lee, Weijie Liu, Saiyong Yang, Yunfang Wu
Abstract: Recent Large Reasoning Models (LRMs) achieve strong performance by leveraging long-form Chain-of-Thought (CoT) reasoning, but uniformly applying overlong reasoning at inference time incurs substantial and often unnecessary computational cost. To address this, prior work explores various strategies to infer an appropriate reasoning budget from the input. However, such approaches are unreliable in the worst case, as estimating the minimal required reasoning effort is fundamentally difficult, and they implicitly fix the trade-off between reasoning cost and accuracy during training, limiting flexibility under varying deployment scenarios. Motivated by these limitations, we propose ORBIT, a controllable multi-budget reasoning framework with well-separated reasoning modes triggered by input. ORBIT employs multi-stage reinforcement learning to discover Pareto-optimal reasoning behaviors at each effort, followed by on-policy distillation to fuse these behaviors into a single unified model. Experiments show that ORBIT achieves (1) controllable reasoning behavior over multiple modes, (2) competitive reasoning density within each mode, and (3) integration of these frontier policies into a single unified student model while preserving clear mode separation and high per-mode performance.
Authors: Rui Sun, Jie Ding, Chenghua Gong, Tianjun Gu, Yihang Jiang, Juyuan Zhang, Liming Pan, Linyuan L\"u
Abstract: Optimizing communication topology in LLM-based multi-agent system is critical for enabling collective intelligence. Existing methods mainly rely on spatio-temporal interaction paradigms, where the sequential execution of multi-round dialogues incurs high latency and computation. Motivated by the recent insights that evaluation and debate mechanisms can improve problem-solving in multi-agent systems, we propose TopoDIM, a framework for one-shot Topology generation with Diverse Interaction Modes. Designed for decentralized execution to enhance adaptability and privacy, TopoDIM enables agents to autonomously construct heterogeneous communication without iterative coordination, achieving token efficiency and improved task performance. Experiments demonstrate that TopoDIM reduces total token consumption by 46.41% while improving average performance by 1.50% over state-of-the-art methods. Moreover, the framework exhibits strong adaptability in organizing communication among heterogeneous agents. Code is available at: https://github.com/Sundiasy/TopoDIM.
Authors: Badri N. Patro, Vijay S. Agneeswaran
Abstract: The field of artificial intelligence has undergone a revolution from foundational Transformer architectures to reasoning-capable systems approaching human-level performance. We present LLMOrbit, a comprehensive circular taxonomy navigating the landscape of large language models spanning 2019-2025. This survey examines over 50 models across 15 organizations through eight interconnected orbital dimensions, documenting architectural innovations, training methodologies, and efficiency patterns defining modern LLMs, generative AI, and agentic systems. We identify three critical crises: (1) data scarcity (9-27T tokens depleted by 2026-2028), (2) exponential cost growth ($3M to $300M+ in 5 years), and (3) unsustainable energy consumption (22x increase), establishing the scaling wall limiting brute-force approaches. Our analysis reveals six paradigms breaking this wall: (1) test-time compute (o1, DeepSeek-R1 achieve GPT-4 performance with 10x inference compute), (2) quantization (4-8x compression), (3) distributed edge computing (10x cost reduction), (4) model merging, (5) efficient training (ORPO reduces memory 50%), and (6) small specialized models (Phi-4 14B matches larger models). Three paradigm shifts emerge: (1) post-training gains (RLHF, GRPO, pure RL contribute substantially, DeepSeek-R1 achieving 79.8% MATH), (2) efficiency revolution (MoE routing 18x efficiency, Multi-head Latent Attention 8x KV cache compression enables GPT-4-level performance at $<$$0.30/M tokens), and (3) democratization (open-source Llama 3 88.6% MMLU surpasses GPT-4 86.4%). We provide insights into techniques (RLHF, PPO, DPO, GRPO, ORPO), trace evolution from passive generation to tool-using agents (ReAct, RAG, multi-agent systems), and analyze post-training innovations.
Authors: Haowei Zhang, Shudong Yang, Jinlan Fu, See-Kiong Ng, Xipeng Qiu
Abstract: Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated significant improvement in offline video understanding. However, extending these capabilities to streaming video inputs, remains challenging, as existing models struggle to simultaneously maintain stable understanding performance, real-time responses, and low GPU memory overhead. To address this challenge, we propose HERMES, a novel training-free architecture for real-time and accurate understanding of video streams. Based on a mechanistic attention investigation, we conceptualize KV cache as a hierarchical memory framework that encapsulates video information across multiple granularities. During inference, HERMES reuses a compact KV cache, enabling efficient streaming understanding under resource constraints. Notably, HERMES requires no auxiliary computations upon the arrival of user queries, thereby guaranteeing real-time responses for continuous video stream interactions, which achieves 10$\times$ faster TTFT compared to prior SOTA. Even when reducing video tokens by up to 68% compared with uniform sampling, HERMES achieves superior or comparable accuracy across all benchmarks, with up to 11.4% gains on streaming datasets.
Authors: Yuxing Chen, Guoqing Luo, Zijun Wu, Lili Mou
Abstract: Large Language Models (LLMs) exhibit social biases, which can lead to harmful stereotypes and unfair outcomes. We propose \textbf{Multi-Persona Thinking (MPT)}, a simple inference-time framework that reduces social bias by encouraging reasoning from multiple perspectives. MPT guides the model to consider contrasting social identities, such as male and female, together with a neutral viewpoint. These viewpoints then interact through an iterative reasoning process to identify and correct biased judgments. This design transforms the potential weakness of persona assignment into a mechanism for bias mitigation. We evaluate MPT on two widely used bias benchmarks with both open-source and closed-source models across different scales. Results show that MPT achieves lower bias than existing prompting-based methods while maintaining core reasoning ability.
Authors: Aditya Kumar, Mario A. Cypko, Oliver Amft
Abstract: We investigate whether temporal embedding models trained on longitudinal electronic health records can learn clinically meaningful representations without compromising predictive performance, and how architectural choices affect embedding quality. Model-guided medicine requires representations that capture disease dynamics while remaining transparent and task agnostic, whereas most clinical prediction models are optimised for a single task. Representation learning facilitates learning embeddings that generalise across downstream tasks, and recurrent architectures are well-suited for modelling temporal structure in observational clinical data. Using the MIMIC-IV dataset, we study patients with chronic kidney disease (CKD) and compare three recurrent architectures: a vanilla LSTM, an attention-augmented LSTM, and a time-aware LSTM (T-LSTM). All models are trained both as embedding models and as direct end-to-end predictors. Embedding quality is evaluated via CKD stage clustering and in-ICU mortality prediction. The T-LSTM produces more structured embeddings, achieving a lower Davies-Bouldin Index (DBI = 9.91) and higher CKD stage classification accuracy (0.74) than the vanilla LSTM (DBI = 15.85, accuracy = 0.63) and attention-augmented LSTM (DBI = 20.72, accuracy = 0.67). For in-ICU mortality prediction, embedding models consistently outperform end-to-end predictors, improving accuracy from 0.72-0.75 to 0.82-0.83, which indicates that learning embeddings as an intermediate step is more effective than direct end-to-end learning.
Authors: Rongzheng Wang, Yihong Huang, Muquan Li, Jiakai Li, Di Liang, Bob Simons, Pei Ke, Shuang Liang, Ke Qin
Abstract: Large Language Models (LLMs) have advanced the field of Combinatorial Optimization through automated heuristic generation. Instead of relying on manual design, this LLM-Driven Heuristic Design (LHD) process leverages LLMs to iteratively generate and refine solvers to achieve high performance. However, existing LHD frameworks face two critical limitations: (1) Endpoint-only evaluation, which ranks solvers solely by final gap to a reference solution, ignoring the convergence process and runtime efficiency; (2) High adaptation costs, where distribution shifts necessitate re-adaptation to generate specialized solvers for heterogeneous instance groups. To address these issues, we propose Dynamics-Aware Solver Heuristics (DASH), a framework that co-optimizes solver search mechanisms and runtime schedules guided by a convergence-aware metric, thereby identifying efficient and high-performance solvers. Furthermore, to mitigate expensive re-adaptation, DASH incorporates Profiled Library Retrieval (PLR), which maintains group-specialized solvers for profile-aware warm starts. These solvers are archived concurrently during evolution, allowing DASH to reuse matched specialists across heterogeneous distributions without restarting adaptation. Experiments on four combinatorial optimization problems demonstrate that DASH improves runtime efficiency by over 4 times while outperforming prior LHD baselines in the overall balance between gap and runtime across diverse problem scales. Furthermore, by enabling profile-aware warm starts, DASH maintains lower gap under distribution shift while reducing LLM adaptation costs by about 90%.
Authors: Junhui He, Zhihui Fu, Jun Wang, Qingan Li
Abstract: Large Language Models (LLMs) and Vision-Language Models (VLMs) have demonstrated remarkable capabilities. However, their deployment is hindered by significant computational costs. Existing structured pruning methods, while hardware-efficient, often suffer from significant accuracy degradation. In this paper, we argue that this failure stems from a stage-agnostic pruning approach that overlooks the asymmetric roles between the prefill and decode stages. By introducing a virtual gate mechanism, our importance analysis reveals that deep layers are critical for next-token prediction (decode) but largely redundant for context encoding (prefill). Leveraging this insight, we propose Prefill-Only Pruning (POP), a stage-aware inference strategy that safely omits deep layers during the computationally intensive prefill stage while retaining the full model for the sensitive decode stage. To enable the transition between stages, we introduce independent Key-Value (KV) projections to maintain cache integrity, and a boundary handling strategy to ensure the accuracy of the first generated token. Extensive experiments on Llama-3.1, Qwen3-VL, and Gemma-3 across diverse modalities demonstrate that POP achieves up to 1.37$\times$ speedup in prefill latency with minimal performance loss, effectively overcoming the accuracy-efficiency trade-off limitations of existing structured pruning methods.
Authors: Joachim Schaeffer, Arjun Khandelwal, Tyler Tracy
Abstract: Future AI deployments will likely be monitored for malicious behaviour. The ability of these AIs to subvert monitors by adversarially selecting against them - attack selection - is particularly concerning. To study this, we let a red team create attack policies that attempt to insert attacks into code without being caught by a monitor in the concentrated BigCodeBench backdooring setting. We decompose attack selection into two problems: mapping attacks to a quality score and mapping quality scores to submission probabilities. We frame attack selection as a classification problem and show that safety is significantly more sensitive to FPR than TPR. We find that prompting the attacker model to reason about the monitor while being cautious with attack selection reduces safety from a baseline of 99% to 59% at 0.5% auditing budget, emphasizing that eliciting attack selection capabilities of models is vital to avoid overly optimistic safety scores in control evaluations.
Authors: Zihao Fan, Xin Lu, Yidi Liu, Jie Huang, Dong Li, Xueyang Fu, Baocai Yin
Abstract: Powered by multimodal text-to-image priors, diffusion-based super-resolution excels at synthesizing intricate details; however, models trained on synthetic low-resolution (LR) and high-resolution (HR) image pairs often degrade when applied to real-world LR images due to significant distribution shifts. We propose Bird-SR, a bidirectional reward-guided diffusion framework that formulates super-resolution as trajectory-level preference optimization via reward feedback learning (ReFL), jointly leveraging synthetic LR-HR pairs and real-world LR images. For structural fidelity easily affected in ReFL, the model is directly optimized on synthetic pairs at early diffusion steps, which also facilitates structure preservation for real-world inputs under smaller distribution gap in structure levels. For perceptual enhancement, quality-guided rewards are applied to both synthetic and real LR images at the later trajectory phase. To mitigate reward hacking, the rewards for synthetic results are formulated in a relative advantage space bounded by their ground-truth counterparts, while real-world optimization is regularized via a semantic alignment constraint. Furthermore, to balance structural and perceptual learning, we introduce a dynamic fidelity-perception weighting strategy that emphasizes structure preservation at early stages and progressively shifts focus toward perceptual optimization at later diffusion steps. Extensive experiments on real-world SR benchmarks demonstrate that Bird-SR consistently outperforms state-of-the-art methods in perceptual quality while preserving structural consistency, validating its effectiveness for real-world super-resolution. Our code can be obtained at https://github.com/fanzh03/Bird-SR.
Authors: Markus Knauer, Samuel Bustamante, Thomas Eiband, Alin Albu-Sch\"affer, Freek Stulp, Jo\~ao Silv\'erio
Abstract: Foundation models have demonstrated impressive capabilities across diverse domains, while imitation learning provides principled methods for robot skill adaptation from limited data. Combining these approaches holds significant promise for direct application to robotics, yet this combination has received limited attention, particularly for industrial deployment. We present a novel framework that enables open-vocabulary skill adaptation through a tool-based architecture, maintaining a protective abstraction layer between the language model and robot hardware. Our approach leverages pre-trained LLMs to select and parameterize specific tools for adapting robot skills without requiring fine-tuning or direct model-to-robot interaction. We demonstrate the framework on a 7-DoF torque-controlled robot performing an industrial bearing ring insertion task, showing successful skill adaptation through natural language commands for speed adjustment, trajectory correction, and obstacle avoidance while maintaining safety, transparency, and interpretability.
Authors: Anupam Purwar, Aditya Choudhary
Abstract: Current evaluation frameworks and benchmarks for LLM powered agents focus on text chat driven agents, these frameworks do not expose the persona of user to the agent, thus operating in a user agnostic environment. Importantly, in customer experience management domain, the agent's behaviour evolves as the agent learns about user personality. With proliferation of real time TTS and multi-modal language models, LLM based agents are gradually going to become multi-modal. Towards this, we propose the MM-tau-p$^2$ benchmark with metrics for evaluating the robustness of multi-modal agents in dual control setting with and without persona adaption of user, while also taking user inputs in the planning process to resolve a user query. In particular, our work shows that even with state of-the-art frontier LLMs like GPT-5, GPT 4.1, there are additional considerations measured using metrics viz. multi-modal robustness, turn overhead while introducing multi-modality into LLM based agents. Overall, MM-tau-p$^2$ builds on our prior work FOCAL and provides a holistic way of evaluating multi-modal agents in an automated way by introducing 12 novel metrics. We also provide estimates of these metrics on the telecom and retail domains by using the LLM-as-judge approach using carefully crafted prompts with well defined rubrics for evaluating each conversation.
Authors: Charles Ye, Jasmine Cui, Dylan Hadfield-Menell
Abstract: Language models remain vulnerable to prompt injection attacks despite extensive safety training. We trace this failure to role confusion: models infer the source of text based on how it sounds, not where it actually comes from. A command hidden in a webpage hijacks an agent simply because it sounds like a user instruction. This is not just behavioral: in the model's internal representations, text that sounds like a trusted source occupies the same space as text that actually is one. We design role probes which measure how models internally perceive "who is speaking", showing that attacker-controllable signals (e.g. syntactic patterns, lexical choice) control role perception. We first test this with CoT Forgery, a zero-shot attack that injects fabricated reasoning into user prompts or ingested webpages. Models mistake the text for their own thoughts, yielding 60% attack success on StrongREJECT across frontier models with near-0% baselines. Strikingly, the degree of role confusion strongly predicts attack success. We then generalize these results to standard agent prompt injections, introducing a unifying framework that reframes prompt injection not as an ad-hoc exploit but as a measurable consequence of how models represent role.
Authors: Hanwen Shen, Ting Ying, Jiajie Lu, Shanshan Wang
Abstract: Although debiased large language models (LLMs) excel at handling known or low-bias prompts, they often fail on unfamiliar and high-bias prompts. We demonstrate via out-of-distribution (OOD) detection that these high-bias prompts cause a distribution shift, degrading static model performance. To enable real-time correction, we propose CAP-TTA, a test-time adaptation framework. CAP-TTA triggers context-aware LoRA updates only when a bias-risk score exceeds a set threshold. By utilizing an offline precomputed diagonal preconditioner, it ensures fast and stable optimization. Across multiple benchmarks and human evaluations, CAP-TTA effectively reduces toxicity/bias score with significantly lower latency than standard optimization methods (e.g., AdamW or SGD). Furthermore, it prevents catastrophic forgetting, and substantially improves narrative fluency over state-of-the-art baselines without compromising debiasing performance.
Authors: Rui Hong, Shuxue Quan
Abstract: When VLMs answer correctly, do they genuinely rely on visual information or exploit language shortcuts? We introduce the Tri-Layer Diagnostic Framework, which disentangles hallucination sources via three metrics: Latent Anomaly Detection (perceptual awareness), Visual Necessity Score (visual dependency, measured via KL divergence), and Competition Score (conflict between visual grounding and instruction following). Using counterfactual interventions (blind, noise, and conflict images) across 7 VLMs and 7,000 model-sample pairs, our taxonomy reveals that 69.6% of samples exhibit Visual Sycophancy--models detect visual anomalies but hallucinate to satisfy user expectations--while zero samples show Robust Refusal, indicating alignment training has systematically suppressed truthful uncertainty acknowledgment. A scaling analysis (Qwen2.5-VL 7B to 72B) shows larger models reduce Language Shortcuts but amplify Visual Sycophancy, demonstrating scale alone cannot resolve the grounding problem. Diagnostic scores further enable a post-hoc selective prediction strategy achieving up to +9.5pp accuracy at 50% coverage with no additional training cost.
Authors: Benjamin Lange
Abstract: When providers update AI companions, users report grief, betrayal, and loss. A growing literature asks whether the norms governing personal relationships extend to these interactions. So what, if anything, is morally significant about them? I argue that this debate has missed a prior structural question: who controls the relationship, and from where? Human-AI companion interaction is a triadic structure in which the provider exercises constitutive control over the AI. I identify three structural conditions of normatively robust dyads that the norms characteristic of personal relationships presuppose and show that AI companion interactions fail all three. This reveals what I call Unilateral Relationship Revision Power (URRP): the provider can rewrite how the AI interacts from a position where these revisions are not answerable within that interaction. I argue that URRP is pro tanto wrong in interactions designed to cultivate the norms of personal relationships, because the design produces expectations that the structure cannot sustain. URRP has three implications: i) normative hollowing, under which the interaction elicits commitment but no agent inside it bears the resulting obligations; ii) displaced vulnerability, under which the user's emotional exposure is governed by an agent not answerable to her within the interaction; and iii) structural irreconcilability, under which the interaction cultivates norms of reconciliation but no agent inside it can acknowledge or answer for the revision. I propose design principles that partially substitute for the internal constraints the triadic structure removes. A central and underexplored problem in relational AI ethics is therefore the structural arrangement of power over the human-AI interaction itself.
Authors: Domenique Zipperling, Lukas Schmidt, Benedikt Hahn, Niklas K\"uhl, Steven Kimbrough
Abstract: Current clinical decision support systems (CDSSs) typically base their predictions on correlation, not causation. In recent years, causal machine learning (ML) has emerged as a promising way to improve decision-making with CDSSs by offering interpretable, treatment-specific reasoning. However, existing research often emphasizes model development rather than designing clinician-facing interfaces. To address this gap, we investigated how CDSSs based on causal ML should be designed to effectively support collaborative clinical decision-making. Using a design science research methodology, we conducted a structured literature review and interviewed experienced physicians. From these, we derived eight empirically grounded design requirements, developed seven design principles, and proposed nine practical design features. Our results establish guidance for designing CDSSs that deliver causal insights, integrate seamlessly into clinical workflows, and support trust, usability, and human-AI collaboration. We also reveal tensions around automation, responsibility, and regulation, highlighting the need for an adaptive certification process for ML-based medical products.
Authors: Badri Narayana Patro
Abstract: Every year, 10 million pets enter shelters, separated from their families. Despite desperate searches by both guardians and lost animals, 70% never reunite, not because matches do not exist, but because current systems look only at appearance, while animals recognize each other through sound. We ask, why does computer vision treat vocalizing species as silent visual objects? Drawing on five decades of cognitive science showing that animals perceive quantity approximately and communicate identity acoustically, we present the first multimodal reunification system integrating visual and acoustic biometrics. Our species-adaptive architecture processes vocalizations from 10Hz elephant rumbles to 4kHz puppy whines, paired with probabilistic visual matching that tolerates stress-induced appearance changes. This work demonstrates that AI grounded in biological communication principles can serve vulnerable populations that lack human language.
Authors: Jiazhou Zhou, Yucheng Chen, Hongyang Li, Qing Jiang, Hu Zhou, Ying-Cong Chen, Lei Zhang
Abstract: Multimodal Large Language Models (MLLMs) have achieved remarkable success, yet they remain prone to perception-related hallucinations in fine-grained tasks. This vulnerability arises from a fundamental limitation: their reasoning is largely restricted to the language domain, treating visual input as a static, reasoning-agnostic preamble rather than a dynamic participant. Consequently, current models act as passive observers, unable to re-examine visual details to ground their evolving reasoning states. To overcome this, we propose V-Reflection, a framework that transforms the MLLM into an active interrogator through a "think-then-look" visual reflection mechanism. During reasoning, latent states function as dynamic probes that actively interrogate the visual feature space, grounding each reasoning step for task-critical evidence. Our approach employs a two-stage distillation strategy. First, the Box-Guided Compression Module (BCM) establishes stable pixel-to-latent targets through explicit spatial grounding. Next, a Dynamic Autoregressive Compression (DAC) module maps the model's hidden states into dynamic probes that interrogate the global visual feature map. By distilling the spatial expertise of the BCM teacher into the DAC student, V-Reflection internalizes the ability to localize task-critical evidence. During inference, both modules remain entirely inactive, maintaining a purely end-to-end autoregressive decoding in the latent space with optimal efficiency. Extensive experiments demonstrate the effectiveness of our V-Reflection across six perception-intensive benchmarks, significantly narrowing the fine-grained perception gap. Visualizations confirm that latent reasoning autonomously localizes task-critical visual evidence.
Authors: Jiahao Xu, Rui Hu, Olivera Kotevska, Zikai Zhang
Abstract: Multi-bit watermarking has emerged as a promising solution for embedding imperceptible binary messages into Large Language Model (LLM)-generated text, enabling reliable attribution and tracing of malicious usage of LLMs. Despite recent progress, existing methods still face key limitations: some become computationally infeasible for large messages, while others suffer from a poor trade-off between text quality and decoding accuracy. Moreover, the decoding accuracy of existing methods drops significantly when the number of tokens in the generated text is limited, a condition that frequently arises in practical usage. To address these challenges, we propose \textsc{XMark}, a novel method for encoding and decoding binary messages in LLM-generated texts. The unique design of \textsc{XMark}'s encoder produces a less distorted logit distribution for watermarked token generation, preserving text quality, and also enables its tailored decoder to reliably recover the encoded message with limited tokens. Extensive experiments across diverse downstream tasks show that \textsc{XMark} significantly improves decoding accuracy while preserving the quality of watermarked text, outperforming prior methods. The code is at https://github.com/JiiahaoXU/XMark.
Authors: Honghao Fu, Miao Xu, Yiwei Wang, Dailing Zhang, Jun Liu, Yujun Cai
Abstract: Scaling multimodal large language models (MLLMs) to long videos is constrained by limited context windows. While retrieval-augmented generation (RAG) is a promising remedy by organizing query-relevant visual evidence into a compact context, most existing methods (i) flatten videos into independent segments, breaking their inherent spatio-temporal structure, and (ii) depend on explicit semantic matching, which can miss cues that are implicitly relevant to the query's intent. To overcome these limitations, we propose VideoStir, a structured and intent-aware long-video RAG framework. It firstly structures a video as a spatio-temporal graph at clip level, and then performs multi-hop retrieval to aggregate evidence across distant yet contextually related events. Furthermore, it introduces an MLLM-backed intent-relevance scorer that retrieves frames based on their alignment with the query's reasoning intent. To support this capability, we curate IR-600K, a large-scale dataset tailored for learning frame-query intent alignment. Experiments show that VideoStir is competitive with state-of-the-art baselines without relying on auxiliary information, highlighting the promise of shifting long-video RAG from flattened semantic matching to structured, intent-aware reasoning. Codes and checkpoints are available at https://github.com/RomGai/VideoStir.
Authors: Wenyue Hua, Sripad Karne, Qian Xie, Armaan Agrawal, Nikos Pagonas, Kostis Kaffes, Tianyi Peng
Abstract: AI agents are increasingly deployed in real-world applications, including systems such as Manus, OpenClaw, and coding agents. Existing research has primarily focused on server-side efficiency, proposing methods such as caching, speculative execution, traffic scheduling, and load balancing to reduce the cost of serving agentic workloads. However, as users increasingly construct agents by composing local tools, remote APIs, and diverse models, an equally important optimization problem arises on the client side. Client-side optimization asks how developers should allocate the resources available to them, including model choice, local tools, and API budget across pipeline stages, subject to application-specific quality, cost, and latency constraints. Because these objectives depend on the task and deployment setting, they cannot be determined by server-side systems alone. We introduce AgentOpt, the first framework-agnostic Python package for client-side agent optimization. We first study model selection, a high-impact optimization lever in multi-step agent pipelines. Given a pipeline and a small evaluation set, the goal is to find the most cost-effective assignment of models to pipeline roles. This problem is consequential in practice: at matched accuracy, the cost gap between the best and worst model combinations can reach 13-32x in our experiments. To efficiently explore the exponentially growing combination space, AgentOpt implements ten search algorithms, including UCB-E, UCB-E with Low-Rank Factorization, Arm Elimination, Epsilon-LUCB, Threshold Successive Elimination, and Bayesian Optimization. Across four benchmarks, UCB-E recovers near-optimal accuracy while reducing evaluation budget by 62-76\% relative to brute-force search. Code and benchmark results available at https://agentoptimizer.github.io/agentopt/.
Authors: Tristan Simas
Abstract: Any rigorously specified problem determines an admissible-output relation $R$, and exact correctness depends only on the induced classes $s \sim_R s' \iff \mathrm{Adm}_R(s)=\mathrm{Adm}_R(s')$. Exact relevance certification asks which coordinates recover those classes. Decision, search, approximation, statistical, randomized, horizon, and distributional guarantees all reduce to this same quotient-recovery problem. Tractable cases still admit a finite primitive basis, but optimizer-quotient realizability is maximal, so quotient shape alone cannot mark the frontier. For frontier theorems, orbit gaps are the exact obstruction. Exact classification by closure-law-invariant predicates succeeds exactly when the target is constant on closure orbits; on a closure-closed domain, this is equivalent to disjointness of the positive and negative orbit hulls, and when it holds there is a least exact closure-invariant classifier. Across four natural candidate tractability predicates, a uniform pair-targeted affine witness produces same-orbit disagreements and rules out exact structural classification on the full binary pairwise domain. Because the canonical optimizer-set exact specifications of that witness class are already rigorously specified problems, no universal exact-certification characterization over formal problems escapes the same obstruction; this is by internal witness class, not by asserting that every problem is binary pairwise. Restricting the domain helps only by removing orbit gaps. Approximation also has a strict limit: without explicit gap control, arbitrarily small perturbations can flip relevance and sufficiency.
Authors: Shiwan Zhao, Zhihu Wang, Xuyang Zhao, Jiaming Zhou, Caiyue Xu, Chenfei Liu, Liting Zhang, Yuhang Jia, Yanzhe Zhang, Hualong Yu, Zichen Xu, Qicheng Li, Yong Qin
Abstract: Post-training has become central to turning pretrained large language models (LLMs) into aligned, capable, and deployable systems. Recent progress spans supervised fine-tuning (SFT), preference optimization, reinforcement learning (RL), process supervision, verifier-guided methods, distillation, and multi-stage pipelines. Yet these methods are often discussed in fragmented ways, organized by labels or objectives rather than by the behavioral bottlenecks they address. This survey argues that LLM post-training is best understood as structured intervention on model behavior. We organize the field first by trajectory provenance, which defines two primary regimes: off-policy learning on externally supplied trajectories and on-policy learning on learner-generated rollouts. We then interpret methods through two recurring roles -- effective support expansion, which makes useful behaviors more reachable, and policy reshaping, which improves behavior within already reachable regions -- together with a complementary systems-level role, behavioral consolidation, which preserves, transfers, and amortizes useful behavior across stages and model transitions. Under this view, SFT may serve either support expansion or policy reshaping; preference optimization is usually off-policy reshaping, though online variants move closer to learner-generated states. On-policy RL often improves behavior on learner-generated states, but stronger guidance can also make hard-to-reach reasoning paths reachable. Distillation is often better understood as consolidation rather than only compression, and hybrid pipelines emerge as coordinated multi-stage compositions. Overall, the framework helps diagnose post-training bottlenecks and reason about stage composition, suggesting that progress increasingly depends on coordinated systems design rather than any single dominant objective.
Authors: Pankayaraj Pathmanathan, Furong Huang
Abstract: While the wide adoption of refusal training in large language models (LLMs) has showcased improvements in model safety, recent works have highlighted shortcomings due to the shallow nature of these alignment methods. To this end, the work on Deliberative alignment proposed distilling reasoning capabilities from stronger reasoning models, thereby instilling deeper safety in LLMs. In this work, we study the impact of deliberative alignment in language models. First, we show that despite being larger in model size and stronger in safety capability, there exists an alignment gap between teacher and student language models, which affects both the safety and general utility of the student model. Furthermore, we show that models aligned through deliberative alignment can retain unsafe behaviors from the base model despite learning the reasoning patterns of larger reasoning models. Building upon this observation, we propose a BoN sampling method that attributes the unsafe behavior back to the base LLMs in the latent space, thereby down-ranking unsafe responses to gain a meaningful improvement in model safety across multiple safety benchmarks with minimal loss in utility. In particular, across 7 teacher models and 6 student models of different classes and sizes, we show an average attack success rate (ASR) reduction of 28.2% in DAN, 31.3% in WildJailbreak and 35.4 % in StrongREJECT benchmarks. We further show that these safety gains prevail post RL training, thus highlighting the uncertainty in safety reasoning and it's explicit attribution to the base model.
Authors: Mehdi Fatan Serj, C. Alejandro Parraga, Xavier Otazu
Abstract: We study how far structured architectural bias can compensate for the absence of end-to-end gradient-based representation learning in visual recognition. Building on the VisNet tradition, we introduce a modular hierarchical framework combining: (i) fixed multi-frequency Gabor decomposition into F=7 parallel streams; (ii) within-stream competitive learning with Hebbian and Oja updates and anti-Hebbian decorrelation; (iii) an associative memory module inspired by modern Hopfield retrieval; and (iv) iterative top-down modulation using local prediction and reconstruction signals. Representational layers are trained without end-to-end backpropagation through the full hierarchy; only the final linear readout and top-down projection matrices are optimized by gradient descent. We therefore interpret the model as a hybrid system that is predominantly locally trained but includes a small number of gradient-trained parameters. On CIFAR-10, the full model reaches 80.1% +/- 0.3% top-1 accuracy, linear probe), compared with 71.0% for a Hebbian-only baseline and 83.4% for a gradient-trained model on the same fixed Gabor basis. On CIFAR-100, performance is 54.8%. Factorial analysis indicates that multi-frequency streams, associative memory, and top-down feedback contribute largely additively, with a significant Streams x TopDown interaction (p=0.02). These results suggest that carefully chosen architectural priors can recover a substantial fraction of the performance typically associated with global gradient training, while leaving a measurable residual gap. Experiments are limited to CIFAR-10/100.
Authors: Jihyeon Yun, Abdullah Yasin Etcibasi, Ming Shi, C. Emre Koksal
Abstract: We develop a queueing-theoretic framework to model the temporal evolution of cyber-attack surfaces, where the number of active vulnerabilities is represented as the backlog of a queue. Vulnerabilities arrive as they are discovered or created, and leave the system when they are patched or successfully exploited. Building on this model, we study how automation affects attack and defense dynamics by introducing an AI amplification factor that scales arrival, exploit, and patching rates. Our analysis shows that even symmetric automation can increase the rate of successful exploits. We validate the model using vulnerability data collected from an open source software supply chain and show that it closely matches real-world attack surface dynamics. Empirical results reveal heavy-tailed patching times, which we prove induce long-range dependence in vulnerability backlog and help explain persistent cyber risk. Utilizing our queueing abstraction for the attack surface, we develop a systematic approach for cyber risk mitigation. We formulate the dynamic defense problem as a constrained Markov decision process with resource-budget and switching-cost constraints, and develop a reinforcement learning (RL) algorithm that achieves provably near-optimal regret. Numerical experiments validate the approach and demonstrate that our adaptive RL-based defense policies significantly reduce successful exploits and mitigate heavy-tail queue events. Using trace-driven experiments on the ARVO dataset, we show that the proposed RL-based defense policy reduces the average number of active vulnerabilities in a software supply chain by over 90% compared to existing defense practices, without increasing the overall maintenance budget. Our results allow defenders to quantify cumulative exposure risk under long-range dependent attack dynamics and to design adaptive defense strategies with provable efficiency.
Authors: Vu Tuan Truong, Long Bao Le
Abstract: Large Language Models (LLMs), despite their impressive capabilities across domains, have been shown to be vulnerable to backdoor attacks. Prior backdoor strategies predominantly operate at the token level, where an injected trigger causes the model to generate a specific target word, choice, or class (depending on the task). Recent advances, however, exploit the long-form reasoning tendencies of modern LLMs to conduct reasoning-level backdoors: once triggered, the victim model inserts one or more malicious reasoning steps into its chain-of-thought (CoT). These attacks are substantially harder to detect, as the backdoored answer remains plausible and consistent with the poisoned reasoning trajectory. Yet, defenses tailored to this type of backdoor remain largely unexplored. To bridge this gap, we propose Critical-CoT, a novel defense mechanism that conducts a two-stage fine-tuning (FT) process on LLMs to develop critical thinking behaviors, enabling them to automatically identify potential backdoors and refuse to generate malicious reasoning steps. Extensive experiments across multiple LLMs and datasets demonstrate that Critical-CoT provides strong robustness against both in-context learning-based and FT-based backdoor attacks. Notably, Critical-CoT exhibits strong cross-domain and cross-task generalization. Our code is available at hthttps://github.com/tuanvu171/Critical-CoT.
Authors: Yinuo Yang, Zixian Ma, Manasi Ganti, Jieyu Zhang, Ranjay Krishna
Abstract: We present a discriminative multimodal reward model that scores all candidate responses in a single forward pass. Conventional discriminative reward models evaluate each response independently, requiring multiple forward passes, one for each potential response. Our approach concatenates multiple responses with separator tokens and applies cross-entropy over their scalar scores, enabling direct comparative reasoning and efficient $N$-way preference learning. The multi-response design also yields up to $N\times$ wall-clock speedup and FLOPs reduction over conventional single-response scoring. To enable $N$-way reward evaluation beyond existing pairwise benchmarks, we construct two new benchmarks: (1) MR$^2$Bench-Image contains human-annotated rankings over responses from 8 diverse models; (2) MR$^2$Bench-Video is a large-scale video-based reward benchmark derived from 94K crowdsourced pairwise human judgments over video question-answering spanning 19 models, denoised via preference graph ensemble. Both benchmarks provide 4-response evaluation variants sampled from the full rankings. Built on a 4B vision-language backbone with LoRA fine-tuning and a lightweight MLP value head, our model achieves state-of-the-art results on six multimodal reward benchmarks, including MR$^2$Bench-Image, MR$^2$Bench-Video, and four other existing benchmarks. Our model outperforms existing larger generative and discriminative reward models. We further demonstrate that our reward model, when used in reinforcement learning with GRPO, produces improved policy models that maintain performance across standard multimodal benchmarks while substantially improving open-ended generation quality, outperforming a single-response discriminative reward model (RM) baseline by a large margin in both training stability and open-ended generation quality.
Authors: Jialu Pan, Yufeng Zhang, Nan Hu, Zhenbang Chen, Ji Wang, Keqin Li
Abstract: We study the problem of characterizing the stability of Kullback-Leibler (KL) divergence under Gaussian perturbations beyond Gaussian families. Existing relaxed triangle inequalities for KL divergence critically rely on the assumption that all involved distributions are Gaussian, which limits their applicability in modern applications such as out-of-distribution (OOD) detection with flow-based generative models. In this paper, we remove this restriction by establishing a sharp stability bound between an arbitrary distribution and Gaussian families under mild moment conditions. Specifically, let $P$ be a distribution with finite second moment, and let $\mathcal{N}_1$ and $\mathcal{N}_2$ be multivariate Gaussian distributions. We show that if $KL(P||\mathcal{N}_1)$ is large and $KL(\mathcal{N}_1||\mathcal{N}_2)$ is at most $\epsilon$, then $KL(P||\mathcal{N}_2) \ge KL(P||\mathcal{N}_1) - O(\sqrt{\epsilon})$. Moreover, we prove that this $\sqrt{\epsilon}$ rate is optimal in general, even within the Gaussian family. This result reveals an intrinsic stability property of KL divergence under Gaussian perturbations, extending classical Gaussian-only relaxed triangle inequalities to general distributions. The result is non-trivial due to the asymmetry of KL divergence and the absence of a triangle inequality in general probability spaces. As an application, we provide a rigorous foundation for KL-based OOD analysis in flow-based models, removing strong Gaussian assumptions used in prior work. More broadly, our result enables KL-based reasoning in non-Gaussian settings arising in deep learning and reinforcement learning.
Authors: Haofu Yang, Jiaji Liu, Chen Huang, Faguo Wu, Wenqiang Lei, See-Kiong Ng
Abstract: Developing non-collaborative dialogue agents traditionally requires the manual, unscalable codification of expert strategies. We propose \ours, a method that leverages large language models to autonomously induce both strategy actions and planning logic directly from raw transcripts. METRO formalizes expert knowledge into a Strategy Forest, a hierarchical structure that captures both short-term responses (nodes) and long-term strategic foresight (branches). Experimental results across two benchmarks show that METRO demonstrates promising performance, outperforming existing methods by an average of 9%-10%. Our further analysis not only reveals the success behind METRO (strategic behavioral diversity and foresight), but also demonstrates its robust cross-task transferability. This offers new insights into building non-collaborative agents in a cost-effective and scalable way. Our code is available at https://github.com/Humphrey-0125/METRO.
Authors: Pengfeng Li, Chen Huang, Chaoqun Hao, Hongyao Chen, Xiao-Yong Wei, Wenqiang Lei, See-Kiong Ng
Abstract: Contextual causal reasoning is a critical yet challenging capability for Large Language Models (LLMs). Existing benchmarks, however, often evaluate this skill in fragmented settings, failing to ensure context consistency or cover the full causal hierarchy. To address this, we pioneer METER to systematically benchmark LLMs across all three levels of the causal ladder under a unified context setting. Our extensive evaluation of various LLMs reveals a significant decline in proficiency as tasks ascend the causal hierarchy. To diagnose this degradation, we conduct a deep mechanistic analysis via both error pattern identification and internal information flow tracing. Our analysis reveals two primary failure modes: (1) LLMs are susceptible to distraction by causally irrelevant but factually correct information at lower level of causality; and (2) as tasks ascend the causal hierarchy, faithfulness to the provided context degrades, leading to a reduced performance. We belive our work advances our understanding of the mechanisms behind LLM contextual causal reasoning and establishes a critical foundation for future research. Our code and dataset are available at https://github.com/SCUNLP/METER .
Authors: Miit Daga, Swarna Priya Ramu
Abstract: Fine-tuning pretrained image classifiers is standard practice, yet which individual samples are forgotten during this process, and whether forgetting patterns are stable or architecture dependent, remains unclear. Understanding these dynamics has direct implications for curriculum design, data pruning, and ensemble construction. We track per-sample correctness at every epoch during fine-tuning of ResNet-18 and DeiT-Small on a retinal OCT dataset (7 classes, 56:1 imbalance) and CUB-200-2011 (200 bird species), fitting Ebbinghaus-style exponential decay curves to each sample's retention trace. Five findings emerge. First, the two architectures forget fundamentally different samples: Jaccard overlap of the top 10 percent most-forgotten is 0.34 on OCTDL and 0.15 on CUB-200. Second, ViT forgetting is more structured (mean $R^2 = 0.74$) than CNN forgetting ($R^2 = 0.52$). Third, per-sample forgetting is stochastic across random seeds (Spearman $\rho \approx 0.01$), challenging the assumption that sample difficulty is an intrinsic property. Fourth, class-level forgetting is consistent and semantically interpretable: visually similar species are forgotten most, distinctive ones least. Fifth, a sample's loss after head warmup predicts its long-term decay constant ($\rho = 0.30$ to $0.50$, $p < 10^{-45}$). These findings suggest that architectural diversity in ensembles provides complementary retention coverage, and that curriculum or pruning methods based on per-sample difficulty may not generalize across runs. A spaced repetition sampler built on these decay constants does not outperform random sampling, indicating that static scheduling cannot exploit unstable per-sample signals.
Authors: Hiroyuki Chuma, Kanji Otsuka, Yoichi Sato
Abstract: This paper reports an unexpected finding: in a deterministic hyperdimensional computing (HDC) architecture based on Galois-field algebra, a path-dependent semantic selection mechanism emerges, equivalent to spike-timing-dependent plasticity (STDP), with magnitude predictable a priori by a closed-form expression matching large-scale measurements. This addresses limitations of modern AI including catastrophic forgetting, learning stagnation, and the Binding Problem at an algebraic level. We propose VaCoAl (Vague Coincident Algorithm) and its Python implementation PyVaCoAl, combining ultra-high-dimensional memory with deterministic logic. Rooted in Sparse Distributed Memory, it resolves orthogonalisation and retrieval in high-dimensional binary spaces via Galois-field diffusion, enabling low-load deployment. VaCoAl is a memory-centric architecture prioritising retrieval and association, enabling reversible composition while preserving element independence and supporting compositional generalisation with a transparent reliability metric (CR score). We evaluated multi-hop reasoning on about 470k mentor-student relations from Wikidata, tracing up to 57 generations (over 25.5M paths). Using HDC bundling and unbinding with CR-based denoising, we quantify concept propagation over DAGs. Results show a reinterpretation of the Newton-Leibniz dispute and a phase transition from sparse convergence to a post-Leibniz "superhighway", from which structural indicators emerge supporting a Kuhnian paradigm shift. Collision-tolerance mechanisms further induce path-based pruning that favors direct paths, yielding emergent semantic selection equivalent to STDP. VaCoAl thus defines a third paradigm, HDC-AI, complementing LLMs with reversible multi-hop reasoning.
Authors: Hiranya V. Peiris
Abstract: The Claude Mythos Preview system card deploys emotion vectors, sparse autoencoder (SAE) features, and activation verbalisers to study model internals during misaligned behaviour. The two primary toolkits are not jointly reported on the most alignment-relevant episodes. This note identifies two hypotheses that are qualitatively consistent with the published results: that the emotion vectors track functional emotions that causally drive behaviour, or that they are a projection of a richer situational-context structure onto human emotional axes. The hypotheses can be distinguished by cross-referencing the two toolkits on episodes where only one is currently reported: most directly, applying emotion probes to the strategic concealment episodes analysed only with SAE features. If emotion probes show flat activation while SAE features are strongly active, the alignment-relevant structure lies outside the emotion subspace. Which hypothesis is correct determines whether emotion-based monitoring will robustly detect dangerous model behaviour or systematically miss it.
Authors: Itay Itzhak, Eliya Habba, Gabriel Stanovsky, Yonatan Belinkov
Abstract: Evaluating LLMs is challenging, as benchmark scores often fail to capture models' real-world usefulness. Instead, users often rely on ``vibe-testing'': informal experience-based evaluation, such as comparing models on coding tasks related to their own workflow. While prevalent, vibe-testing is often too ad hoc and unstructured to analyze or reproduce at scale. In this work, we study how vibe-testing works in practice and then formalize it to support systematic analysis. We first analyze two empirical resources: (1) a survey of user evaluation practices, and (2) a collection of in-the-wild model comparison reports from blogs and social media. Based on these resources, we formalize vibe-testing as a two-part process: users personalize both what they test and how they judge responses. We then introduce a proof-of-concept evaluation pipeline that follows this formulation by generating personalized prompts and comparing model outputs using user-aware subjective criteria. In experiments on coding benchmarks, we find that combining personalized prompts and user-aware evaluation can change which model is preferred, reflecting the role of vibe-testing in practice. These findings suggest that formalized vibe-testing can serve as a useful approach for bridging benchmark scores and real-world experience.