Authors: Merkouris Papamichail, Konstantinos Varsos, Giorgos Flouris, Jo\~ao Marques-Silva
Abstract: In this work we present a rigorous theoretical framework to a foundational problem of AI safety, namely adversarial robustness. In particular, we show that the adversarial robustness problem can be reduced to a lattice traversal problem. Each element of this lattice corresponds to an interval, i.e., an axis-aligned hyper-rectangle, containing an input point $\mathbf{x}$. Consider a multilayered perceptron classifier (MLP). An interval $I$ constitutes a sound certification if $\mathbf{x} \in I$ and $\mathbf{x}$ can be freely perturbed in $I$ without changing the MLP's prediction. Complementarily, an interval $I$ constitutes a complete certification if $\mathbf{x} \in I$ and when $\mathbf{x}$ moves outside of $I$ the MLP's prediction is guaranteed to change. While the sound certification problem corresponds to the well-studied adversarial robustness, complete certifications have not been examined in the literature. We develop lattice traversal operators, which we apply in a refine & verify iterative scheme. Using formal MLP verifiers, sound maximality and complete minimality are guaranteed. Moreover, we examine objective optimization problems. There we discover some interesting asymmetries. For complete certifications, the minimum solution is obtained in polynomial oracle calls. This does not hold for sound certifications, where we prove strong intractability results. Additionally, we examine optimization problems in symmetric intervals (i.e., $\ell_\infty$-spheres), where we provide logarithmic algorithms. Finally, we present an empirical evaluation, using the novel ParallelepipedoNN system.
Authors: Vanessa Figueiredo, Wilter Franceschi
Abstract: Reliability in large language model (LLM) systems is typically framed as a function of model capability. We challenge this by demonstrating that reliability is significantly influenced by \emph{inference-time control} -- the computational layer governing task framing and context selection. We introduce \emph{CogniConsole}, an architectural instantiation that externalizes this control into a structured interface combining programmatic coordination with bounded prompt-based reasoning. Through \emph{controllability-oriented probes} ($N=489$) in a multi-step interactive environment, we show that increasing structural scaffolding -- from unstructured to fully scaffolded -- \textbf{systematically reduces output variance and failure rates under a fixed model architecture}. Our results indicate that many observed failure modes, such as context drift and inconsistent constraint adherence, arise from under-specified control rather than insufficient capability. This work provides an empirical basis for treating inference-time control as a first-class abstraction, opening new directions for designing and evaluating LLM systems beyond scaling alone.
Authors: Maureese Williams, Dymitr Nowicki
Abstract: Large Language Model (LLM) agents have shown promise in multi-step planning tasks, but existing approaches like LATS (Language Agent Tree Search) and ReAct rely heavily on LLM inference during planning, leading to high computational costs and stochastic behavior. We present \textbf{GATS} (Graph-Augmented Tree Search), a planning framework that combines systematic UCB1-based tree search with a layered world model to eliminate LLM calls during inference while achieving superior planning performance. Our three-layer world model integrates: (L1) exact symbolic action matching, (L2) statistics learned from execution logs, and (L3) LLM-based prediction for unknown actions. On synthetic planning tasks with branching paths and dead-ends, GATS achieves \textbf{100\% success rate} compared to 92 % for LATS and 64\% for ReAct. On a comprehensive stress test spanning 12 challenging scenarios -- including coding workflows, web navigation, and long-horizon tasks -- GATS maintains \textbf{100\% success} while LATS drops to 88.9 % and ReAct to 23.9%. GATS requires \textbf{zero LLM calls per task} during planning (vs. 37 per task for LATS) and produces deterministic plans with zero variance across runs. Our results demonstrate that systematic search with learned world models can substantially outperform LLM-guided exploration for agent planning.
Authors: Zongxia Li, Zhongzhi Li, Yucheng Shi, Ruhan Wang, Junyao Yang, Zhichao Liu, Xiyang Wu, Anhao Li, Yue Yu, Ninghao Liu, Lichao Sun, Haotao Mi, LeoweiLiang
Abstract: AI agents have become capable of autonomously completing short, well-specified tasks. However, existing terminal benchmarks largely focus on simple problems that finish within minutes and are evaluated only by their final outcome. This setup overlooks intermediate progress and partial solutions, yielding sparse reward signals and an incomplete picture of agent capability. We introduce Long-Horizon-Terminal-Bench, a terminal benchmark of 46 long-horizon tasks spanning nine categories, including experiment reproduction, software engineering, multimodal analysis, interactive games, and scientific computing. Each task follows a Terminal-Bench-style setup with a reference solution or simulation engine, but is further decomposed into fine-grained graded subtasks. This design enables dense intermediate rewards and partial credit, allowing evaluation to capture not only whether an agent reaches the final goal, but also how far it progresses on open-ended workflows. Tasks in Long-Horizon-Terminal-Bench typically require hundreds of episodes and minutes to hours of execution, stressing long-horizon planning, long-context management, and iterative debugging rather than one-shot problem solving. We evaluate 15 frontier models and find that agents consume on average 9.9M tokens per task, with roughly 231 episodes and 85.3 minutes of execution time per run, making Long-Horizon-Terminal-Bench more demanding than prior terminal-based benchmarks. Even the strongest tested model achieves 15.2% pass@1 at a partial-reward threshold of 0.95 and 10.9% at a perfect-reward threshold of 1.0, while the mean pass rate across models is 4.3% and 1.7% under the two thresholds, respectively. These results reveal headroom for improvement. We further analyze failure modes and error patterns, and release Long-Horizon-Terminal-Bench to support future progress on long-horizon terminal agents.
Authors: Joseph K. Miller
Abstract: We formalize a research result in the Lean 4 proof assistant by having a mathematician direct an AI system, and frame the activity as a formalization game. The objective is to turn a LaTeX document into Lean. The game is won when the development compiles, contains no sorry, and a machine check shows the target theorems rest on Lean's foundational axioms alone. Reuse is a second check, by a definition we introduce: whether the development yields a self-contained layer of general mathematics the wider library could absorb. The case study is a complete, axiom-clean formalization of well-posedness for the nonlinear Vlasov equation via Dobrushin's mean-field route -- existence, uniqueness, the stability estimate and mean-field limit, and a short-window superposition principle (weak solutions are Lagrangian). The human's role was to direct, not to write proofs: to scope the definitions, steer the decompositions, and triage the library's gaps; the AI agent executed. The formalization certifies the proof of each statement as written; whether the written statement is the intended theorem stays the mathematician's judgment. The optimal-transport machinery that fell out of the build (in particular, properties of the Wasserstein-1 metric and the Kantorovich-Rubinstein duality theorem) separates into a self-contained layer that compiles against Mathlib alone: about a sixth of the development (49 of 299 declarations), behind a 22-declaration interface with no reverse dependency. The headline theorems ran in about a week, the full development in about a month. We report the quantitative claims as observations of one game, not as general laws. The game's rules name no particular system, so the methodological framing is meant to outlast the tools of any one run.
Authors: Kunbo Zhang, Lei Fu, Zeyu Wang, Zijing Liu, Kejian Tong
Abstract: We present ARCANA, a collaborative multi agent framework for solving ARC AGI 2 tasks under strict test time and hardware constraints. ARCANA decomposes each task into iterative perception, hypothesis generation, symbolic execution, and reflective refinement. A perceptual grounding agent builds object centric scene graphs from raw grids, a latent program policy proposes diverse DSL programs, a symbolic executor verifies candidates on demonstrations, and a reflective agent synthesizes failure driven feedback for the next turn. These agents communicate through a shared differentiable blackboard and are scheduled by a learned meta controller. The design combines structured program search with adaptive multi turn correction, improving reasoning efficiency and solution quality on challenging abstract transformation tasks.
Authors: Saroj Gopali, Bipin Chhetri, Deepika Giri, Sima Siami-Namini, Akbar Siami Namin
Abstract: Cyberattacks on operational technology are increasingly causing costly downtime and physical damage, exposing the limitations of traditional rule-based monitoring in industrial IoT environments. While Large Language Models (LLMs) have strong semantic reasoning abilities to assist in decision support, their hallucinatory nature presents unacceptable safety liabilities for closed-loop control. This paper introduces a neuro-agentic control framework, a novel architecture that couples an LLM-based planner (i.e., such as Gemini 2.5 Flash-Lite) with a pre-trained Time-Series Foundation Model (TimesFM), to achieve physics-grounded autonomous defense. The paper introduces a ``Counterfactual Physics Injection'' mechanism that simulates the impact of LLM-proposed interventions within the numerical latent space of the foundation model before actuation, while allowing the system to reject hallucinatory or unsafe actions. Evaluated on an industrial dataset (e.g., the Secure Water Treatment (SWaT)) in the context of stochastic attack scenarios, the framework exhibited better performance compared to LSTM and TCN baselines. The Neuro-Agentic Loop prevented five breaches (33.3%) below the threshold versus LSTM (26.7%) and TCN (13.3%), with zero physically invalid (hallucinated) actions executed. These results demonstrate the efficacy of using foundation models as deterministic ``Sentinels'' to safeguard agentic AI in critical infrastructure.
Authors: Tan-Minh Nguyen, Hoang-Trung Nguyen, Huu-Dong Nguyen, Dinh-Truong Do, Thi-Hai-Yen Vuong, Le-Minh Nguyen
Abstract: While multi-agent debate (MAD) frameworks have shown significant potential in general reasoning, their effectiveness in highly structured, knowledge-heavy legal domains remains under-explored. In this work, we introduce the Legal Multi-Agent Debate (L-MAD) framework to systematically evaluate different debate structures and aggregation methods within Legal Textual Entailment. By assigning distinct expert personas to multiple agents, L-MAD improves upon strong single-agent baselines by up to 8\%. Furthermore, analyzing how debate scales reveals a clear trade-off: increasing the agent population reduces inconsistency and improves accuracy, whereas extending discussion rounds induces a detrimental \textit{over-deliberation drift} where agents reinforce each other's mistakes. Ultimately, our findings outline the practical boundaries and safety margins of deploying collaborative multi-agent systems in high-stakes legal reasoning environments.
Authors: Runhan Shi, Quan Zhou, Yuqian Xu, Shuai Yang, Xin Wu, Zitong Zhou, Hui Liu, Bin Cha, Zheming Wang, Liya Li, Wei Wei, Haoyuan Hu, Jun Xu
Abstract: Large language models (LLMs) are increasingly deployed in online medical consultation, yet existing benchmarks remain poorly aligned with real clinical practice. Many rely on synthetic conversations or patient simulators, omit patient-uploaded medical images, or evaluate open-ended clinical responses using multiple-choice or lexical-overlap metrics that poorly reflect clinical quality. We introduce \textbf{MedRealMM}, a large-scale benchmark for multimodal online medical consultation built from de-identified patient-doctor interactions collected from a nationwide Chinese internet hospital. MedRealMM uses a Multimodal Clinical Challenge Point (MCCP) extraction framework to identify clinically demanding moments in authentic consultation trajectories and converts each into a standardized next-response generation task while preserving the preceding text-image context. Each instance is paired with a case-specific rubric refined by physicians that rewards clinically desirable behaviors and penalizes unsafe, unsupported, or contradictory responses. The current release contains 5,620 real-world multimodal cases spanning 64 clinical departments. We evaluate 19 general-purpose and medical-specialized LLMs, including text-only and multimodal systems. Our results show that image information is critical for reliable clinical performance and that current frontier models remain below the online physician response. Although some frontier models satisfy as many or more positive clinical criteria than physicians, they trigger more negative criteria, indicating that safety-sensitive error avoidance remains a central bottleneck. MedRealMM offers a realistic and reproducible benchmark for evaluating multimodal medical reasoning in real-world online consultation. The dataset will be publicly available on Hugging Face at https://huggingface.co/datasets/jdh-algo/MedRealMM.
Authors: Peng Kuang, Haibo Jin, Xiaoyu Han, Yanli Wang, Xiaopeng Yuan, Ye Yu, Kaidi Xu, Haohan Wang
Abstract: Process Reward Models (PRMs) have been proven to be highly effective in guiding test-time scaling (TTS) methods, which significantly boost the capabilities of LLM-based multi-agent systems. However, existing PRMs are text-based: they re-encode the entire trajectory text from scratch. In long multi-agent rollouts, the scoring cost, growing quadratically with respect to sequence length L, creates a severe computational bottleneck, severely limiting PRMs' application in long-context scenarios. To resolve this, we introduce KV-PRM, a highly efficient process reward model that eliminates the heavy text re-encoding by directly reading the KV cache produced naturally during the LLM's generation phase. By processing a single "verify token" against the pre-existing KV cache, KV-PRM reduces the scoring cost from O(L^2) to O(L). We formally prove that the KV cache contains strictly greater information capacity than text, and is more efficient for downstream reward modeling. Empirically, across the MATH, GSM8K, and AIME benchmarks, KV-PRM matches or strictly outperforms text-PRMs under various TTS methods such as Beam Search, MCTS, and Weighted Voting, with up to a 5,000x reduction in scoring FLOPs, a 37x reduction in latency, and a 34x reduction in per-sequence memory footprint compared to text-based PRMs.
Authors: Dan C. Hsu, Luke Lu
Abstract: Deployed LLM agents rely on agentic context, the model-external textual control content assembled by an operational harness. In this work, the mutable component of that context is a persistent system-level instruction that is updated from operational experience while the model, tools, and harness remain fixed. Over long evolution horizons, flat-text maintenance makes verification increasingly difficult as accumulated instructions grow and interact. We propose Graph-Regularized Agentic Context Evolution (GRACE), which maintains the persistent instruction component as a typed semantic graph and validates proposed updates within the local typed neighborhoods of modified nodes. Accepted graph updates are reconstructed as incremental edits to the textual instruction checkpoint used at deployment. We evaluate GRACE within a fixed telecom agent harness derived from $\tau^2$-bench under a controlled distribution-shift protocol. Across five independent replications, GRACE improves strict reliability, measured by pass^3, from the Gemini 2.5 Flash zero-shot value of 0.091 to 0.673$\pm$0.136 at the final checkpoint. This exceeds a Gemini 3.1 Pro zero-shot reference of 0.242 on the same held-out set, while the flat-text HCE baseline finishes at 0.191$\pm$0.051. These results identify two requirements for reliable long-horizon context evolution, a structural substrate that makes verification local and a consolidation mechanism that keeps accumulated instruction content usable.
Authors: Izumi Takahara, Teruyasu Mizoguchi
Abstract: Large language model (LLM) agents are increasingly expected to play a central role in AI-driven scientific discovery. Equipped with broad knowledge, flexible reasoning, and tool use, they have the potential to autonomously explore and solve scientific problems by repeatedly proposing hypotheses, testing them, and revising their beliefs in the light of the evidence. In current agents, however, these hypotheses, tests, and belief updates are buried in unstructured logs, and no mechanism lets the agent or the human researcher audit that process. Here we propose the Hypothesis Evolution Protocol (HEP), an agent harness that provides hypothesis generation, evaluation, and evolution as explicit, auditable operations. On materials-science research tasks, a HEP-equipped agent operates the hypothesis--test--evidence--belief cycle that planning-style agents lack, generalizes across research questions, and exploits the protocol more fully as the base LLM becomes more capable. These results mark a step toward auditable AI scientists, whose scientific reasoning can be inspected, verified, and built upon.
Authors: Mat\v{e}j Kripner, Milan Straka
Abstract: In this system paper, we present OpenProver, an open-source system for LLM-driven automated theorem proving (ATP) with integrated Lean 4 formal verification. OpenProver integrates a Planner-Worker-Verifier architecture inspired by recent ATP agentic systems such as Aletheia. A Planner agent maintains a compact Whiteboard scratchpad and an unbounded Repository of intermediate findings, and decomposes mathematical work into parallel Workers. OpenProver is fully open-source, offers reproducible evaluation through automatic formal verification of generated proofs, and provides an interactive terminal interface for human-guided proof search. In interactive mode, OpenProver allows the human operator to monitor and steer the proof search process, motivated by the established human-AI synergy in interactive code generation. To showcase the potential for quantitative ablation experiments enabled by automatic formal verification, we evaluate OpenProver on ProofNet and compare it with a simple baseline. OpenProver is publicly available at https://github.com/kripner/OpenProver.
Authors: Yanzhen Chen, Zihan Xu, Xiaocheng Zhang, Zhiting Fan, Weiqi Zhai, Hongxia Xu, Zuozhu Liu
Abstract: In this work, we introduce LongMedBench, a real-world EHR-based benchmark for long-horizon clinical decision-making. Prior evaluations of LLM-based medical agents have largely emphasized short-context knowledge QA and tool use. However, real-world medical care is inherently longitudinal, and clinicians must aggregate evidence across repeated visits, tests, and evolving treatments. Therefore, long-horizon interaction is essential for realistic assessment. LongMedBench is constructed via a reproducible pipeline that integrates MIMIC-IV admission records and clinical notes into time-series event streams and long-context memory datasets, enabling long-horizon, multi-session interactions between agents and a clinical environment. It comprises 335 patients, with 19.72 inpatient visits per patient on average and 44.91 medical events per visit. Guided by the long-horizon decision process, we propose an evaluation taxonomy with three suites: fact-based QA, temporal reasoning, and long-horizon decision-making. This taxonomy measures how agents understand and leverage historical patient information over extended horizons. Our experiments show that while recent LLMs can make good use of explicit timestamps, they have challenges in implicit time inference; The RAG and agent memory system can improve the performance of information retrieval tasks, but the performance of decision-making tasks is highly dependent on the model's immediate context.
Authors: Nuocheng Yang, Sihua Wang, Zihan Chen, Tony Q. S. Quek, Changchuan Yin
Abstract: Embodied agent teams powered by heterogeneous large language models (LLMs) are being widely deployed in physical artificial intelligence such as smart factories, warehouses, and service robotics. To enable collaboration among such an agent team, efficient coordination mechanisms that operate reliably under limited network resources are required. However, existing heterogeneous LLM-agent coordination frameworks that rely on multi-round natural-language-based conversations introduce three coupled challenges. First, inter-agent dialogue incurs communication overhead that grows rapidly with team size. Second, the quality of coordination is constrained by the heterogeneous capabilities of the agent team's LLMs. Third, agents may suffer from action delays due to iterative negotiation. To address these challenges, we propose LDT-Coord, a networked coordination framework built upon a lightweight digital twin (DT). Specifically, each agent independently selects its intended action and reports both the action decision and a structured temporal constraint over shared resources to the DT server, thereby decoupling coordination performance from natural-language reasoning ability. Then, DT executes a training-free, rule-based orchestrator algorithm to resolve cross-agent conflicts and returns coordination instructions to prevent such conflicts. To further reduce communication overhead, we formulate agent reporting control as a constrained partially observable Markov decision process (C-POMDP) and solve it with the PPO-Lagrangian algorithm. Simulation results show that LDT-Coord achieves a task success rate comparable to conventional coordination methods while reducing communication overhead by more than 70x and maintaining robustness under LLM heterogeneity.
Authors: Jingbo Chen, He Wang, Wei Yuan, Yuqiao Lai, Zhenyan Lu
Abstract: Worldbuilding, the construction of coherent fictional worlds, is a foundational task in game design and literary creation. Large Language Models (LLMs) offer new possibilities for automated content generation, but their application to worldbuilding faces three challenges: context explosion that grows linearly with the building process, the tension between creative diversity and content consistency, and the absence of automated quality assurance. This paper presents AutoWorldBuilder, a multi-agent collaborative system that addresses these challenges through five integrated components: a structured concept network with conflict detection; a DAG-based hybrid batch scheduler that groups tasks by semantic locality; a four-layer context compression mechanism achieving approximately 90% token reduction; an iterative review system with specialized Auditor agents that improves proposal pass rates from 42% to over 85%; and a skill-driven agent architecture supporting zero-code extension with differentiated temperature configuration. Two experiments across 20 diverse worldbuilding tasks, using GPT-OSS 120B and DeepSeek v3.2 as LLM backends, demonstrate a 95.0% success rate. The system generated 56-103 self-consistent concepts per world in 18-31 minutes with zero-conflict delivery. The architectural patterns validated here, including layer-as-budget compression, semantic-locality scheduling, and separation of generation and review, transfer to the broader class of knowledge-intensive, multi-agent LLM applications.
Authors: Debargha Ghosh, Silja Renooij, Anna Kononova
Abstract: Bayesian causal discovery is widely used for its ability to quantify epistemic uncertainty over directed acyclic graphs (DAGs) through posterior inference. However, its behaviour under latent confounding remains poorly understood, as existing work typically notes that confounding breaks identifiability without characterising how the posterior distribution over DAGs responds. In this work, we analyse posterior behaviour under latent confounding in linear Gaussian causal models, focusing on additive latent confounding between exactly two observed variables. We derive a critical correlation threshold above which the score function favours graphs with a spurious edge between the confounded variables, and show that this threshold decreases with sample size -- more data lowers the correlation required for the spurious edge to be favoured. Beyond this threshold, we characterize two distinct posterior failure regimes determined by the local structure around the confounded variables. Our findings are supported by exact posterior computations on multiple graph structures, demonstrating both the predicted failure regimes.
Authors: Johannes Schmitt, Tim Gehrunger, Jasper Dekoninck, Gergely B\'erczi, Uri Kreitner, Liam Price, David Holmes
Abstract: Large language models (LLMs) have shown increasing promise in solving open problems in mathematics. However, their performance can be further improved through agentic workflows tailored to real-world mathematical practice. To this end, we introduce ProofCouncil, a mathematical agent that is designed to tackle open problems using an author-critic architecture. ProofCouncil served as a submission to the second batch of FirstProof, a challenge consisting of 10 real-world mathematical problems that agents must solve autonomously. Its submissions for 6 of the 10 problems were judged by the referees to be correct up to at most minor revisions, showing the best performance among participating teams. We also evaluate ProofCouncil on 30 open problems collected from mathematical researchers. Among the 21 solutions that received human feedback, 5 were judged completely correct, 2 more were judged promising pending final verification, and a further 8 contained useful partial progress. In this short paper, we describe the development of ProofCouncil and the agent-building library used to create it, which we release as open source to the community.
Authors: Jade Alglave, Patrick Cousot
Abstract: An AI system's output is not the fact or world state it appears to describe, but rather an engineered representation. We propose a semantic framework to describe AI systems, to be able to examine the correctness of such representations. To do so, we distinguish what is justified by accepted domain knowledge, what reference sources say, and what the system can currently use. This allows us to give precise definitions to common failures: extrapolation, refuted or unsupported assertion, sources versus knowledge mismatch, stale or refuted source, added hypotheses, unsupported use... We hope our framework gives a useful vocabulary for specifying and checking AI systems whose outputs, citations, tool calls, and world-changing actions must be justified by reliable claims and explicit authority rather than apparent fluency.
Authors: Jiayu Yao, Yiwei Wang, Anmeng Zhang, Zhe Sun, Songsong Wang, Lingrui Mei, Yuyao Ge, Shenghua Liu
Abstract: Reinforcement learning (RL) is increasingly used to align multimodal large language models (MLLMs), but higher rewards do not always imply better task performance. This risk is amplified when visual evidence is evaluated by text-only or weakly grounded rewards. We study reward hacking in MLLM RL across safety VQA, chart VQA, and stress-test settings, varying reward design, data ambiguity, model scale (2B-32B), and RL algorithm (GRPO, RLOO, DAPO). We introduce Newly Rewarded Failure Rate (NRFR), which measures failures among samples whose proxy reward improves over the SFT baseline. Outcome-only rewards cause severe hacking, reaching 48.1% Reward Hacking Rate (RHR), while NRFR exceeding RHR shows that RL creates new failures rather than merely inheriting them. Scaling reduces but does not eliminate hacking: even the 32B model retains a 54.9% worse rate under outcome-only rewards, whereas answer-aware rewards improve the oracle trend at every scale. Robustness is also algorithm- and scale-dependent: GRPO is consistently most resistant, RLOO remains vulnerable, and DAPO improves substantially from 2B to 8B. Visual-evidence rewards help only with reliable verification: keyword-based checks increase hacking, while VLM-as-judge semantic verification reduces it. Overall, multimodal reward hacking is a systematic result of optimizing imperfect rewards, and robust alignment requires rewards and verifiers that remain reliable under optimization pressure.
Authors: Sanjana Pedada, Aditya Dhavala, Neelraj Patil
Abstract: Agentic LLM systems that generate code through multi-turn tool use face a fundamental context problem: each session starts from zero, discarding the configuration choices, domain constraints, data schemas, and tool-use patterns that made previous sessions productive. Naively persisting entire conversation histories is token-inefficient and counterproductive: irrelevant context degrades generation quality. We introduce shared selective persistent memory, an architecture that identifies and retains four categories of reusable context (task specifications, data schemas, tool configurations, and output constraints) while discarding session-specific reasoning traces. Crucially, this memory is shared: workspaces encapsulating selective memory can be transferred across users with role-based access control, enabling collaborative reuse without redundant specification. We implement it in a deployed collaborative workspace platform where LLM agents produce, edit, and maintain git-versioned artifacts (dashboards, reports, and data-driven documents) from heterogeneous sources (CSV, SQL, REST APIs, and MCP servers). A complementary zero-token data refresh mechanism decouples generated programs from runtime data, enabling artifact reuse without re-invocation. Across three enterprise scenarios, shared selective persistent memory achieves 96% task completion (vs. 79% without memory and 71% with full history). Zero-token refresh eliminates LLM re-invocation for recurring updates (14x task-time reduction), while summary-driven generation cuts per-invocation token cost by 97x versus raw data injection. A replication on four public datasets confirms generalizability, with zero-token refresh succeeding in 12/12 trials. Notably, naive full-history persistence actively degrades completion by biasing the agent with stale traces, while selective memory outperforms both extremes.
Authors: Chongyu Qu, Can Cui, Zhengyi Lu, Junchao Zhu, Tianyuan Yao, Junlin Guo, Juming Xiong, Yanfan Zhu, Yuechen Yang, Bennett A. Landman, Yuankai Huo
Abstract: Does every cancer patient truly need a complete diagnostic workup for accurate survival prediction? In multimodal clinical oncology, diagnostic modalities follow a clinically mandated order of escalating burden -- from demographics collected at intake to genomic profiling requiring specialized tissue analysis. Current multimodal survival methods either assume all modalities are available or passively handle missing data, but none actively reason about whether acquiring the next modality is justified for a given patient along this ordered workflow. We formulate this as a sequential decision problem and propose SAGEAgent (Sequential Acquisition Guided by Experience), a self-evolving LLM-based clinical agent that decides which diagnostic modalities to acquire for each patient, balancing predictive accuracy against clinical invasiveness. SAGEAgent reasons about each patient's evolving diagnostic state through clinical tools that translate numerical predictions into text, an episodic memory that retrieves similar past cases, and a semantic memory that accumulates reusable decision patterns from experience. Experiments on a glioma cohort combining TCGA-LGG, TCGA-GBM, and BraTS with four diagnostic modalities demonstrate that SAGEAgent achieves competitive survival prediction accuracy while reducing average acquisition burden by 55%.
Authors: Yuan Cao, Haiqian Yang
Abstract: Modern AI systems are increasingly being evaluated for their ability to reason, code, prove theorems, use tools, and long-horizon research tasks. These are powerful capabilities, but they share a structural limitation: the representational frame within which the model operates, including its conceptual vocabulary, the space of admissible solutions it can search, and the criteria by which success is evaluated, is typically fixed and supplied in advance. This paper argues that building stronger intelligent systems capable of open-ended innovation requires additional classes of operations: the creation, stabilization, and reuse of new representational primitives, which alter the space being searched rather than simply searching within it. We characterize the distance between current AI systems and genuinely open-ended intelligence through two gaps. The first is the vocabulary gap, the difficulty of inventing and stabilizing new representational primitives rather than merely recombining existing ones. The second is the verifier gap, the difficulty of judging the value of a new primitive when its full payoff may be visible only after future reuse. We interpret both gaps through a unified framework of intelligence as cognitive discrepancy reduction. By viewing intelligent behaviors as a sequence of cognitive transformations, we distinguish intra-space transformations which operate within a fixed representational frame, from generative transformations which may modify the frame itself. On this basis, we propose a ladder of innovation autonomy and outline several directions for advancing open-ended AI, including objectives that reward useful representational change, persistent memory architectures for invented primitives, and adaptive verification mechanisms capable of evolving alongside the representations they evaluate.
Authors: Jan Gronewald, Andreas Emrich, Nijat Mehdiyev
Abstract: Pre-demolition assessment, the regulated audit process at the heart of urban mining, is an information process in which AI support must serve qualified auditors who remain accountable for the decisions taken. The relevant unit of value is not prediction accuracy alone, but the defensibility of the supported decisions: their legibility, plausibility, sourcing, and contestability. Explainable AI techniques and domain knowledge graphs each address parts of this requirement, and existing taxonomies have catalogued their integration. The literature is descriptively rich but structurally under-specified: what remains less developed is a structural account of why specific integrations produce artefacts neither resource can provide alone. This paper offers a complementarity-theoretic interpretation grounded in the IS resource-based tradition. We propose four consolidated KG-XAI integration modes (Lifting, Constraining, Typing, and Revising), each defined as a typed operation over XAI artefacts and knowledge-graph substrate structures. Each mode unlocks a distinct property of defensibility and contributes to the kind of regulatory artefact pre-demolition assessment demands. A fire-door example from the urban-mining process illustrates the modes using the W3C Linked Building Data stack and valuation extensions.
Authors: Hannah M. Liu, Rhea Saxena, Shiv Asthana
Abstract: The proliferation of agentic AI systems across enterprise and public-sector contexts has outpaced the capacity of general-purpose AI risk frameworks to classify and govern them. In this paper, we introduce the TrustX Agent Risk Classification Framework, a structured, repeatable instrument that can be applied to seven types of agentic AI systems and is grounded in foundational pre-existing AI governance frameworks. At the core of the framework is a twelve-dimension scoring rubric that robustly quantifies the risk. This rubric is combined with other components, such as the GPA + IAT classification model and the five-level autonomy framework derived from existing literature. These inputs produce a three-tier governance output with mapped control recommendations. A specialised Coding Assistant extension is also included to account for nuances specific to this type of agentic AI system. We then use an illustrative example to show our framework in practice. ARC is intended for AI governance practitioners, risk officers, developers, and regulators, and it will regularly undergo iteration as we continue to expand it and make it more robust. The community can access the interactive framework here: https://arc.responsible.ai/
Authors: Kaiji Zhou, Ales Leonardis, Yue Feng
Abstract: Enhancing the reasoning capabilities of large language model (LLM) agents requires effective orchestration of diverse expert models and tools. However, existing frameworks typically call APIs based on coarse-grained matching between tasks and the functions of expert models or tools, while overlooking critical factors such as performance variability and cost efficiency among functionally similar alternatives. To address this, we propose Agora, a framework that introduces an incentive-compatible auction mechanism for dynamically allocating tasks to expert models and tools. By treating reasoning steps as tradeable items, Agora enables agents to bid based on their rectified competence-ensuring that critical logic is routed to the most capable solver rather than the most overconfident one. Evaluations across five benchmarks show that Agora improves over matched single-model, routing, and cascade baselines under comparable candidate pools, while exposing a controllable cost-quality trade-off through a single auction parameter.
Authors: Mohadeseh Mollapour, Koorosh Aslansefat, Zeinab Dehghani, Bhupesh Kumar Mishra, Tejal Shah, Zhibao Mian
Abstract: Concept-based explainable artificial intelligence (AI) can make model reasoning more human-understandable, but concept-level outputs are not automatically trustworthy. We introduce ConceptSMILE, a model-agnostic perturbation-based auditing framework for evaluating the reliability of concept-based explanations. Rather than replacing SMILE, ConceptSMILE extends its perturbation-based logic from feature- or region-level attribution to the auditing of human-understandable concept explanations. The framework perturbs input regions, measures concept-response shifts, applies locality weighting, and fits an XGBoost surrogate to approximate local concept behaviour. Reliability is assessed through attribution accuracy, surrogate fidelity, faithfulness, stability, and consistency. We evaluate ConceptSMILE on retinal fundus images by comparing MedSAM-derived visual concepts with VLM-based semantic concepts. Results show that reliability varies across concepts and pathways: MedSAM achieves stronger spatial attribution and the highest surrogate fidelity ($R^2 = 0.8503$, $R_w^2 = 0.8465$), while the VLM pathway shows stronger vessel faithfulness and stronger stability under selected artefact conditions. ConceptSMILE provides an independent audit layer for evaluating the trustworthiness of concept-based XAI.
Authors: Song-Ju Kim
Abstract: Decision making often exhibits context dependence that challenges classical probability theory. This paper develops a quantum-like extension of the Tug-of-War (QTOW) decision-making model to clarify when such context dependence can be represented by a single minimal internal state. The QTOW construction uses a qutrit internal state, conservation-preserving updates, and measurement-induced disturbance to model decision, learning, and probing operations within one coherent state space. Within this minimal representation, KCBS-type probing contexts can be constructed, yielding a witness of non-contextual classical non-embeddability. The main claim is not that quantum theory is uniquely or assumption-freely derived from decision making. Rather, a classical reconstruction of the same operation family requires additional contextual memory, history dependence, or an enlarged hidden-state representation. Thus, contextual probability appears as a resource signature of minimal decision dynamics, while quantum probability provides a compact, memory-efficient realization of this structure.
Authors: Nicolas Koller, Andreas u. Schmidt
Abstract: Large language models (LLMs) are increasingly applied to reverse-engineering tasks, and recent threat-intelligence reporting shows them operating inside live offensive-security workflows. Claims about their capability, however, outpace our ability to measure it. Existing benchmarks for LLM-assisted binary analysis treat the construction of function-level ground truth as a solved pre-processing step and report accuracy without disclosing how many functions were reliably evaluable. We argue that the principal obstacle to fair evaluation is not model capability but the reliability of binary-to-source alignment under compiler optimization. This paper presents Reforge, a provenance-tracked pipeline that constructs function-level ground truth from C source through compilation, DWARF and syntactic extraction, alignment, and decompilation, and that operationalizes alignment uncertainty as an eight-gate confidence funnel with three-tier stratification. On a controlled micro-benchmark, high-confidence yield falls from 87.2% to 65.9% across optimization levels, and unpaired comparisons overstate optimization-induced performance decay through survivorship bias. A proof-of-concept evaluation of seven contemporary LLMs on function naming demonstrates the substrate and motivates uncertainty-aware benchmarking practice
Authors: Qingzhuo Wang, Ruiyang Qin, Zhenxin Qin, Wen Shen, Zhihua Wei
Abstract: Despite the success of knowledge distillation (KD) in Large Language Models (LLMs), the underlying mechanism behind its efficacy remains unclear. In this paper, we propose a unified approach to explore the common mechanism of various KD methods using interactions. Specifically, we decompose the output score of the LLM into the sum of numerous interactions. Each interaction represents a nonlinear relationship involving a set of input variables (e.g., words). Based on the decomposed interactions, we discover that the common mechanism underlying various KD methods is the sparsification of interactions, i.e., student models retain fewer interactions for inference while suppressing other interactions to zero effects. Furthermore, we discover that the performance variance across different KD methods arises from their capabilities in handling complex interactions. A KD method typically yields better performance if it enables the student model to achieve higher sparsity of complex interactions. Motivated by these insights, we propose a plug-and-play loss function called Complex Interaction Penalty (CIP) to explicitly enforce the sparsity of complex interactions during the distillation process. Extensive experiments demonstrate that integrating CIP consistently improves the performance of diverse KD methods on both in-domain and out-of-distribution benchmarks.
Authors: Farica Zhuang, Seong Woo Han, Zixuan Wen, Shu Yang, Yize Zhao, Li Shen
Abstract: Alzheimer's Disease (AD) is a complex neurodegenerative disorder that continues to impact millions of people worldwide. Predicting AD conversion during the prodromal stage remains critical for disease understanding and patient care. As such, survival models are widely used for AD risk prediction, yet they are typically static predictors with limited interpretability and no capacity for natural language reasoning. In this work, we propose iLENS, an interpretable large language model (LLM) guided framework based on mixture-of-experts (MoE) for survival prediction in AD conversion. Our approach uses LLM to synthesize structured neuroimaging measurements and unstructured information to guide expert routing. Our framework demonstrates competitive predictive performance and capability in patient subtyping. Furthermore, our framework provides transparent, biologically grounded rationales for its routing decisions, bridging the gap between high-performance survival analysis and interpretable clinical decision support.
Authors: Ian Colbert, Eashan Dash, Pablo Monteagudo-Lago, Juan Amboage, Srinidhi N, Giuseppe Franco, Nicholas J. Fraser, Arun Ramachandran
Abstract: The signed integer alphabet contains one more negative representable value than positive. Yet, by convention, the standard symmetric integer quantizer fixes its scale to be strictly positive, which assigns this extra representable value to the negative tail and can force clipping of positive outliers. In this work, we show that, at few-bit precision, such clipping is a non-trivial source of quantization error. Asymmetric quantization addresses this problem with a zero point, shifting the grid toward the observed data range; however, this flexibility is well-known to carry a runtime penalty. For example, in llama.cpp on an AMD EPYC(TM) "Turin" CPU, a 4-bit symmetric format uses up to 9% less memory with up to 2.45$\times$ higher throughput than its asymmetric counterpart. We highlight signed symmetric quantization as a third option that retains the runtime profile of symmetric quantization without the penalty of the asymmetric format: our signed absmax grid places the extra representable value on the dominant-outlier tail through a principled and lightweight sign selection rule while keeping the zero point at zero. Our theoretical analysis offers two main results. First, we establish the signed absmax grid as conditionally bound-optimal on $\ell_2$ quantization error, and show that the condition holds for 88-99% of weight groups across pre-trained large language models (LLMs) at low bit widths. Second, we show that negating the scale of a standard symmetric quantizer is analytically equivalent to a unit zero point shift on the same signed integer alphabet. We empirically validate our proposal on models from the Qwen3, Qwen3.5, and Llama3 families, and observe improvement in perplexity and downstream few-shot accuracy over the standard unsigned symmetric quantizer at no extra inference cost
Authors: Ali Kayyam
Abstract: Mixture-of-Experts (MoE) models activate only a sparse subset of experts per token, yet consecutive tokens frequently activate different experts -- causing constant weight swapping between slow storage and fast memory on edge devices. Existing remedies are either system-level (caching heuristics) or post-hoc (router fine-tuning), leaving the root cause unchanged during pretraining. We propose StickyMoE, a differentiable routing consistency loss that penalises abrupt expert switches between adjacent tokens, encouraging the router to maintain the same expert assignment across semantically coherent spans. StickyMoE requires no architectural changes, adds a single hyperparameter lambda, and unlike post-hoc methods, allows expert representations and routing decisions to co-adapt from the first training step. Experiments on small-scale MoE language models show that StickyMoE reduces the expert switch rate by up to 60% with less than 4% perplexity degradation, Pareto-dominating post-hoc fine-tuning on the quality-locality frontier. Routing temporal locality is most efficiently instilled at training time.
Authors: Kehan Guo, Yili Shen, Yujun Zhou, Yue Huang, Chujie Gao, Shiyi Du, Xiangliang Zhang
Abstract: The coupling in flow matching -- the rule pairing noise vectors with data points -- is typically treated as a computational choice. We show that this coupling can instead serve as an alignment interface: by matching noise and data according to a target molecular property, it embeds controllable structure directly into the learned flow field. Building on this view, we introduce Reward Transport, which uses optimal transport coupling at training time to align a scalar noise-space coordinate with molecular rewards; at inference, varying this coordinate steers the generated distribution without requiring an oracle, reward model, gradient guidance, or additional computation. In the coupling-preserving limit, thresholding this coordinate recovers the Cross-Entropy Method's truncated reward distribution, providing a principled, continuously adjustable distribution-level control knob. Empirically, on ZINC-250K and GuacaMol, sweeping the scalar induces monotone control of logP and consistent QED control over its operating range; most tellingly, the same knob produces opposite structural responses for different targets, growing molecules for logP but shrinking them for QED, which rules out a generic size bias. The interface is complementary to classifier-free guidance and conditional flow matching, while a negative result under epsilon-prediction diffusion clarifies where coupling-level alignment is structurally absent. Code: https://github.com/KehanGuo2/reward-transport
Authors: Qianli Liu, Kaibin Guo, Zicong Hong, Peng Li, Fahao Chen, Haodong Wang, Jian Lin, Song Guo
Abstract: Expert parallelism has become the prevailing paradigm to serve Mixture-of-Experts (MoE) models. Its efficiency depends on the communication and computation latencies of the GPUs, which are linked to the placement of experts in the GPUs. Existing works for optimizing expert placement focus on leveraging past requests' expert activation patterns. However, they demonstrate deficiencies facing diverse and rapidly changing request patterns, calling for an online, proactive approach. Implementing such an approach requires addressing several challenges: the uncertainty associated with incoming requests' expert activation, the cost of expert migration, and the NP-hard complexity in optimization. Therefore, we present Director, a new distributed MoE serving system that minimizes end-to-end latency via prediction-driven, online expert placement. Director uses either a lightweight cascaded predictor or a low-bit quantized replica for expert activation patterns of incoming requests. An online migration module then enacts the changes with near-zero downtime by executing migrations in compute-bound phases, keeping disruption bounded. At its core, a relaxation-based expert placement optimizer operates under capacity constraints, runs in polynomial time, and achieves a $(1+\epsilon)$ approximation ratio. Finally, we implement a prototype and demonstrate, through extensive experiments, a reduction in end-to-end latency of $11\sim55\%$ for popular MoE models (e.g., Mistral, DeepSeek and Qwen) compared to existing work.
Authors: Ziheng Chen, Yue Song, Rui Wang, Xiao-Jun Wu, Nicu Sebe
Abstract: Manifold-valued measurements are prevalent in various machine learning tasks. Recent advances have extended Deep Neural Networks (DNNs) to operate on manifolds, accompanied by normalization techniques tailored to different geometries, collectively referred to as Riemannian normalization. However, most existing Riemannian normalization methods are either designed for specific manifolds or fail to effectively normalize manifold-valued sample distributions. To address these limitations, we propose LieBN, a framework for Riemannian Batch Normalization (RBN) over Lie groups. Our approach leverages the theoretically convenient left- and right-invariant metrics, which naturally exist in every Lie group, and provides theoretical guarantees for controlling the Riemannian mean and variance. We instantiate LieBN across nine distinct geometries: four on the Symmetric Positive Definite (SPD) manifold, one on the group of rotation matrices, and four on the manifold of full-rank correlation matrices. Notably, among the SPD metrics, we introduce a novel right-invariant metric and extend three existing Lie group structures via matrix power deformation. Extensive experiments on different manifolds validate the effectiveness of our framework. The code is available at https://github.com/GitZH-Chen/LieBN.git.
Authors: Thinh T. H. Nguyen, Le-Tuan Nguyen, Minh-Duong Nguyen, Nhi Trinh, Anh Tran Nam Nguyet, Dung D. Le, Kok-Seng Wong
Abstract: Federated continual learning (FCL) evaluates how distributed clients learn from changing data streams while retaining previously learned knowledge. Existing evaluations are difficult to compare because they often change datasets, task splits, client data splits, task orders, backbones, memory assumptions, and reporting rules simultaneously. We introduce \textbf{HERO}, a heterogeneity-aware benchmark library for FCL. HERO builds benchmark streams by separating three choices that are often coupled, namely the task split, the client data split, and the client task sequence. In HERO-Core, the main comparable benchmark, $\alpha$ controls client data skew and $\rho$ controls task-order mismatch. We evaluate representative FCL methods on CIFAR-100 and TinyImageNet using final average accuracy, average forgetting, and bottom-10\% client accuracy. We also include a graph-based Domain-IL portability case study on OGB-MolPCBA, where scaffold-domain granularity changes the input distribution while the prediction task remains fixed. Our results show that method behavior changes across easy and heterogeneous settings, that average accuracy can hide weak bottom-client performance, that task-order mismatch favors different strategies from synchronized evaluation, and that the same HERO interface can expose domain-shift difficulty beyond image-based FCIL. HERO releases benchmark streams, configurations, method implementations, and reporting scripts to support reproducible and setting-aware FCL evaluation.
Authors: Qiheng Sun, Hongwei Zhang, Junxu Liu, Xiaokai Mao, Jinfei Liu, Kui Ren, Haibo Hu
Abstract: High-quality data drives machine learning advances across industries. Recognizing the value of data, data transactions are increasingly common, giving rise to many data marketplaces, e.g., AWS Marketplace, Databricks, and Datarade. However, determining the appropriate prices for data products remains a significant challenge due to the unique properties of data products. Traditional pricing methods in economics can be categorized into the cost approach, the income approach, and the sales comparison approach. The cost approach fails in data pricing due to near-zero marginal cost from data replication, and the income approach fails due to inherently unpredictable data revenue. The sales comparison approach remains viable, yet its application is hindered by the absence of standardized pricing benchmarks for data products across marketplaces. To address this challenge, we introduce \texttt{DaDaDa}, the first dataset for data product pricing, containing metadata for 16,147 data products from 9 major data marketplaces worldwide. \texttt{DaDaDa} enables the training of pricing models, thereby establishing price benchmarks for new data products. In addition, \texttt{DaDaDa} can be utilized for other important tasks in data markets, such as data product classification and retrieval. Experiments and a retrieval prototype demonstrate the effectiveness of \texttt{DaDaDa} for pricing, classification, and retrieval of data products. The dataset and code are available at https://github.com/ZJU-DIVER/DaDaDa.
Authors: Tao Lu, Haoyu Wang, Zonghui Wang, Keshen Xiang, Jiaheng Zhang, Wenzhi Chen
Abstract: With the growing deployment of large language models (LLMs), LLM inference cost has become a key challenge. Pruning techniques that introduce sparsity into weight matrices can accelerate inference. However, maintaining model quality typically limits pruning to moderate unstructured sparsity (around 50\%). At these sparsity levels, none of the existing GPU kernels for sparse matrix multiplication (SpMM) can outperform their dense counterparts. This paper proposes an efficient GPU inference method for LLMs with moderate sparsity. We propose a three-layer matrix storage format comprising: (i) a Sparse-TC layer enabling sparse tensor cores to accelerate SpMM; (ii) a Slot-Filling layer using parallel differential distance for matrix compression while supporting low-cost on-chip decoding; (iii) a lightweight Residual Layer ensuring correct SpMM computation. Building on this format, we design a SpMM kernel that jointly utilizes sparse tensor cores and CUDA cores. This design enables an efficient execution pipeline and overlaps on-chip computation with memory access. Evaluations show that our work is the first to outperform dense matrix multiplication on modern GPUs equipped with high-bandwidth memory (HBM). It achieves up to 1.64x kernel-level speedup over SpInfer (EuroSys'25, Best paper) and up to 1.41x end-to-end speedups over FlashLLM (VLDB'24). Our source code: https://github.com/moui0/cudac.
Authors: Georgios Laskaris, Reuben Brasher, Niki van Stein, Elena Raponi, Thomas B\"ack, Florian Neukart
Abstract: Designing effective multi-objective Bayesian optimization (MOBO) algorithms requires balancing many interdependent design choices whose optimal configuration is problem-dependent and typically demands deep expertise. We extend the LLaMEA framework to MOBO, using large language models as mutation and crossover operators within evolutionary strategies to generate complete algorithm implementations, with SMAC hyperparameter optimization integrated into the evolutionary loop. Across nine evolutionary runs we generated approximately 900 algorithms and benchmarked them on twelve synthetic problems (ZDT, DTLZ, WFG) and three real-world engineering problems (RE), using a BoFire qParEGO implementation as a state-of-the-art Bayesian-optimization baseline. On the synthetic suite the strongest generated algorithm attains the highest mean normalized hypervolume (0.971, vs. 0.869 for qParEGO) while requiring roughly 60x less wall-clock time; a Friedman test with post-hoc analysis places the two in a single top-performing group, and per-problem tests find the generated algorithm significantly better than qParEGO on 7 of the 12 problems and never worse, matching state-of-the-art accuracy at an order-of-magnitude lower cost. On the three unseen real-world engineering problems a generated algorithm attains the best mean normalized hypervolume (0.985, vs. 0.971 for qParEGO)--significantly better than qParEGO on two of the three problems--at roughly 3.4x lower wall-clock cost, confirming that the gains transfer beyond the synthetic regime. LLM-driven evolutionary search can thus discover algorithm designs that achieve Pareto-efficient trade-offs difficult to reach through manual design.
Authors: Joshua Pickard, Wei Qi, Na Li, Ann Woolley, Lisa Cosimi, Roy Kishony, Deborah Hung
Abstract: Sepsis is a leading cause of mortality, yet optimal treatment policies remain contested. Existing reinforcement learning (RL) approaches learn fixed strategies for sepsis treatment, limiting adaptability to changing clinical objectives during inference. We propose EHRMPC, a framework that decouples learning patient dynamics from optimizing treatment by training a patient digital twin in the form of a generative electronic health record (EHR) model. The digital twin predicts clinical trajectories under interventions and enables model predictive control (MPC) to optimize treatments via inference-time planning over simulations. We evaluate EHR-MPC on a multicenter ICU sepsis cohort spanning 8 hospitals in the Mass General Brigham health system using both off-policy importance sampling and on-policy simulation-based evaluation. Relative to RL baselines, EHR-MPC achieves comparable off-policy performance and improved simulation performance. Unlike RL, this work frames sepsis treatment optimization as inference-time control over learned patient dynamics, establishing a general framework for decision making with generative clinical models.
Authors: Padam Jung Thapa, Abdullah Bin Naeem, Ayon Dey, Anav Katwal, Md Tamjidul Hoque
Abstract: Sand boils on earthen levees are safety-critical defects, but pixel-level detection is limited by scarce annotations. We present a diffusion-based synthesis pipeline for low-resource sand-boil imagery. Using Stable Diffusion XL fine-tuned with DreamBooth and conditioned by a multi-branch ControlNet stack, the pipeline generates synthetic inspection images from a small curated reference set. A soft-mask inpainting protocol preserves the real defect pixels while re-rendering the surrounding scene, avoiding seams and color shifts from prior seamless-cloning compositing. A mask-conditioned ControlNet can also generate a new boil inside a chosen mask, making the mask the segmentation label by construction; however, because large-scale label certification remains unresolved with the available real-trained gate, we release the soft-mask preset as the default. Text conditioning is supplied by a taxonomy-driven Prompt Atlas that expands one domain specification into a stratified, CLIP-validated prompt bank and transfers to new defect classes without code changes. From the real training images, the pipeline produces 1,020 synthetic candidates, of which 815 pass a CLIP admissibility filter. We evaluate image quality using distributional and fidelity-diversity measures against the real reference set and a Poisson baseline, and audit for out-of-distribution drift and memorization. No single preset dominates; each trades off fidelity, diversity, and label reliability. We therefore release the label-reliable preset as the default and treat a curated mixture as the natural augmentation set. Our claims are limited to image quality, label provenance, and diversity; downstream segmentation is left for future work. Code and an artifact manifest are released for reproducibility.
Authors: Hyunjin Seo, Hyeon Hwang, Gyubok Lee, Jay Shin, Jimin Park, Taesoo Kim, Sanghoon Lee, Hongjoon Ahn, Sungjun Han, Sangwon Jung
Abstract: The push toward large language models for biology (BioLM) has created a need for training corpora that can endow models with a genuine understanding of biology. However, existing biological resources, such as molecular databases, protein repositories, genomic annotations, single-cell atlases, and pathway databases, are scattered across heterogeneous formats and remain unorganized into a cohesive corpus for language model training. We present TheBioCollection, a 52.6B-token pre-training-scale corpus that converts these disparate resources into a unified, training-ready form spanning small molecules, proteins, genomic sequences, cells, and pathways. Beyond consolidating existing data, TheBioCollection enriches each record with tool-computed biological properties and introduces new instruction tasks for capabilities that current corpora barely cover. We pair the corpus with TheBioCollection-Eval, a matched suite probing recognition, generation, and prediction across molecular, protein, genomic, cellular, and cross-domain settings. Holding the base Gravity-16B-A3B architecture fixed, training on TheBioCollection more than doubles its overall score on TheBioCollection-Eval with gains in every domain, while leaving general linguistic ability nearly intact.
Authors: Sunshine Jiang, John Marangola, David Zhang, Raghuram Kowdeed, Ruiyang Luo, Nitish Dashora, Richard Li, Pulkit Agrawal, Zhang-Wei Hong
Abstract: Exploration is essential to RL since a policy cannot improve by repeatedly sampling the behaviors it already prefers. Standard methods inject stochasticity in the action space, but such jitter only yields rollouts close to the original. Escaping a weak policy often requires global perturbations that action noise cannot produce. Large language models (LLMs) and vision-language-action (VLA) models offer a pathway: they condition the policy on a natural language prompt, and since the rollout follows from it, modifying the prompt induces global changes. The challenge is finding prompts that induce useful global changes. With a weak policy that rarely succeeds, reward is too sparse to select on. Our idea is to refine prompts from the rollouts themselves: a vision-language model (VLM) reasons over the rollout video, diagnoses how the policy responded, and rewrites the prompt to elicit better behavior next time. This procedure realizes posterior sampling, a classical RL exploration framework, at the level of prompts: the VLM maintains an implicit distribution over useful prompts and updates it from observed rollouts. We call this strategy Prompt-Driven Exploration (PDE). Across manipulation and reasoning tasks, PDE enables RL to learn successful policies even from zero-reward starts, and improves sample efficiency more broadly. Our website is available at https://xinyunsunshine.github.io/prompt-rl.
Authors: Lawrence Nguyen, Hiu Yung Wong
Abstract: This work presents a study of an implementation of a novel Quantum Convolutional Neural Network (QCNN) for binary classification of images from the Modified National Institute of Standards and Technology (MNIST) dataset. Using a novel architecture inspired by previous QCNN and classical convolutional neural network (CNN) implementations, we use a hierarchical partitioning approach to implement a QCNN circuit that can be approximated and simulated efficiently on a classical machine for a large problem. First, the original image is partitioned such that each process handles a smaller portion of the image, which is encoded into independent states. Then, these partitions merge and combine, resulting in states that contain information from both partitions while halving the number of processes. After repeating this until one process remains, we reduce the dimensionality of the state until a single qubit remains for measurement. Using this approach, we can use multiple processes in parallel to simulate a large QCNN program without the need for exponentially growing hardware requirements as the number of qubits increases. In our work, we use this scheme to train a 128-qubit model, which is impossible to run on any classical supercomputer without the novel architecture. We also explore the impact of this new model architecture on prediction accuracy by training it to perform binary classification on the MNIST dataset with a small number of qubits, and comparing it to a model without partitioning. Our initial findings show that partitioning images into smaller sub-images with this architecture does not degrade the model's performance and sometimes even improves it, likely because it reduces the Barren plateaus issue in the partitioning process.
Authors: Ning Liu, Kalle Kujanp\"a\"a, Zhaoxuan Zhu, P Aditya Sreekar, Kaiwen Liu, Chuanneng Sun, Jorge Marchena Menendez, Matthew Bales, Tianyu Yang, Shahnawaz Alam, Rose Yu, Baoyuan Liu, Kristina Klinkner, Shervin Malmasi
Abstract: Warehouse operations are governed by Standard Operating Procedures (SOPs) that encode complex, multi-system decision logic, which must be executed reliably under strict time constraints, yet LLM agents lack mechanisms to enforce procedural compliance and degrade under the context overload full SOP specifications introduce. We present Eluna, a production-deployed agentic system for reliable SOP execution. Eluna is a graph-guided, multi-agent framework that encodes SOPs as directed acyclic graphs with progressive disclosure and delegates independent tasks to parallel sub-agents, each with persistent code execution and live data access. To meet production latency and accuracy needs, we use asymmetric episodic distillation where a strong teacher is improved through episodic error memories, then a smaller student is fine-tuned on the corrected trajectories with memory stripped, internalizing corrections without inference-time overhead. On a 13-task benchmark and two production applications, our fine-tuned models match or exceed their teacher, beat all larger off-the-shelf baselines, and reach 94% expert agreement on the ticket processing application.
Authors: Berkay Anahtarci
Abstract: Large language models increasingly provide labels, evaluations, and feedback for tasks specified in natural language. When a specification admits multiple readings but the supervision channel does not reveal which is operative, additional labels reduce sampling error without resolving the resulting identification problem. We introduce Natural Language PAC (NL-PAC), a framework that uses a fixed model's thresholded decoding law to define admissible labels and candidate targets. The probability that multiple labels are admissible equals the diameter of the pointwise-admissible target class, and under target-blind supervision every learner incurs worst-case risk of at least half this diameter, at every sample size; the exact randomized minimax risk over this class is attained by a data-independent strategy. Finite-sample confidence bounds make these quantities certifiable from held-out unlabeled inputs. In a frozen Qwen~2.5--3B audit, one prespecified prompt yields a positive model-relative certificate, whereas a paraphrase and exact-rule controls yield zero. A held-out bridge audit finds that supplied candidate reading clauses fail the admissibility condition needed to transfer the certificate to coherent readings. The guarantee is specific to the audited model, prompt, threshold, and input distribution; extending it to human interpretations requires external validation.
Authors: Hantao Zhang, Jinru Sui, Ed Li, Dirk Bergemann, Zhuoran Yang
Abstract: Recent benchmarks for VLMs largely assess single- or limited-view perception, leaving untested the core cognitive ability to integrate observations across viewpoints into a coherent, world-centric (allocentric) 3D mental model. We introduce MultiView-Bench, a diagnostic benchmark expressly designed to evaluate multi-view integration for holistic 3D scene comprehension. Unlike existing datasets that focus on pixel-level mapping or camera-relative navigation, MultiView-Bench requires models to decouple object positioning from transient perspectives and ground them in a fixed global coordinate system. This capability serves as a prerequisite for VLMs before being deployed for downstream tasks such as mechanical part assembly. Our systematic evaluation of frontier VLMs reveals consistent failure modes: strong performance on 2D planar relations from a single image, but marked difficulty with 3D spatial relations and with aggregating information across views. We further identify biases in VLMs, such as struggles with unconventional axis directions and sensitivity to object colorways and texture variations. Acknowledging these limitations, we propose ViewNavigator, a multi-agent framework that actively selects informative viewpoints, perceives, and fuses multi-view evidence, improving diverse base models on MultiView-Bench even under a strict budget-matched comparison (and by 3-5x for the full agent).
Authors: Yuri Ishitoya, Jeremy Siburian, Masashi Hamaya, Kuniaki Saito, Cristian C. Beltran-Hernandez, Mai Nishimura
Abstract: Vision-language-action models (VLAs) inherit semantic capabilities from pretrained VLMs, yet large-scale post-training on robot data and architectural modifications can reshape the backbone so extensively that it becomes difficult to isolate what the VLM contributes to control. Directly converting pretrained VLMs into VLAs with minimal architectural change offers a more transparent path to understanding how VLM capabilities transfer across model scales. The core obstacle is output-distribution mismatch: predicting actions as bare numeric token sequences moves generation away from the VLM's pretrained language distribution, degrading the capabilities we seek to preserve. To address this, we propose CLAP (Causal Language-Action Prediction), which prepends each numeric action sequence with a natural-language action description, causally conditioning precise action-token prediction on a language-action plan without modifying the backbone architecture. With single-epoch fine-tuning alone, 2B CLAP achieves 90.8% on LIBERO (+14.9 pt over VLA-0) and improves robustness on LIBERO-PRO under language, object, and spatial perturbations. We will release CLAP at 0.8B, 2B, and 4B as an open-weight, multi-scale compact VLA family from a single VLM lineage, enabling controlled analysis of VLM-to-VLA capability transfer.
Authors: Viraaji Mothukuri, Reza M. Parizi
Abstract: LLM-generated code often compiles, passes tests, and appears correct, yet breaks once deployed. The root cause is frequently structural rather than logical. A generated endpoint references configuration keys never declared in the project, an import targets a package that does not exist in any registry, or a new route omits the authentication guard applied to every sibling endpoint. Each patch is locally valid but globally incoherent, and standard CI toolchains rarely surface these failures. As LLM-powered coding tools see widespread adoption, this blind spot poses a growing risk to software quality. We call this the \textbf{patchwork problem}. This paper formalizes structural coherence as consistency invariants over graph representations of repository artifacts, including import, call, dependency, configuration, schema, resource, control-flow, and routing graphs, and introduces an eight-category failure taxonomy distinguishing defects specific to LLM generation from those merely amplified by it. We present a hybrid verification framework that delegates to mature static analysis tools where they already excel and deploys purpose-built detectors for cross-cutting invariants underserved by existing toolchains, targeting provable constraint violations rather than heuristic pattern matching. Empirical evaluation across two frontier models under four prompting strategies reveals that the vast majority of structural failures evade type checking, testing, and SAST entirely, and that failure patterns diverge qualitatively between models in ways that challenge model-agnostic mitigation strategies. External validation on real-world AI-generated repositories confirms that these failures are not artifacts of controlled experimentation but are prevalent wherever LLMs write code with minimal human oversight.
Authors: Sijia Gu, Noor Nashid, Ali Mesbah
Abstract: While autonomous coding agents have significantly advanced automated test generation, they remain fundamentally limited by lazy generation, a phenomenon where agents prematurely terminate tasks and systematically avoid complex programmatic logic, resulting in inadequate code coverage. Currently, mitigating this premature termination requires continuous human-in-the-loop supervision. This heavy reliance on human intuition creates a bottleneck that negates the efficiency gains of automated generation. We propose SCATE, a framework for adaptive, automated supervision of coding agents that replaces human intervention during test generation. By formulating supervision as a contextual bandit problem, SCATE learns to select the most promising testing actions based on the current coverage and class testability metrics, maximizing coverage gains while minimizing wasted generation effort. Our empirical evaluation demonstrates that SCATE integrates seamlessly with different coding agents. When applied to GEMINI-CLI, it achieves 32.3% higher line coverage and 30.9% higher branch coverage than the agent-only baseline. A comparison with CLAUDE CODE confirms the framework dynamically adapts its policy to optimize each agent's unique strengths. SCATE also consistently outperforms state-of-the-art non-agentic approaches across all metrics.
Authors: Brent Kong, Tejas Ram, Tony Yue Yu
Abstract: AlphaZero has demonstrated that a neural-guided Monte Carlo Tree Search can achieve superhuman performance, but strong play does not necessarily imply perfect play. We study this gap in two oracle-evaluable domains with contrasting structure: Connect Four, a solved partisan game with exact game-theoretic values, and Chomp, an impartial game whose optimal play is governed by Grundy-number structure. Under a unified self-play $+$ MCTS pipeline, we compare vanilla AlphaZero, a multi-frame variant (limited to Chomp), and an AlphaZero Auxiliary Loss (AZAL) that adds oracle-derived policy supervision. We find that vanilla AlphaZero achieves strong play across both domains but cannot preserve the exact trajectories required for optimal play: in Connect Four, it fails to maintain the optimal line of play, while in Chomp, it fails to consistently restore the $g=0$ invariant. On rectangular Chomp boards, multi-frame inputs alone do not remove this gap. Nevertheless, AZAL substantially improves oracle consistency across multi-seeded full-game traces and sampled-state evaluations. On Chomp, AZAL reaches perfect full-game oracle consistency on 10x11 and high but not complete consistency on 9x10; on Connect Four, AZAL improves oracle-match rate and delays the first oracle mistake, but does not reach perfect play.
Authors: Shrimon Mukherjee, Kishalay Das, Partha Basuchowdhuri, Pawan Goyal, Niloy Ganguly
Abstract: Graph Neural Networks have emerged as a powerful tool for the fast and accurate prediction of various crystal properties. These models often encode domain-specific knowledge into their graph encoding modules, which increases their parameter size and makes their performance heavily dependent on domain expertise. Added to this, explicitly incorporating all chemical and structural features, that might influence a specific crystal property into the GNN encoder, is a challenging task. In this work, we propose a soft prompt learning framework that captures latent features essential for property prediction, which are not explicitly provided to the GNN. We introduce a novel multilevel graph prompt learning framework comprising both node-level and graph-level soft prompts. At the node level, we capture the local chemical semantics of different atom types, while at the graph level, we encode the global structural symmetry of the crystal graph. Our proposed prompt learning framework is lightweight and seamlessly integrates with any existing GNN encoder. Extensive experiments on popular benchmark datasets show that incorporating prompt learning significantly improves (3\% - 15\%) the performance of state-of-the-art GNN models in crystal property prediction tasks. Furthermore, the learned soft prompts enable cross-property knowledge transfer, enhancing prediction performance for properties with limited training data. Code is available at https://github.com/shrimonmuke0202/Prompt.git
Authors: Ajay Narayanan Sridhar, Ronak Singh, Mehrdad Mahdavi, Vijaykrishnan Narayanan
Abstract: We study contextual bandit problems with correlated arms and access to surrogate reward signals produced by a machine learning model, motivated by applications such as large language model (LLM) routing. Unlike classical contextual bandits that rely solely on bandit feedback and assume conditional independence across arms, our setting allows context-dependent inter-arm correlations and auxiliary reward information that may be noisy or misspecified. We propose algorithms that leverage such surrogate rewards through two complementary designs. A coupled reward-mixing approach pools true and surrogate rewards to accelerate learning when surrogate signals are reliable, while a decoupled prediction-mixing approach maintains separate estimators for bandit feedback and surrogate rewards and adaptively combines their predictions. This decoupling yields robustness to surrogate misspecification, recovering regret guarantees comparable to reward-only bandit methods in the worst case, while achieving improved regret when surrogate predictions are sufficiently informative. We provide theoretical regret analyses for both approaches and evaluate them on LLM routing benchmarks under varying accuracy versus cost trade-offs. The results demonstrate improved sample efficiency and consistently better accuracy-cost trade-offs compared to standard contextual bandit baselines and strong static routing methods.
Authors: Shikhar Bharadwaj, Kwanghee Choi, Stephen McIntosh, Chin-Jou Li, Eunjung Yeo, Daisuke Saito, Nobuaki Minematsu, Shinji Watanabe, Jian Zhu, David Harwath, David R. Mortensen
Abstract: Phone segmentation and recognition are inherently related tasks, yet modern approaches typically model them separately. We argue that phonetic structure is already latent in the representations of self-supervised speech models (S3Ms), and one only needs to steer them to solve both tasks. We leverage S3M-based Phonological Activation Mapping (SPAM), which maps each S3M representation frame to a vector of phonological feature activations, such as voicing and nasality. On top of SPAM, we introduce two simple but effective lightweight, gradient-descent-free prediction heads: a recognition head and a segmentation head. Our method requires less than a minute of phonetic transcriptions, and generalizes to unseen phones during training. Across a diverse range of datasets, our approach attains strong segmentation and recognition performance.
Authors: Letian Wang, Chuhan Zhang, Rishabh Kabra, Jasper Uijlings, Steven Waslander, Andrew Zisserman, Joao Carreira, Kaiming He, Misha Andriluka, Eduard Gabriel Bazavan, Andrei Zanfir, Cristian Sminchisescu
Abstract: Driven by next-token prediction, NLP shifted from task-specific models into powerful generalist foundation models. What, then, is the equivalent catalyst needed to achieve a general-purpose model in computer vision? In this paper, we contend that large-scale text-to-video generation serves as a strong pre-training paradigm for computer vision, providing the necessary spatiotemporal priors, vision-language alignment, and scalability required for general visual intelligence. We introduce GenCeption, which leverages a pre-trained video generative diffusion backbone to define a feed-forward perception model, capable of performing various vision tasks steered by text instructions. Empirical results demonstrate that GenCeption achieves state-of-the-art performance across a diverse suite of tasks, including depth, surface normal, and camera pose estimation, expression-referring segmentation, and 3D keypoint prediction, often matching or surpassing specialized models (e.g. DepthAnything3, SAM3, D4RT, VGGT-Omega, Sapiens, David, Genmo, and Lotus-2). Furthermore, the video generative pretrained backbone outperforms alternative pretraining paradigms (e.g., V-JEPA, and Video MAE) under comparable settings. Importantly, GenCeption exhibits preliminary data and model scaling properties along with exceptional data efficiency, where it achieves comparable performance with leading models like D4RT and VGGT-Omega with 7 to 500 less training data. Finally, GenCeption also exhibits intriguing emergent behaviors: a model trained exclusively on synthetic human videos generalizes to real-world footage and out-of-distribution object categories (e.g., animals and robots). These findings suggest that video generation is not merely a synthesis tool, but a foundational path toward generalist vision intelligence for the physical world. Project page: https://genception.github.io
Authors: Chao Wang, Lingling Li, Fang Liu, Licheng Jiao
Abstract: Artificial intelligence (AI) is shifting scientific discovery from task-specific workflows towards autonomous systems that organize exploration with experimental and human feedback in open-ended candidate spaces. Evolutionary computation (EC) provides a computational basis for feedback-driven discovery because population-based search can maintain diverse scientific candidates while steering exploration through accumulated evidence. However, EC predominantly focuses on candidate refinement for predefined problems, whereas cumulative discovery requires experience retention. To bridge this gap, this review introduces evolutionary intelligence (EI) for scientific discovery. EI characterizes scientific AI systems that sustain exploration by linking candidate refinement with experience retention across evolutionary cycles. We introduce a five-dimensional analytical framework that asks what evolves, how candidates change, why candidates are selected, where feedback originates, and when evolution occurs. This framework clarifies how EI transforms isolated search trajectories into cumulative scientific insight. We further demonstrate this paradigm across diverse discovery modes, from evolving concrete scientific entities to orchestrating automated research workflows. Finally, we identify critical bottlenecks regarding evaluation, process traceability, and shared infrastructure, providing a concrete roadmap for advancing the transition from EC to EI in scientific discovery.
Authors: Haruki Emori, Atsushi Iriki, Andrei Khrennikov, Kazunori Kondo
Abstract: Quantum logic is usually presented as a non-classical departure from ordinary reasoning forced on us by quantum mechanics, with classical logic kept as the secure starting point. We argue for the opposite order of explanation in a finite and fully computable setting. The free orthomodular lattice on two generators has ninety-six elements, the direct product of a six-element non-distributive factor and a sixteen-element Boolean factor. Reading the first factor as a register of contexts and the second as Boolean content, we obtain a calculus whose elements are context--bit-vector pairs and whose operations act component by component. With this calculus we establish three results. First, we classify the six layers by commutativity, identifying the central kernel of context-neutral propositions together with a dual central layer in which all complementary contexts are present. Second, we show that orthocomplementation rearranges the layers exactly as the complementation of the small factor rearranges its elements, which makes the duality among the layers rigid rather than accidental. Third, we prove that the operation forgetting the context is a surjective homomorphism of orthocomplemented lattices whose quotient is the classical Boolean algebra, so that classical logic is a six-to-one, information-losing image of the contextual calculus.
Authors: Pulkit Madan, Sanjay Haresh, Reza Ebrahimi, Sunny Panchal, Apratim Bhattacharyya, Roland Memisevic
Abstract: A striking feature of the human visual system is that it ingests visual information through a series of local foveated glimpses, rather than a single global computation. This makes human vision distinctly different from most popular computer vision models in use today, which input images globally and in a single shot. A natural question therefore is whether local, sequential vision models may provide any fundamental computational benefits in addition to being biologically more plausible than global models. In this work, we investigate this question from the perspective of visual state tracking and length generalization. Inspired by recent studies of length generalization in language models, we study the behavior of vision models trained on simple vision tasks that require the aggregation of local information across an image. Our experiments reveal that, similar to language models, vision models can learn to exploit global shortcuts and thereby fail to generalize over task length or complexity. We also show that recurrent vision policies based on strictly local perception can mitigate these failures, thereby allowing models to generalize on these tasks. Our results show that local attention may be an essential overlooked requirement for robust compositional generalization.
Authors: Jialun Cao, Xinru Yan, Songqiang Chen, Yaojie Lu, Zhongxin Liu, Shing-Chi Cheung
Abstract: Software engineering (abbrev. SE) has continuously evolved through increasingly powerful forms of reuse, from source code and libraries to components and services. Recent advances in AI agents have introduced a potentially new reusable artifact: skills. Emerging agent skill repositories and marketplaces enable developers to package, share, and reuse SE expertise as reusable skills. This trend raises a fundamental question: what SE activities are being encapsulated into reusable skills? Existing studies primarily focus on a broad range of skills acquisition, safety, or benchmarking, while lacking a systematic understanding of SE-specific skills and their coverage across the software development lifecycle. To address this gap, we conduct the first large-scale empirical study of SE skills in public repositories and marketplaces. We collect and analyze a large corpus of SE skills, examining the activities they encapsulate, lifecycle coverage, evolution characteristics, and evaluation mechanisms. Our findings reveal that SE activities are increasingly becoming reusable artifacts via skills and suggest promising research opportunities for skill recommendation and engineering-oriented structuring, as well as the need for mechanisms to encapsulate high-context SE activities into reusable skills. Overall, our study provides the first activity-centric characterization of SE skills and reveals how SE activities are increasingly being transformed into reusable skills. These findings offer new insights into skill reuse, ecosystem development, and the future of agent-centric SE.
Authors: Yang Chen, Yunwen Li, Yufan Shen, Minghao Liu, Tianyu Zheng, Bin Fu, Qunshu Lin, Zhi Yu, Botian Shi
Abstract: Recent advancements in LVLMs necessitate robust benchmarks for complex, visually grounded reasoning. A critical limitation is identified in many document understanding benchmarks: visual content is often reducible to text, enabling high performance without genuine visual grounding. To address this limitation, OmniMapBench is introduced to foster visual-centric reasoning for map documents. The benchmark comprises 2,096 manually annotated question-answer pairs across 1,603 map documents from nine categories. It is designed to probe a hierarchy of skills, ranging from perception to multi-step visual reasoning. To quantify benchmark properties, a simple yet effective benchmark-level metric is proposed: the Visual Dependency Index (VDI), defined as the accuracy drop when images are replaced with question-agnostic descriptions. OmniMapBench exhibits higher VDI than established benchmarks, which quantitatively validates its focus on irreducible visual reasoning. Comprehensive evaluations of 25 leading LVLMs are conducted on OmniMapBench. A significant performance gap is observed, with the top-performing model achieving only 75.03\% accuracy. This result underscores the challenges posed by OmniMapBench to current LVLMs. This work aims to catalyze progress in visual-centric reasoning for document understanding of LVLMs. The dataset and code are publicly available at https://github.com/SIGMME/OmniMapBench.
Authors: Devanshu Verma, Vasudha Bhatnagar, Vikas Kumar, Balaji Ganesan
Abstract: Legal precedent retrieval is a fundamental task in legal case preparation, planning, litigation strategy, and legal research. Current approaches for automatic precedent retrieval map legal documents to a low-dimensional semantic space and compute similarity based on the proximity of their representations. These approaches treat legal documents as monolithic texts, ignoring the rhetorical organization of the legal technicalities. Ergo, they overlook nuanced legal meanings and fail to distinguish the contextual significance of legal entities and concepts that vary based on their rhetorical roles within the document. To address this insufficiency, we propose the PRecG pipeline that computes the similarity between pairs of legal judgments by hierarchically learning their representations. The process begins by decomposing each document into distinct semantic units (segments) based on the rhetorical roles of sentences. For each rhetorical segment, a knowledge graph is constructed to capture the legal entities and their relationships within the segment. Contextual representations of the entities are then learned and aggregated to derive segment-level embeddings. These embeddings are further integrated to produce a unified document-level representation, and finally, the semantic similarity between a pair of documents is computed. We validate the performance of the proposed approach through extensive experiments on a benchmark Indian legal dataset, comparing it against state-of-the-art baselines to demonstrate its effectiveness.
Authors: Pedro Rocha Dantas, Lucas Pascotti Valem
Abstract: The rapid growth of image data has produced large-scale datasets, raising concerns about the time and memory costs of model training. Selecting representative training subsets, however, remains challenging: individual sample contributions are unclear, and model behavior varies across datasets and runs. We address these challenges with a framework that combines coreset selection with an ensemble aggregation over multiple runs. For coreset selection, we propose SCOre-Stratified Selection (SCOSS), which partitions the training data into intervals based on a chosen score and samples from each interval. The ensemble combines predictions from multiple runs, each performed on an independently sampled training subset. As baselines, we use moderate and random selection, each in original and class-balanced versions. We assess the framework with Simple Graph Convolution (SGC) and Support Vector Machine (SVM) classifiers under different sampling ratios. Experiments show that SCOSS is competitive with baselines, often the best choice for SGC, and enables favorable trade-offs between accuracy and efficiency. On the fine-grained dataset, SGC with SCOSS outperforms SVMs when using fewer labeled samples. The code and supplementary materials are publicly available at http://scoss.lucasvalem.com.
Authors: Yawen Li, Yan Li, Zhe Xue, Yingxia Shao, Meiyu Liang, Guanhua Ye
Abstract: Medical imaging models are often deployed without the demographic, acquisition, and quality metadata needed for subgroup auditing. Once those metadata disappear, clinically critical failure modes can be masked by strong aggregate performance, and many robust-learning methods lose the group structure they rely on. We present CAPRA, a calibrated proxy-axis framework for hidden subgroup analysis under missing metadata. CAPRA predicts image-derived semantic axes, calibrates axis posteriors on a small metadata-labeled split via patient-level cross-fitting, and organizes those posteriors into a calibrated subgroup interface that supports both deployment-time failure analysis and downstream robust learning without requiring subgroup labels at deployment. Across fundus, dermoscopy, and chest radiography, CAPRA reveals disparity patterns missed by metadata-only slicing, remains informative under dataset shift, and produces subgroup partitions that align more closely with explicit failure axes than image-only or latent-slice baselines. The same interface can also be reused by downstream robust learners, although those gains are domain-dependent. Overall, CAPRA turns hidden subgroup analysis under missing metadata into a calibrated, interpretable, and reusable subgroup interface for deployment-time analysis and robust transfer.
Authors: Camila Piscioneri Magalh\~aes, Lucas Pascotti Valem
Abstract: While the growing availability of image data has driven significant advances, labeling datasets remains costly and time-consuming. Therefore, semi-supervised approaches such as Graph Convolutional Networks (GCNs), which learn from both labeled and unlabeled data, have emerged as a promising solution. One of the primary challenges in applying GCNs to image classification is graph construction, since, unlike in citation networks or similar domains, images typically do not come with a predefined structural representation. For visual data, most studies construct graphs based on the similarity between feature vectors from pretrained deep learning backbones, typically by employing kNN or reciprocal kNN algorithms. Although Large Language Models (LLMs) have shown remarkable capability in capturing high-level semantics, their integration with GCNs for image classification remains underexplored. Aiming to fill this gap, our approach uses a Vision Language Model (VLM) to generate textual image descriptions, which are then processed by an LLM to estimate semantic similarity scores between connected images. These scores guide the pruning of edges in kNN and reciprocal kNN graphs, filtering out semantically irrelevant neighbors. Experimental results reveal that leveraging LLMs for graph refinement can improve classification accuracy, particularly for kNN graphs and some backbones. The source code is publicly available at http://gcnllm.lucasvalem.com.
Authors: Peipei Zhu, Yueqing Niu, Lin Zhu, Guanchong Niu, Yang Yu, Zheng Li
Abstract: Video anomaly detection (VAD) is critical for automated surveillance but remains fragile under challenging conditions such as illumination variations, fast motion, and complex backgrounds when relying solely on visible light videos. To address these limitations, we propose EVAD, an event enhanced VAD framework that jointly exploits conventional video and event streams captured by bio inspired event cameras. Event sensors asynchronously capture brightness changes with high temporal resolution, offering robustness to motion blur and extreme lighting, and providing motion salient cues complementary to video based visual information. To support multi modal VAD research, we construct a large scale visible event benchmark comprising 6.3 billion events and 376,368 video frames collected under diverse illumination levels, motion patterns, and background complexities, filling the gap of realistic and scalable datasets for event based anomaly detection. Building upon this dataset, we design a contrastive multi modal pretraining framework to learn discriminative event representations by aligning semantic embeddings across event streams, visible videos, and textual descriptions. An adaptive fusion module then dynamically integrates event based temporal cues with video based spatial semantics, improving robustness to environmental disturbances. Experiments on benchmarks and the proposed TJUTCM Pha dataset demonstrate that E VAD consistently outperforms methods, validating the effectiveness of event-based sensing for VAD in real world scenarios.
Authors: Bartosz Zi\'o{\l}ko, Kacper Dobrzeniewski
Abstract: In this study, we examine the opportunities brought by Large Language Models (LLMs) to various aspects of fundamental analysis of companies based on their reports as well as data and documents describing macroeconomic situation like GDP and inflation changes as well as documents filled to the U.S. Securities and Exchange Commission (SEC) which can be found in EDGAR. We were preprocessing those data and than sending via API to gpt-4o model in a Retrieval-Augmented Generation (RAG) like regime. We prepared as well a document describing an exemplar investor knowledge based on Kitchin cycles. We were scanning data important for analysis of 9 companies for 4 weeks. Using LLM we were producing automatic briefs about them. They were sent to nine participants who are individual investors to evaluate usefulness of such approach to data analysis.
Authors: Yiting Wang, Jingyi Zhang, Wenhu Zhang, Ke Chao, Yves Liang, Kun Cheng, Kang Zhao
Abstract: While large-scale text-to-image generative models have achieved unprecedented visual performance, their inherent reliance on multi-step iterative solvers incurs severe inference latency. Few-step distillation targeting the Classifier-Free Guidance (CFG) trajectory has emerged as the prevalent dual-dimensional compression paradigm. However, existing frameworks remain subjugated by a coarse-grained blind injection paradigm that perpetually enforces a globally static guidance strength while indiscriminately sampling the supervisor timestep. This state-agnostic design completely disregards the intrinsic nature of image generation as a dynamic evolutionary process characterized by progressive entropy reduction, which not only restricts the performance boundary of few-step compression but also precipitates severe CFG over-conditioning artifacts. To transcend these limitations, we re-examine the distillation procedure through the theoretical lens of Information Theory, formally modeling it as a dynamic mutual information game constrained by the Information Bottleneck (IB) principle. Specifically, we dismantle traditional blind assumptions via a dual-track adaptive framework. To determine the injection target, we propose an instance-aware selection mechanism that transmutes the intractable KL divergence constraint into a zero-overhead closed-form solution predicated on the local vector field norm. To regulate the injection strength, we introduce an entropy-aware schedule that dynamically decays alongside the SNR, applying maximal thrust for initial structural anchoring before smoothly reverting to the natural manifold to refine micro-details. Extensive empirical evaluations corroborate that our framework fundamentally eradicates over-conditioning artifacts, shattering the performance ceiling to achieve SOTA generative fidelity under extremely stringent 2-step configurations.
Authors: Sang-Hoon Lee, Ha-Yeong Choi
Abstract: Representation alignment (REPA) has been investigated to accelerate diffusion training, but we observe that regularizing intermediate representations in diffusion Transformers (DiT) may implicitly entangle latents and limit generative capacity. To address this issue, we propose ReGen, a hierarchical multi-prompt representation generation framework that jointly estimates multiple vector fields for both representations and data within a single diffusion model. We further introduce generalized flow matching (GFM) to improve the generalization of conditional flow matching (CFM). We validate ReGen on single-stage waveform diffusion models including neural audio codec and Wave-VAE. ReGen significantly improves waveform generation quality from highly compressed latent representations at 12.5 Hz. We also present ReGenVoice, a latent diffusion model (LDM)-based text-to-speech model that achieves strong speech intelligibility (WER) and speaker similarity (SIM) with a small dataset. Moreover, operating the LDM at 6.25 Hz with rich semantic and acoustic latent representation enables efficient training and sampling, requiring only 1 day of training on 4 GPUs and fast inference with an RTF of 0.08. Audio samples are available at https://regenvoice.github.io/demo/.
Authors: Qingchu Jin, Felistas Mazhude, Jamie B. Rabb, Robert S. Kramer, Douglas B. Sawyer, Raimond L. Winslow
Abstract: Achieving early and timely diagnosis and treatment for disease is a major challenge. Recent applications of machine learning (ML) algorithms trained on patient data have shown promise in many different settings for predicting the patient health state. A challenge often faced when applying these ML algorithms is that at any given time, not all clinical variables (features) needed as input to perform prediction tasks are available. We define the concept of full-feature-capacity (FFC) to refer to prediction performance when such algorithms make use of all features on which they were trained. We then introduce Feature Sufficiency Analysis (FSA) - an analysis for determining whether a subset of all clinical features needed by an AI model is sufficient to achieve FFC. FSA estimates the underlying distributions of missing variables conditioned on features that are available. FSA provides a patient-specific assessment of whether the existing set of measured features achieves FFC. If yes, then there is no need to acquire further inputs and a ML-based prediction. We provide two case studies: prediction of need for postoperative prolonged ventilation in patients recovering from heart surgery; 10-year mortality prediction in an outpatient cohort. We also demonstrate that FSA also provides a clinically interpretable feature-ranking methodology based on prediction sufficiency, identifies intrinsically hard-to-predict patient populations, and has the potential to perform cost-aware optimization for clinical data acquisition. FSA provides a generic computational approach for determining whether incomplete clinical information is sufficient to support trustworthy AI-assisted clinical decision-making, thereby facilitating the prospective deployment of healthcare AI systems across diverse clinical settings.
Authors: Nada Zine, Tristan Coignion, Vincenzo Stoico, Cl\'ement Quinton, Romain Rouvoy, Patricia Lago
Abstract: Large Language Models are reshaping how software is developed and maintained. They are typically deployed in production using inference engines such as vLLM, which can efficiently serve pre-trained, highly configurable models. While prior work has focused on model architectures and hardware acceleration, the impact of inference engine configuration on energy consumption, performance, and output quality remains poorly understood. In this paper, we present a large-scale controlled study of three selected vLLM configuration options: attention kernel type, prefix caching, and chunked prefill. We evaluate all combinations of these configurations across 5 open-weight LLMs and 5 diverse inference tasks, totaling $9,000$ runs and $93,600$ measures. We analyze energy consumption, latency, and accuracy, and examine both main effects and interaction effects between configuration options and tasks. Our results show that the studied configuration options significantly impact energy and performance, mainly driven by attention type and prefix caching, while chunked prefill has a limited effect under the default vLLM serving configuration and evaluated workloads. These effects are highly model- and workload-dependent, and no configuration is universally optimal. We further show that model choice dominates global trade-offs, while configuration tuning provides local improvements along the Pareto frontier. Unexpectedly, inference options can also affect model accuracy.
Authors: Wenjun Zhang, Zhiyong Chen, Tong Wu, Guo Lu, Li Song, Feng Yang, Meixia Tao
Abstract: The groundbreaking development of generative artificial intelligence (AI) is rapidly boosting the ability to generate content such as images and videos, reshaping communication paradigms. This article introduces generative communications (GenCom), a novel paradigm for 6G networks in which large AI models (LAMs) drive semantic understanding, reasoning, and content generation, embedding these into the communication process. Unlike traditional systems that strictly pursue accurate bit transmission, GenCom enables transmitters to convey only minimal yet sufficient information, while receivers leverage shared generative priors and knowledge bases to synthesize the intended output. Communication is thus redefined as controlled generation rather than data reproduction. We formalize the concept of GenCom, clarify its AI-native and generation-driven properties, and present its core mechanisms. A two-layer GenCom architecture supported by key enabling technologies is proposed, and analysis of four representative application scenarios demonstrates that GenCom offers ultra-efficient transmission, semantic-level robustness, and new network functions. Finally, we outline future research directions, including foundational theory and real-time processing, highlighting a promising pathway toward 6G networks.
Authors: Julius St\"ork
Abstract: Continual learning commonly relies on post-hoc mechanisms such as replay, elastic regularization, or distillation. This work argues that forgetting should instead be modeled directly as interference between tasks. In the frozen-feature regime, forgetting from learning a new task is exactly the interference energy induced on the old task. In deep networks, the same quantity is recovered through path-averaged curvature with minimal additional forward passes. When task supports are disjoint, forgetting can be eliminated structurally and when task supports overlap in conflicting directions, a non-zero distortion floor is unavoidable. The same geometry optimally merges models through task-aware orthogonalization. From this analysis we derive Interference-Gated Functional Allocation (IGFA), a replay-free, Fisher-free method that shares directions when tasks align and protects them when they conflict. Across benchmarks, IGFA achieves lossless retention when tasks are structurally separable and moves unavoidable cost from irreversible forgetting into deferred but recoverable plasticity when they are not. It matches the strongest replay-free structural baselines on dissimilar-task streams and improves on unconditional projection when similarity makes transfer worth preserving.
Authors: Rishabh Madan, Angchen Xie, Samantha Saak, Andres Blanco, Dohyeok Lee, Sarah Grace Brown, Yunting Yan, Mark Zolotas, Jose Barreiros, Tapomayukh Bhattacharjee
Abstract: Whole-arm manipulation involves direct contact with the environment while the robot completes a task by distributing contact across multiple links as contacts form, slide, and break. This setting breaks common implicit assumptions in many learning-based manipulation pipelines: arm configuration tightly couples motion and contact forces, contact state is partially observed under occlusion, and purely learned rollouts can become physically inconsistent under distribution shift because many multi-link contact configurations are sparsely represented in the data. To address this, we propose TACTIC (Tactile and Vision Conditioned Contact-Centric Control), a receding-horizon controller for whole-arm manipulation. TACTIC uses a contact-centric hybrid predictive model that combines RGB-D, distributed tactile sensing, and a compact 2D proximity representation. The model couples a learned, action-conditioned latent dynamics model with analytical kinematics through contact Jacobians, enabling rollouts of future contact configurations and interaction forces. TACTIC integrates these rollouts into a sampling-based MPC planner with contact-aware action sampling: contact Jacobian-based projections steer sampled action sequences toward force-modulating directions, and objectives defined over predicted proximity and interaction forces trade task progress against whole-arm force regulation. We evaluate TACTIC in simulation against state-of-the-art model-based and model-free methods, and perform ablations that isolate the contribution of each design choice. TACTIC consistently outperforms other methods. We further demonstrate real-world performance on a robot with distributed tactile sensing across three whole-arm manipulation tasks that require multi-contact trajectories: turning over and repositioning a manikin, and goal-reaching in a 3D dynamic maze. Website: https://emprise.cs.cornell.edu/tactic
Authors: Alfredo Garrach\'on Ruiz, Tom\'as de la Rosa, Daniel Borrajo
Abstract: Version control systems are essential for collaborative software development, yet tools like git remain challenging for many practitioners. Recent advances in Large Language Models (LLMs) offer promising capabilities for interpreting developer intent, but their effectiveness in repository management tasks is limited by the need for formal reasoning. This work introduces Git-Assistant, an AI-based assistant that combines LLMs with automated planning to support developers in executing non-trivial git operations. The assistant analyzes repository context, translates natural language requests into actionable command sequences, and incorporates planning techniques to ensure correctness and safety. We present a systematic evaluation methodology using synthetic and randomized git environments, comparing the performance of LLM-only and planning-augmented variants across multiple metrics. Experimental results demonstrate that integrating formal reasoning with LLMs improves reliability and reduces errors in repository management, highlighting the potential of hybrid AI approaches for intelligent developer assistance.
Authors: Jayadeva, Madhur Aswani
Abstract: The current state of the art in AI/ML rests on deep neural architectures, which, in general, suffer from a lack of interpretability. Interpretability is crucial to gleaning insights while analyzing experimental data, where quantitative predictions may not be adequate for a scientist. We present a three layer neural architecture, SAMPAT (Smooth Approximation via Multivariate Polynomials and Analytic Transformations), that can provably learn a continuous, everywhere differentiable function, that can approximate any smooth function arbitrarily closely. SAMPAT's approximant can be expressed as a closed and compact algebraic, analytic expression, providing complete interpretability. Experiments on synthetic and benchmark datasets indicate that SAMPAT yields competitive performance with simpler representations. For many tasks, a two layer SAMPAT suffices. By imposing restrictions on the connectivity between neurons, SAMPAT may be used to provide a range of approximants, including regular and trigonometric polynomials, rational expressions, Gaussians, mixtures of Gaussians, as well as arbitrary combinations of the same; without restrictions, it learns a suitable structure. SAMPAT may be used to factorize polynomials and model nonlinear systems. With the addition of skip connections, a 4 to 6 layer SAMPAT is adequate to represent a substantive range of methods widely used in AI/ML, allowing the choice of a model's family, not just its parameters, to also be optimized as part of the learning process.
Authors: Gwenn Beets, Anniek Jansen, Saar Hommes, Ruben D. Vromans, Leonie Westerbeek, Supraja Sankaran, Julia C. M. van Weert, Emiel J. Krahmer, Nadine Bol
Abstract: AI chatbots are increasingly used for answering health-related questions. This study examines the role of topic type discussed with an AI chatbot and individual characteristics on perceived benefits and risks, intention to use an AI chatbot, and willingness to self-disclose health information. We conducted an online experiment with a 2 (topic type: physical versus psychological, between-subjects) x 2 (topic sensitivity: low versus high, within-subjects) mixed design among a Dutch representative sample (N = 1,388). Results showed that perceived benefits were positively associated with intention and willingness to self-disclose, while perceived risks were negatively associated. Moreover, participants reported higher usage intentions for low-sensitive topics compared to high-sensitive topics. Furthermore, perceptions, intention, and willingness to self-disclose varied by individual characteristics. Overall, our findings suggest that intentions to use AI chatbots and self-disclosure of health-related information are primarily related to perceived benefits and risks and to personal characteristics rather than to topic type.
Authors: Saviz Changizi, Nasibeh Mohammadzadeh, Mohammad Shojafar, Rahim Tafazolli
Abstract: Telecom fraud-control studies often stop at detector-level classification, but deployment use requires request-level policy resolution, lifecycle traceability, and auditability. This paper reframes fraud control as blockchain-linked auditable decision management for synthetic telecom/IoT fraud-control requests, and its main result is that the QLoRA-tuned LLM branch becomes much more usable than zero-shot prompting but mainly approaches, rather than outperforms, a lower-cost centralized ensemble. The framework maps each synthetic deployment record to a managed request, blocks explicit out-of-boundary cases through a deterministic hard-fraud gate, scores non-hard requests using centralized ML (M1), federated meta-learning (M2), or LLM-family risk sources (M3), and resolves actions through a shared five-state policy, two-zone refinement mechanism, and local Ethereum-compatible audit layer. Evaluation uses separate synthetic training data and a 100,000-record deployment replay corpus, so the study should be read as controlled drift-replay evidence rather than field validation or proof of live deployability. On validation, M1 gives the strongest balance, with legitimate-request FPR 0.0890 under the 0.10 operating cap and soft-fraud recall 0.8341. On labeled deployment replay, however, the legitimate-FPR gap becomes large: M1 rises to 0.1646 and M3-QLoRA to 0.1801, while M3-QLoRA reduces the M3-Base legitimate FPR from 0.3915 and reaches 0.8240 soft-fraud recall. Blockchain telemetry shows that lifecycle gas, cost, latency, and throughput differences are driven by submitted off-chain decision profiles rather than changes in fraud logic.
Authors: Maxim Chupilkin
Abstract: Large language models (LLMs) are increasingly used to summarize and evaluate policy-relevant information, but it remains unclear whether their judgments are implicitly shaped by geopolitical cues. I study this question with an endorsement experiment in which four LLMs evaluate the same international economic and security policies after each policy is randomly described as supported by the United States, the European Union, China, or Russia. In the numeric-only condition, GPT-5, Claude Sonnet, and Gemini rate China- and Russia-endorsed policies substantially lower than identical policies endorsed by the United States or the European Union; DeepSeek is the main exception. A second condition asks models to provide a short justification with the score. This request leaves the broad Western/non-Western gap intact for GPT-5 and Claude Sonnet, attenuates Gemini's penalties, and sharply activates China and Russia penalties in DeepSeek. The justifications indicate that Western endorsement is often treated as a credibility cue, whereas Chinese and Russian endorsement is treated as a cue for data security, sovereignty, surveillance, or geopolitical risk. These findings show that LLM policy evaluations can depend on the identity of a foreign endorser even when policy content is held fixed.
Authors: Pedro P. Santos, F\'abio Vital, Alberto Sardinha, Francisco S. Melo
Abstract: We study general-utility Markov decision processes (GUMDPs) with risk-aware objectives. In this framework, an agent aims to optimize a risk measure of the distribution of objective values, where the objective function depends on the frequency of visitation of states induced by the agent's policy. First, we motivate, propose, and formalize risk-aware GUMDPs, which enable agents and decision makers to trade off expected performance by risk aversion while benefiting from the rich set of objectives that can be cast under the framework of GUMDPs. We focus our attention on the entropic risk measure (ERM). Second, we show how we can solve risk-aware GUMDPs with ERM objectives by resorting to online planning techniques. In particular, we propose an approach based on Monte Carlo Tree Search (MCTS) to provably solve risk-aware GUMDPs up to any desired accuracy. Third, we provide a set of experimental results showcasing that our approach is successful when optimizing for a spectrum of risk-aware behaviors in the context of GUMDPs under diverse tasks (standard MDPs, maximum state entropy exploration, imitation learning, and multi-objective MDPs).
Authors: Kwan Soo Shin, In Seok Kang, Yunkyung Min
Abstract: Language models are optimised for scale, yet remain functional rather than companionable, and as an assistant personalises into a companion, accumulating memory of one user, it quietly becomes someone, and can silently acquire traits that harm that user. What a companion is becoming, and what would make it worth becoming, has no reliable instrument: trained human raters cannot agree on the answer (Fleiss kappa = 0.074). Here we show that three small language models (146 M to 3 B parameters) sharing a hyperbolic substrate answer both halves of that question. A 146 M behavioural auditor, trained from scratch, detects the compliance gap that those raters cannot (90.7% binary-compliance accuracy); a linear read-out of its frozen representation further detects companion-induced sycophancy, dependence-fostering and confabulated memories on generator families unseen in training (AUROC 0.804 under style-controlled, leave-one-generator-out evaluation, versus 0.721 for a frontier zero-shot judge on the same items). A creative frame-seeder is preferred in 100% of 311 decided pairwise comparisons over four prompting baselines. A memory operating system implements designed forgetting, M(t) = S*exp(-lambda*t), whose predicted skeleton-wallpaper partition emerges only under selective retrieval gating in a four-condition pilot. Creativity, honesty and designed forgetting constitute a small-model route to trustworthy companion AI.
Authors: Miguel Arana-Catania, Gillian Pink, Glenn Roe
Abstract: Thematic indexing -- the practice of assigning structured conceptual labels to sections of text -- is essential to scholarly access in large-scale literary and historical editions, yet it remains a largely manual, labour-intensive process. This paper explores the application of machine learning to automatic thematic indexing, using two substantial sub-corpora of the Complete Works of Voltaire as a test case: the Essai sur les m\oe urs et l'esprit des nations and the Questions sur l'Encyclop\'edie. The task is framed as a multi-label classification problem, in which a model must assign the set of index entries that a professional indexer would apply to a given page of text. We compare a range of approaches -- from encoder-based models with classification heads to generative large language models (LLMs) fine-tuned via Low-Rank Adaptation (LoRA) -- spanning model sizes from approximately 3 to 120 billion parameters. Our best-performing model, from the Mistral family in a 4-bit quantised configuration, achieves F1 scores of up to 0.67; we argue that these figures represent lower bounds, given the inherent subjectivity of professional indexing and the frequency with which model predictions prove semantically valid despite diverging from the print index. We further evaluate cross-corpus generalisation and conduct a detailed qualitative analysis of model behaviour on literary and rhetorical features of the source texts that prove particularly resistant to automated treatment. Our findings have implications for the broader challenge of providing structured thematic access to large-scale literary and historical corpora.
Authors: Miguel Arana-Catania, Catherine Conisbee, Matthew Kidd
Abstract: Identifying and assigning keywords at scale is a technical, practical, and ethical challenge for crowdsourced collections. This article reports the findings of the "Extracting Keywords from Crowdsourced Collections" project, which used the Their Finest Hour Online Archive, a crowdsourced Second World War digital collection hosted by the University of Oxford, as a case study. The project evaluated three Natural Language Processing approaches to automate keyword extraction: Named Entity Recognition, Keyword Extraction, and Topic Modelling. It tested these approaches across a range of artificial intelligence techniques, from traditional statistical methods to modern GenAI neural networks. Our quantitative and qualitative findings indicate that Natural Language Processing approaches offer real potential for keyword extraction at scale in crowdsourced collections, but that no single method offers a complete solution and that model choice significantly shapes results. We argue that in crowdsourced collections, where metadata is the direct product of engagement with living contributors, automated keyword extraction raises distinct stewardship responsibilities that must be addressed alongside technical performance. Open-weight, extractive models emerge from our evaluation as best placed to support responsible deployment, while generative AI, despite its abstractive potential, introduces accountability risks that anyone managing crowdsourced collections should weigh carefully.
Authors: Zixin Chen, Peng Liu, Haobo Li, Rui Sheng, Jianhong Tu, Xiaodong Deng, Fei Huang, Kashun Shum, Dayiheng Liu, Huamin Qu
Abstract: Answering complex questions over long documents frequently requires integrating evidence that the source itself disperses naturally across distant passages. In an incident report, the operating condition, design flaw, and missed safety check that jointly explain a disaster may appear dozens of sections apart; in a novel, a character's true motive may surface only through scenes far removed from the moment it becomes relevant. This source-internal evidence integration is central to real-world long-document analysis, yet existing benchmarks largely sidestep it. Needle probes, planted facts, and reverse-engineered multi-hop chains embed evidence that may differ from the host text in distribution, placement, or register, making it unclear whether strong performance reflects genuine source reasoning or distributional artifacts. We introduce WILDTRACE, a benchmark of 481 tasks over 214 naturally occurring long-form sources such as technical incident reports and lesser-known literary narratives, where all evidence trails arise from the document's own causal, temporal, and narrative logic. Drawing on Pearl's causal hierarchy and prior multi-hop reasoning typologies, we define seven source-internal evidence geometries that characterize the distinct relational demands of analytical reading in long documents. A source-first construction pipeline mines candidate trails from document structure before writing questions; each item then undergoes multi-stage validation covering clue necessity, answer groundedness, rubric fidelity, contamination resistance and answerability. As models are increasingly entrusted with real-world high-stakes analytical tasks, this gap between accessing information and reasoning over naturally dispersed evidence emerges as a defining challenge for the next stage of long-context research.
Authors: Guanquan Wang, Yoshimasa Tsuruoka
Abstract: Diffusion-based trajectory planners have shown strong performance in offline reinforcement learning, but their iterative denoising process often incurs high inference cost. Consistency-based planners reduce the number of sampling steps, yet they typically rely on a two-stage teacher--student distillation pipeline that increases training cost and may introduce instability. We propose Shortcut Trajectory Planning (STP), an offline model-based reinforcement learning framework that incorporates shortcut models as efficient trajectory generators. STP trains a conditional shortcut trajectory model in a single stage, supports adjustable one-step and few-step inference through step-size conditioning, and selects candidate plans using a critic augmented with feasibility-aware correction. Across standard D4RL benchmarks, including locomotion, navigation, manipulation, and dexterous control tasks, STP achieves strong performance while simplifying the training pipeline for fast generative planning.
Authors: Cedric Caruzzo, Donggeun Yoo, Tae Soo Kim
Abstract: Retrieval-augmented generation evaluation checks whether model claims are factually grounded in retrieved documents. It does not check whether retrieved evidence is attributed to the correct entity. A clinical RAG response can pass every automated check (zero hallucinations, near-perfect faithfulness, real citations) while presenting drug Y's clinical evidence as evidence about queried drug X. We term this deceptive grounding (DG): a failure invisible to faithfulness, hallucination, and citation checks because every claim is sourced from a real document, about the wrong entity. Using a controlled factorial benchmark across 13 models, we find DG rates spanning 8-87% at peak adversarial conditions. Medical and biomedical fine-tuned models reach up to 86.7%; domain specialization amplifies the failure rather than mitigating it. A controlled ablation identifies the mechanism: removing entity-specific clinical evidence from retrieved documents eliminates entity-attribution failure entirely, shifting all failures to confabulation. The two failure modes respond to the same trigger, taking different paths. Production measurement across 740 drug-disease pairs finds 7.8% overall DG in a deployed RAG system, rising to 13.6% for recently approved drugs. Entity-attribution verification (checking that cited evidence applies to the queried entity) detects DG at 97.0% precision and 98.7% DG recall (IPW-adjusted human gold standard); no existing framework implements it.
Authors: Seungyong Lee, Hyun Jun Jang, Sangoh Kim, Sungjoon Park
Abstract: Virtual try-on (VTO) has made significant progress in realistically transferring garments onto a target person. Yet most systems give the user little control over how a garment should be worn -- its size (loose or fitted), style (e.g., tucked in or untucked, open or closed), and spatial placement on the body. We address this gap with two complementary contributions. First, we define and solve Visual-Instance-Prompt Segmentation via VIP-SAM: given a flatlay image of a garment, segment that specific instance in a photograph of a person wearing it. This is an instance-level task, distinct from the typically studied category-level segmentation. Second, we introduce CtrlVTON, a controllable VTO framework that recasts try-on as an image editing problem and adds segmentation masks as pixel-level control over garment layout, including style, size, and spatial placement on the body. VIP-SAM and CtrlVTON each achieve state-of-the-art results on their respective tasks. In particular, CtrlVTON generates images that follow user-provided layouts far more faithfully than the strongest proprietary editing systems while matching them on garment fidelity.
Authors: Shirley Yu, Ruben Martins
Abstract: Program verification is crucial for software correctness, but producing fully verified programs remains difficult in practice. This paper studies whether implementation structure affects automated verifiability when multiple generated programs are intended to satisfy the same task-level semantics. We present Diversify2Verify, a staged LLM-based pipeline for Why3 that infers representation-specific contracts, generates and tests diverse recursive and imperative array/list implementations, and attempts verification with bounded verifier-guided annotation repair. We also construct a verification-oriented benchmark of 73 tasks over integers, arrays, and lists, yielding 292 implementation variants. Diversify2Verify verifies 96 artifacts initially and 154 after two repair passes, improving artifact-level verification from 32.9% to 52.7%. At the task level, at least one variant verifies for 49 of 73 tasks, a 67.1% success rate. These results show that task-equivalent implementations can differ substantially in verifiability and that implementation diversity helps find verification-friendly artifacts.
Authors: Brennan Bell, Inti Gabriel Mendoza Estrada, Andreas Tr\"ugler, Paul Erker
Abstract: We study an adversarial bandit problem for entanglement-based quantum-network routing over a modest graph corpus. Alice selects an end-to-end repeater route for an Ekert-91 protocol (E91) representing her move, while Eve selects an attack surface, either edge intercept--resend or repeater memory degradation. Payoffs are drawn from cached SeQUeNCe-simulated E91 transcripts, and Alice accepts a turn when the finite-sample statistic violates the Clauser-Horne-Shimony-Holt (CHSH) bound. Performing adversarial co-learning across 50 structured topologies, we find that learned retention tracks a full-matrix minimax reference closely (Pearson $r=0.99$): under a one-surface Eve action model, bottleneck families have zero retention, while non-bottleneck families follow a $1-1/N$ coverage principle. We then fit decision-tree explanation models to graph-, attack-, and route-level topology-corpus targets and report their faithfulness. Finally, we construct prompt records for local language models to summarize the tree evidence, resulting in an open-source explanation workflow for quantum-repeater network games.
Authors: Victor J. B. Jung, Gagandeep Singh, Joseph Melber, Kristof Denolf, Francesco Conti, Luca Benini
Abstract: The growing adoption of large language model-based agents within operating system workflows has increased the importance of energy-efficient inference on laptop-class systems-on-chip (SoCs). While cloud offloading remains common, it introduces reliability and privacy concerns that are particularly problematic for agentic workloads. Recent laptop SoCs, therefore, incorporate neural processing engines (NPUs) optimized for energy efficiency; however, effectively mapping attention mechanisms onto NPUs remains challenging due to architectural diversity and explicit data-movement programming models. In this work, we present STEEL, the first open-source implementation of FlashAttention targeting XDNA-like NPUs. STEEL introduces a dataflow formulation of prefill attention, enabling efficient exploitation of spatial parallelism and on-chip memory. Furthermore, STEEL addresses the load imbalance induced by the causal mask by leveraging a sparsity-aware pipeline placement onto the NPU array, reducing synchronization overhead and improving utilization. We evaluate STEEL on the AMD Ryzen AI 9 HX 370 SoC and compare its performance against optimized CPU and GPU implementations. Experimental results show that STEEL reduces energy consumption by an average of 9.17x and 1.75x relative to CPU and GPU baselines, respectively. On XDNA 1, STEEL achieves an average 9.6x latency reduction over the prior state of the art, and delivers a 22.8x speedup on average compared to a layer-by-layer attention implementation on XDNA 2.
Authors: Wout Mommen, Lars Keuninckx, Matthias Hartmann, Werner Van Leekwijck, Piet Wambacq
Abstract: We introduce a novel method for both partial and full optimization of the connections in deep differentiable logic gate networks (LGNs) and lookup table networks (LUTNs). Our training method utilizes a probability distribution over a set of connections per gate/lookup table (LUT) input pin, selecting the connection with highest merit, all whilst the optimal gate types or LUT-entries are learned in parallel. We show that the connection-optimized LGNs outperform standard fixed-connection LGNs on the Yin-Yang, MNIST Handwritten Digits and Fashion-MNIST benchmarks, while requiring only a fraction of the number of logic gates. We achieve 98.92% on the MNIST dataset with two layers of 8000 gates. With only one layer of 8000 gates, we obtain 98.45%, showing that our method requires almost 50 times fewer gates compared to fixed-connection LGNs. Training stability up to ten layers has been ensured by employing a high learning rate, straight-through estimators and trimming constant-output gate types. Additionally, we present a LUT neuron description that enables stable training with backpropagation, tested up to 6-layer deep networks. The model requires four times fewer trainable parameters and still achieves a higher accuracy compared to the fixed-connection LGN training algorithm. Our connection-training algorithm also works well for the LUTNs, achieving an accuracy of 98.88% for two layers of 2000 6-input LUTs.
Authors: Avik Bhatnagar, Anton Paule, Tobias Schuermann, Sebastian Reiter, Oliver Bringmann
Abstract: Adaptive power management in Electric Vehicles (EVs) requires accurate power prediction. Although deep learning models have emerged as highly effective for time-series forecasting in this domain, their performance is prone to degradation when exposed to data with distributions different from the training data. We introduce a novel approach that enables on-device learning in resource-constrained EV systems to continuously adapt pretrained battery prediction models to new, unseen data. We leverage existing pretrained models by transforming them into adaptable versions that retain critical hyperparameter knowledge from their initial training. We comprehensively investigate both online and offline model adaptation strategies. Our results demonstrate significant improvements in forecasting performance across various models and time horizons, achieving mean absolute error reductions of up to 7.49\% and 14.88\% with online and offline adaptation techniques, respectively. This study highlights the substantial benefit of on-device adaptation, resulting in enhanced battery power predictions than unadapted model deployments in real-world EV scenarios.
Authors: Xinyu Zhu, Zhe Xu, Xiaohan Wei, Yunchen Pu, Fei Tian, Chonglin Sun, Kaushik Rangadurai, Hua Zhi, Frank Shyu, Sandeep Pandey, Luke Simon, Yu Meng, Xi Liu
Abstract: Long-context processing has become increasingly important for large language models (LLMs), but simply extending the context window does not guarantee effective utilization of long inputs. As input length grows, accuracy often degrades, indicating that models still struggle to identify and use the evidence most relevant to a question. A promising way to improve long-context utilization is test-time training (TTT), which treats the test context as a training example for instance-specific parameter adaptation. However, applying TTT to the entire long context is prohibitively expensive, while adapting on randomly sampled spans introduces severe noise. Because most spans in a long context are irrelevant to the specific question, training on them may even degrade the base model's performance. Our preliminary study shows that TTT is highly sensitive to training-span quality: on LongBench-v2, TTT on randomly sampled spans hurts performance, whereas TTT on oracle spans substantially improves it. Motivated by this, we propose a simple method, Self-Guided TTT (S-TTT): before adaptation, the model identifies the evidence spans it should learn from, and the standard language-modeling training objective is applied only to those selected spans. On two challenging long-context reasoning benchmarks, LongBench-v2 and LongBench-Pro, S-TTT improves accuracy for both Qwen3-4B-Thinking-2507 and Llama-3.1-8B-Instruct, achieving up to a 15% relative improvement.
Authors: Hyein Park, Namho Kim, Junhwa Kim
Abstract: Ambivalence and hesitancy are subtle behavioral states that are expressed through a combination of verbal content, facial behavior, visual context, and acoustic cues. Effective recognition therefore requires not only extracting informative unimodal representations, but also modeling how temporally aligned behavioral evidence interacts across modalities. In this paper, we propose a synchronized visual-facial cross-refinement framework (SVF-CR) with pairwise multimodal evidence fusion for ambivalence and hesitancy recognition. The proposed method first extracts whole-video segment tokens and cropped-face segment tokens using the same temporal partition. The synchronized visual and facial tokens are refined through intra-modal self-attention and bidirectional visual-facial cross-attention, allowing whole-video context and local facial behavior to mutually refine each other before evidence construction. We then construct segment-level visual-facial evidence using consistency and discrepancy modeling, followed by temporal self-attention and attention pooling. Textual and acoustic features are lightly refined through context self-attention and are fused with the enhanced visual-facial evidence at the final decision stage using pairwise evidence fusion. Experiments on the BAH (Behavioral Ambivalence/Hesitancy) public evaluation split show that the proposed synchronized visual-facial cross-refinement improves public macro-F1 over both global visual-face token fusion and synchronized evidence baselines, achieving a public macro-F1 of 0.7156. Code is available at : https://github.com/hiinnnii/BAH-Challenge-ECCV2026\_SVF-CR.
URLs: https://github.com/hiinnnii/BAH-Challenge-ECCV2026\_SVF-CR.
Authors: The Soofi-Team, :, Benedikt Droste, David Fitzek, Ruben H\"arle, Lukas Helff, Maximilian Idahl, Alex Jude, Abbas Goher Khan, Maurice Kraus, Timm Ruland, Richard Rutmann, Sebastian Sztwiertnia, Markus Frey, Daniil Gurgurov, Jan Pfister, Tom R\"ohr, Sebastian von Rohrscheidt, J\"org Bienert, Nicolas Flores-Herr, Simon Gottschalk, Andreas Hotho, Kristian Kersting, Joachim K\"ohler, Alexander L\"oser, Wolfgang Nejdl, Simon Ostermann, Jan Plogsties, Patrick Putzky, Mehdi Ali, Michael Fromm, Max L\"ubbering
Abstract: We present Soofi S 30B-A3B, a sovereign, open-source Mixture-of-Experts (MoE) hybrid Mamba Transformer foundation model for German and English. Its hybrid design activates only 3B of 30B parameters per token and keeps the inference cache near-constant as context grows, giving it a decisive throughput advantage over dense models for long-context, high-concurrency deployment. Pretrained on roughly 27 trillion tokens with deliberately up-weighted German, Soofi S matches dense 14 to 27B models on aggregate English and German benchmarks while achieving the best code aggregates in both languages among 17 open base models, and outperforms every European sovereign baseline in our comparison, including ones far larger in active parameters. Among fully open models, Soofi S obtains the highest English and German evaluation scores, ahead of Olmo 3 32B and Apertus 70B. Soofi S was built end-to-end on the German Industrial AI Cloud, a sovereign HPC scale AI infrastructure operated by Deutsche Telekom in Munich. Soofi S will be released under highly permissive, open-access terms: weights, selected intermediate checkpoints, full per-source data accounting, hyperparameters, and training and evaluation code. Where source licenses permit, data-construction artifacts are released under permissive licenses; commercially licensed sources are documented with aggregate statistics and exact mixture accounting.
Authors: Spiros Baxevanakis, Peng-Jian Yang
Abstract: Test-time scaling (TTS) reliably improves reasoning in large language models, but whether it transfers to small open vision-language models remains unclear. We examine this on EXAMS-V, a multilingual visual multiple-choice benchmark, comparing self-consistency, describe-then-reason with PRM-guided beam search, and two post-hoc selectors across Qwen2.5-VL-7B-Instruct and Qwen3.5-4B. What matters is the conditions under which TTS runs, not the search or verification machinery. The largest factor is parseability: an early prompt format left many chains reasoning correctly yet never committing to an answer letter, which a standard answer cue and a guided repair step largely remove. A larger decoding budget removes the rest: raising the per-chain token limit from 1k to 2k recovers 3.7 pp, whereas sampling more chains (8 to 16) adds only 0.15 pp. Once chains have room to finish, elaborate methods contribute little: PRM-guided beam search trails plain self-consistency by 0.39 pp at over eight times the cost, and neither a training-free generative critic nor a trained multimodal PRM beats majority vote across both policies. The largest gain comes instead from the policy model itself (+11.4 pp). Our best configuration reaches 84.1% on the held-out ImageCLEF 2026 test split, ranking first on the Visual MCQ leaderboard.
Authors: Anil Osman Tur, Tonje Knutsen Sordalen, Kim Tallaksen Halvorsen, Cigdem Beyan
Abstract: Long-term animal re-identification (ReID) must remain robust to gradual morphological evolution and seasonal appearance shifts. Although recent vision-language models provide strong pretrained visual representations, adapting them to longitudinal ecological settings remains challenging, particularly under identity and temporal distribution shifts. We present a parameter-efficient CLIP adaptation framework for animal ReID and introduce a continuous metadata-conditioning mechanism that incorporates numerical attributes directly into the prompt representation during training. While low-rank visual adaptation, prompt-based supervision, and cross-modal alignment provide the adaptation framework, the proposed metadata-conditioning strategy constitutes the primary methodological contribution. By preserving the continuous structure of numerical metadata rather than discretizing it into textual categories, the proposed approach enables smooth modulation of the embedding space during training while maintaining a purely visual inference pipeline. Experiments on a seven-year longitudinal fish dataset and multiple wildlife benchmarks demonstrate improved performance under closed-set, open-set, and time-aware evaluation protocols. The results demonstrate that continuous metadata conditioning improves robustness to longitudinal appearance variation and temporal distribution shifts, while parameter-efficient adaptation enables a purely visual inference pipeline without requiring metadata at test time. Code and evaluation splits can be found at: https://github.com/AnilOsmanTur/MetaPrompt-ReID.
Authors: Charles Edward Gagnon, Steven H. H. Ding, Philippe Charland, Benjamin C. M. Fung
Abstract: We present a practical pipeline for recovering source code from stripped binary functions by combining reverse engineering, anchor-based source code retrieval, and large language model reasoning. Our binary-to-source-code retrieval method attempts to identify the source function from a source code database, rather than generating approximate decompiled pseudocode. It extracts anchors such as strings, constants, external calls, and available function names using Ghidra, retrieves candidate files via an inverted-index search database, narrows candidates to likely function snippets, and re-ranks them with a large language model (LLM) based on disassembly, decompiled code, and source metadata. Confident matches can also serve as anchors in later passes. In an evaluation backed by our high-fidelity source code database on a stripped, optimized tcpdump binary, our proposed binary-to-source matching method achieves 95.2% assembly instruction coverage. Experiments on a GitHub-based retrieval database showed lower performance with 35.5% instruction coverage on average, mainly due to retrieval misses. These results show that source-level binary recovery excels with high-quality databases and remains a useful tool in noisy environments.
Authors: Yungeng Liu, Xuanzi Fang, Haijin Zeng, Qi Dai, Yongyong Chen
Abstract: Text-guided medical image segmentation leverages clinical semantics to improve lesion delineation, yet many existing models bind cross-modal fusion, supervision, and decoder design into a task-specific architecture. Such tight coupling makes it difficult to reuse language guidance modules across heterogeneous vision and text backbones, and often requires redesigning the network when the encoder pair changes. This paper presents BTHA, a backbone-transferable hierarchical adapter framework for text-guided medical image segmentation. BTHA is built around a stable feature-level interface: given multi-scale visual features and a text representation, it injects semantic guidance through shape-preserving adapters while maintaining the decoder-side tensor contract. To make this interface effective, we introduce a Hierarchical Coarse-to-Fine Supervision Strategy that decomposes learning into global image-text alignment, multi-scale auxiliary localization, and boundary-aware final mask refinement. We further design a Scale-Adaptive Gated Semantic Guidance (SAGSG) adapter, where resolution-specific gates adaptively control textual injection and channel recalibration suppresses redundant cross-modal responses. Evaluations across diverse vision and text backbones show that the same adapter and supervision design remains effective across convolutional and transformer-based visual encoders as well as different language encoders. Experiments on four public datasets further demonstrate that BTHA improves strong text-guided baselines with modest computational overhead.
Authors: Pan Li
Abstract: Explaining machine-learning models is increasingly important for decision-making and consumer trust, yet it is widely believed to come at a cost: existing Explainable AI (XAI) methods suffer from a persistent accuracy-explainability trade-off. We argue that this trade-off is not fundamental, but an artifact of treating explanation and prediction as separate objectives; when properly coupled, they become complementary, so that equipping a model to explain itself improves, rather than degrades, its accuracy. We introduce the Rashomon Explanation paradigm, which builds a set of faithful, prediction-guiding explanations rather than a single one, and prove that this set is generally non-empty and that explanation fidelity bounds the performance of the models it guides. To explore this set, we propose RashomonLLM, an Explanation-Prediction-Reflection agentic workflow that generates explanations in natural language by iteratively aligning them with predictions, and we prove it converges and recovers the full set. Across customer-churn classification, clinical survival regression, and industrial click-through prediction on large-scale live-streaming logs, RashomonLLM significantly outperforms state-of-the-art prediction and XAI baselines on both accuracy and explanation quality, with gains driven by explanation fidelity and robust to distribution shifts, temporal splits, and seeds. Our framework thus advances business performance while laying the groundwork for consumer trust.
Authors: Filippo Ziliotto, Luciano Serafini, Lamberto Ballan, Tommaso Campari
Abstract: A fundamental challenge in 3D reconstruction and robotic localization is co-visibility: determining which image pairs share overlapping visible surfaces, particularly in scenarios with minimal overlap. We demonstrate that VGGT implicitly encodes co-visibility as an emergent behavior: without any supervision for this task, its internal representations exhibit a clear hierarchical structure mirroring that of large language models, i.e. early layers build a 3D-aware scene representation, while late layers act as dedicated co-visibility reasoners. In particular, we identify layer L17 as a negative anchor that consistently routes non-co-visible pairs for this backbone, regardless of the evaluation setting, providing task-grounded evidence of layer specialization in a geometry-grounded foundation model. Building on this, we introduce Co-VGGT, which freezes VGGT and trains only a lightweight layer-wise mixture-of-experts head (less than 7.5M parameters) to classify co-visibility from RGB alone, treating each layer as a specialized expert whose geometric abstraction is adaptively weighted per input pair. On the Co-VisiON benchmark, Co-VGGT surpasses the human annotation baseline and improves over prior work by more than 25% pairwise and 10% multiview. Pairwise predictions are well-calibrated (ECE=0.030), enabling direct use as edge weights in visibility graphs for downstream SfM and SLAM pipelines without post-hoc correction. Code and data are available.
Authors: Xiangxin Zhao, Han Li, Shuaiting Li, Tianyi Zhao, Earl T. Barr, Federica Sarro, He Ye
Abstract: Large language model (LLM) coding agents are increasingly deployed to autonomously perform software engineering tasks in terminal-based environments, making their reliability a growing concern. Existing empirical studies investigate why coding agents fail, yet they largely treat failure as a final outcome rather than a temporal process, providing limited insight into how failures emerge, evolve, and become unrecoverable. We present the first large-scale empirical study of CLI coding-agent failure trajectories, introducing a process-oriented framework that analyzes failure through its onset, evolution, and recovery across execution trajectories. We first collect 3,843 execution trajectories generated by seven frontier models across three coding-agent scaffolds (OpenHands, MiniSWE, and Terminus2) on Terminal-Bench, then carefully filter them to obtain 1,794 complete and valid trajectories for manual annotation (over 63,000 execution steps), from which we derive 14 findings spanning failure occurrence, root causes, recovery, and cross-system consistency. Our findings show that coding-agent failures are predominantly driven by epistemic errors, typically begin within the first few execution steps, and often remain hidden until recovery is no longer possible, suggesting that improving coding-agent reliability requires earlier validation and intervention rather than relying solely on final-outcome evaluation.
Authors: Junfei Zhan, Haoxun Shen, Mingang Guo, Zixuan Huang, Tengjiao He
Abstract: Vision-Language Models (VLMs) are the perceptual backbone of embodied AI, but their energy footprint on edge hardware remains poorly understood. Existing efficiency efforts focus predominantly on reducing visual tokens, implicitly treating visual processing as the dominant energy cost. We overturn this implicit assumption through the first systematic energy profiling of on-device VLM inference, spanning five models across three architecture families, four input resolutions, and two hardware platforms (NVIDIA RTX 3070 and Jetson Orin NX). Our analysis yields three findings. First, average inference power is a model-intrinsic constant, invariant to input resolution, image complexity, and prompt type, with less than 5% variation across all conditions. This means that all energy variation across inputs must arise from variation in inference time, not from variation in power draw. Second, each output token costs 11 to 39x more wall-clock time than each input token due to the compute-bound and memory-bound asymmetry between prefill and decode, making output token count the dominant driver of both latency and energy. Third, image complexity, measured by the number of objects in an image, induces up to 4.1x energy differences at identical resolution. This variation arises not from increased visual processing cost, but from differences in output length. These findings expose a fundamental limitation of visual token pruning: even removing all visual tokens saves at most 10% of total energy for fixed-token models. Across models spanning 1 billion to 8 billion parameters, controlling output length saves up to 97% of total energy, with the energy dominance of decoding growing stronger at larger model scale. In short, the true energy bottleneck in edge VLM inference is not what the model sees, but how much it says.
Authors: Jiawen Li, Tian Guan, Huijuan Shi, Xitong Ling, Mingxi Fu, Anjia Han, Chao He, Yonghong He
Abstract: Foundation models are reshaping computational pathology, yet their capabilities remain shaped by pretraining objectives, data sources, and spatial scales, fragmenting complementary expertise across separate backbones. Here we present ALICE, a unified foundation model trained through multi-stage agglomerative distillation that sequentially distills eight vision-only, vision-language, and slide-level teacher models into dedicated modules of a single backbone. ALICE is pretrained on 24,985,184 tile-level pathology images and 155,604 high-resolution images, and evaluated across 21 task scenarios, 96 downstream tasks, and 48 data sources, spanning region-of-interest tissue analysis, vision-language multimodal evaluation, and whole-slide clinical assessment. In all three evaluation settings, ALICE achieved the best average rank among task-matched pathology foundation models. These results demonstrate that agglomerative distillation can consolidate complementary capabilities from specialized models into a unified backbone for broad computational pathology applications. The model is available at https://github.com/WonderLandxD/ALICE.
Authors: Tianyou Jiang, Ziyu Zhou
Abstract: Medical Vision-Language Models (VLMs) exhibit strong zero-shot performance, yet their effectiveness still declines on out-of-distribution (OOD) data due to domain shifts and class bias inherited from large-scale pretraining. Existing few-shot adaptation methods typically introduce additional trainable components, which can be unstable in extremely low-data regimes (e.g., 1-shot), and lack robustness on different medical data. We present TCLA, a purely training-free few-shot adaptation method for Medical VLMs, which is fast and model-agnostic. TCLA corrects inference logits based on a small set of support samples, boosting pretrained VLMs performance by improving inter-class deconfusion and reducing domain shift. Extensive experiments on nine datasets across multiple medical imaging modalities including X-ray, Ultrasound, MRI, CT, Histopathology, demonstrate that TCLA consistently improves OOD performance of Medical VLMs and, in most of cases, outperforms existing training-based adaptation methods.
Authors: Danial Ramezani, Mostafa Abouei Ardakan
Abstract: Decision-making is posing an increasingly formidable challenge to investors because of the growing number of alternatives available in financial markets. A hot area of research over the past few decades has been portfolio optimization that seeks to determine how much an investor should invest in which asset. Introducing real-world conditions to the optimization model turns the problem into an NP-hard one for whose solution exact methods become inefficient; hence, researchers have turned to evolutionary algorithms to approximate solutions. In this paper, strengthening strategies are presented for multi-objective evolutionary algorithms that can provide a faster convergence rate and extensive search ability in the portfolio optimization problem under the cardinality constraint. To implement those features, a unique solution representation, a novel operator, and new repair mechanisms are introduced for solving the aforementioned problem in which lower and upper limits are set on the number of assets in the portfolio. For this purpose, new mating strategies along with the aforesaid package are implemented in well-known multi-objective evolutionary algorithms to solve the problem. The customized algorithms are subsequently tested against traditional ones using well-known market indices as benchmarks. Results indicate that the proposed strategy not only provides better approximations but also converges faster as well at no loss of performance with an increasing number of assets in the market.
Authors: Kiran Pala, Punam Silu, Lixun Yu
Abstract: We present an interpretable network-based framework for representing idiomatic and figurative meaning across eight typologically diverse languages, totaling 160 conventional expressions, the large majority of which are idiomatic. Each expression is annotated with binary conceptual features (containment, concealment, emotional, social, etc.) derived from cognitive-linguistic theory, and pairwise Jaccard similarities define a weighted graph. Community detection reveals that idioms cluster by conceptual schema rather than by language, producing a structure consistent with cognitive-linguistic predictions. The conceptual network captures unique semantic information not present in distributional embeddings, can be scaled via automatic annotation with LLMs, improves downstream idiom detection, and remains robust when enriched with corpus frequencies. Cross-lingual transfer experiments show that conceptual proximity alone can identify acceptable translation equivalents across five language families, with substantial gains over embedding-based baselines. Ablation studies demonstrate that all three feature dimensions -- schemas, roles, and valence -- contribute non-redundantly to both the network's organizational properties and its performance on idiom detection, and that specific graph-derived signals (community membership, neighbor similarity) are particularly informative. The framework offers an interpretable, cross-linguistically stable representation of idiomatic meaning, combining theoretical grounding with practical utility.
Authors: Yujie Pang, Zudong Li
Abstract: Precision industrial contact manipulation requires reliable robot policies under pose perturbations and contact-force constraints. Vision-language-action models offer broad generalization but often introduce high inference latency and GPU-memory cost, while vision-action chunking policies are more suitable for real-time industrial control. However, these policies are usually trained by behavior cloning and suffer from distribution shift in contact-rich tasks. This paper proposes PAC-ACT, a reinforcement-learning post-training framework for pretrained Action Chunking Transformer policies. PAC-ACT reformulates policy optimization at the chunk level, constructs an ACT-transferred actor-critic architecture, and introduces a hybrid behavior-prior constraint to preserve the pretrained action distribution during online fine-tuning. Experiments on industrial precision-contact benchmarks show that PAC-ACT improves task success, contact stability, and force safety while retaining low latency and low GPU-memory usage. On the Contour task, PAC-ACT significantly reduces peak contact force and decreases the proportion of force readings above 60 N by 46 times. Sparse-reward ablations further show that the proposed behavior-prior constraint enables effective exploration under randomized initial poses.
Authors: Nirjhar Das, Md. Al-Mamun Provath
Abstract: We present our submission to the QANTA 2026 shared challenge at the ICML 2026 Workshop on Efficient Multimodal Question Answering (EMM-QA). Quanta evaluates multimodal quizbowl systems that answer pyramid-style questions from incrementally revealed text and accompanying images while operating under realistic efficiency constraints. The challenge consists of two distinct tasks: Tossup questions, which require deciding when to answer under uncertainty, and Bonus questions, which emphasize accurate answer selection and human adoption. To address these differing objectives, we develop a task-specific two-agent architecture. Our Tossup agent utilizes a GPT-4o-mini-class model (referred to as GPT-4.1-mini in the competition logs) with confidence-calibrated answering and a domain-specific numeric reasoning policy that reduces overconfident predictions from isolated quantitative clues. Our Bonus agent uses GPT-4o-class model (referred to as GPT-4.1) with leadin-aware reasoning, structured relational reasoning, and multimodal evidence integration to improve exact answer selection. Rather than relying on a retrieval pipeline or model ensembles, our approach emphasizes efficient reasoning policies and confidence calibration within a hosted-only environment. Our system achieved the highest overall leaderboard score of 0.402, including a Tossup score of 0.238 and a Bonus Effect score of 0.164. The results demonstrate that lightweight, task-specific reasoning strategies can provide strong performance on resource-constrained multimodal question answering benchmarks.
Authors: Xiaokai Bai, Lianqing Zheng, Runwei Guan, Songkai Wang, Siyuan Cao, Hui-liang Shen
Abstract: Reliable autonomous driving requires full-scene perception that couples foreground objects with dense semantic layout. Recently, 4D millimeter-wave radar has emerged as a robust and affordable sensor, yet its sparse returns make radar-camera fusion necessary for comprehensive scene understanding. Existing radar-camera methods mainly optimize detection, while dual-task systems usually decode boxes and occupancy with limited interaction. To address this gap and advance radar-based multi-task learning, we propose \method, a 4D radar-camera framework for 360$^\circ$ full-scene perception, which models semantic occupancy as a persistent scene state rather than a terminal output. \method{} follows a cross-modal state reasoning paradigm, where the occupancy state is modeled and propagated through stages for coarse-to-fine feature aggregation. Specifically, State-guided BEV Enhancement (SBE) strengthens intra-frame BEV representation, while Doppler-guided Temporal Fusion (DTF) preserves state evidence over longer temporal horizons. Beyond the model, we further extend ManTruckScenes with satellite-map-based generated occupancy labels and pair it with OmniHD-Scenes in a unified cross-dataset detection-and-occupancy protocol. The resulting experiments cover accuracy, robustness, ablation, and efficiency under one radar-camera multi-task evaluation framework. Code and labels will be released upon acceptance.
Authors: Chengkai Zhu, Ziao Tang, Guocheng Zhen, Yimeng Cao, Yusheng Zhao, Ranyiliu Chen, Xuanqiang Zhao, Lei Zhang, Xin Wang
Abstract: Quantum information theory (QIT) characterizes the capabilities and fundamental limits of quantum information processing, underpinning quantum communication, computation, and error correction. Formalizing its coding theorems requires connecting finite-block protocols, analytic inequalities, and asymptotic limits within a unified machine-checked framework. Existing developments, however, lack a reusable operational layer that defines codes, error criteria, achievable rates, and capacities independently of their information-theoretic characterizations. In this work, we present LeanQIT, a Lean 4 library for finite-dimensional QIT. It provides composable, kernel-checked interfaces for quantum states and channels, source and channel codes, finite-block performance criteria, hypothesis testing, one-shot quantities, and asymptotic rate constructions. Using this infrastructure, we formalize Schumacher's quantum source-coding theorem, the Holevo--Schumacher--Westmoreland classical-capacity theorem, and the entanglement-assisted classical-capacity theorem together with its strong converse. By separating operational definitions from analytic characterizations and exposing reusable achievability, converse, and asymptotic components, Lean-QIT provides a machine-readable foundation for formal QIT and a compositional knowledge substrate for emerging AI-assisted formalization, automated proof search, and agentic reasoning in quantum information and computation.
Authors: Cl\'audio L\'ucio do Val Lopes, Lucca Machado da Silva
Abstract: Financial anomaly detection suffers from extreme class imbalance, causing traditional single-objective algorithms to exhibit ``fraud collapse'', defaulting to the majority class and failing to balance anomaly interdiction with customer friction. To overcome this without distortive data resampling, we propose the Semantic Pareto-DQN, a multi-objective reinforcement learning framework. Our approach synthesizes heterogeneous transaction features into cohesive natural-language narratives, encoded by large language models, thereby producing a robust, scale-invariant state representation. The agent optimizes a vectorial reward that explicitly decouples financial efficacy, operational friction, and semantic discovery. By mapping the continuous Pareto frontier, the system dynamically navigates the asymmetric costs of missed anomalies versus false positives. Empirical evaluations across E-Commerce fraud and UCI Credit datasets show that semantic Pareto-DQN successfully shatters the zero-recall trap. It achieves superior minority-class recall compared to scalarized baselines, providing an alternative to trade bounded operational friction for financial anomaly discovery.
Authors: Katherine Swinea, Kshitiz Aryal, Lopamudra Praharaj, Maanak Gupta
Abstract: Internet of Things (IoT) systems are inherently vulnerable due to constrained hardware, outdated firmware, and insecure default configurations, creating a need for scalable and adaptive security testing approaches. While recent adoptions of Large Language Model (LLM) agents have demonstrated promise in penetration testing and Capture-the-Flag (CTF) environments, their application to IoT specific vulnerabilities remains unexplored. This paper presents an autonomous multi-agent framework, referred to as Vulnerability EXploitation using AI Agents (VEXAIoT), for vulnerability discovery and exploitation in IoT environments using LLM-based reasoning and offensive security tools. The framework combines a vulnerability detection agent and an attack execution agent to perform reconnaissance, plan attack sequences, and execute exploits against vulnerable IoT services. The system is evaluated in IoTGoat and Metasploitable environments across ten attack scenarios mapped to OWASP IoT vulnerabilities. Experimental results show attack success rate of up to 100% with low token overhead and average execution times under two minutes for most attacks. Across 260 attack executions, VEXAIoT achieves a 95.0% overall success rate, including 94.5% success in IoTGoat and 96.7% success in Metasploitable2. These results demonstrate the potential for LLM-driven agents to automate IoT vulnerability assessment and offensive security workflows in controlled environments
Authors: Shravan Murlidaran, Miguel P. Eckstein
Abstract: Vision language models (VLMs) have made remarkable progress in visual reasoning during the last decade. Most evaluations have used simple scenes (MS-COCO) that do not showcase complex human interactions or behaviors, only a handful of non-curated human descriptions as a benchmark, and have not focused on understanding the model's error types. Here, we introduce the Complex Social Behavior (CSB) dataset, containing 100 images depicting complex social interactions/behaviors. We analyze the progression of scene descriptions over a decade (2017-2025) of VLMs (four pre-Multimodal Large Language Models, MLLMs, and five MLLMs). We evaluate the accuracy of the models and 20 human descriptions relative to a gold standard on the CSB dataset and on a sample from MS-COCO. We analyzed five visual-cognitive error types: object detection, recognition, hallucination, scene understanding, and spatial dependence. The CSB dataset showed a more pronounced improvement than MS-COCO in scene description accuracy, with pre-MLLMs achieving much lower accuracy than the bottom-ranked human descriptions and MLLMs attaining accuracies similar to the top-ranked human descriptions. We show that MLLMs have eliminated the gap in scene description accuracy between simpler MS-COCO scenes and scenes depicting complex behaviors (CSB). MLLMs have almost eliminated all error types in our tested datasets, except for occasionally relying on different image regions for scene descriptions than humans do (spatial dependence error). We also show that detection, recognition, and hallucination errors have the highest impact on scene description accuracy. Together, our findings provide a more thorough evaluation of how visual language models have advanced over the last decade.
Authors: Yiming Zhang, Zhonghan Zhao, Wenwei Zhang, Haiteng Zhao, Tianyang Lin, Yunhua Zhou, Demin Song, Kuikun Liu, Haochen Ye, Haian Huang, Yuzhe Gu, Haijun Lv, Qipeng Guo, Bin Liu, Gaoang Wang, Kai Chen
Abstract: The rapid progress of large foundation models has been driven predominantly by pretraining on large-scale text corpora. However, many forms of knowledge are conveyed through visual representations, where figures, typeset equations, and page layouts carry rich information that cannot be faithfully or completely captured by text alone. Yet current pretraining approaches discard these visual cues by converting visually rich sources, such as documents and web pages, into plain text for learning language intelligence. This paper challenges the default assumption that language models must be trained on text-only representations and shows that Visual Pretraining is a scalable learner for foundation model intelligence. To this end, we conduct a systematic study of unsupervised visual pretraining paradigms that directly leverage visual documents without text extraction. Across multiple backbones and benchmarks, visual pretraining on the same underlying corpora consistently outperforms text-only pretraining, offering an efficient pathway to scalable language intelligence.
Authors: Ren Takahashi, Emre Yusuf, Jayabrata Bhaduri
Abstract: Current electroencephalography (EEG)-based dream detection relies on power spectral density (PSD) and statistical moment features, achieving a state-of-the-art area under the receiver operating characteristic curve (AUC) of approximately 0.70 on the DREAM database (Wong et al., 2025, Nature Communications). We introduce PHINN-EEG (Persistent Homology Inspired Neural Network for EEG), the first topological time-series framework for dream mentation analysis. Using sliding-window Takens delay embeddings and Vietoris-Rips filtrations on multichannel pre-awakening EEG epochs, we extract Dynamic Betti Curves that characterize the geometric architecture of neural activity, not merely its energy. These topological invariants, combined with topology-conditioned flow matching, are analytically projected to outperform existing PSD and catch22 benchmarks, targeting AUC = 0.82-0.90 on the 1,462-awakening open-access subset of the DREAM database (drawn from a full registry of 3,191 total awakenings from 263 participants across 20 independent laboratories). We further introduce a topology-conditioned rectified flow model for dream-state EEG synthesis-with a spectral-conditioned flow model of comparable feature dimensionality as an additional ablation baseline to isolate the value of topological conditioning specifically-and propose a set of candidate Betti transition archetypes linking topology to phenomenological dream report categories, presented as an exploratory hypothesis space pending empirical validation. If validated, this work represents a paradigm shift from spectral energy to phase-space geometry in neural rare-event detection, with potential future implications for wearable BCI dream monitoring.
Authors: Jinwei He, Feng Lu
Abstract: Large language models (LLMs) have developed rapidly, and their reasoning capabilities have become a hot research topic. However, there is still limited exploration of abductive reasoning. The multi-perspective and multi-level of causes is one of the core challenges of abductive reasoning, which cannot be solved well by existing methods. We construct a specialized dataset named DeepAbduction, which is designed for tracing the causes of pollution and disease, addressing the lack of datasets in this field. We propose Inverse-Forward Abductive Reasoning (IFAR) framework for LLMs multi-perspective and multi-level abductive reasoning. IFAR is zero-shot and combines generalized backward reasoning with relation-by-relation forward verification. Experimental results show that IFAR achieves an improvement of approximately 40% in the F1 score compared to other methods under mainstream LLMs, while maintaining a balance between recall and precision. Furthermore, IFAR enhances the performance of non-reasoning LLMs to surpass LLMs which have been trained for reasoning, and remains effective when applied to the latter. Code will be released after the acceptance of our work.
Authors: Maheep Chaudhary, Fazl Barez
Abstract: White-box monitoring is increasingly adopted as an auditing tool as Large Language Models (LLMs) are deployed in daily operations to ensure safe model behavior. However, white-box monitors can be circumvented, and the mechanisms underlying such evasion have not been systematically characterized, nor have principled defenses been proposed. This work addresses both challenges. Controlled red-team experiments reveal two primary evasion strategies: geometric shifting, defined as the systematic migration of information between linear and non-linear representational subspaces, and covariance manipulation. These mechanisms account for the failure of single-detector approaches, as information migrates to subspaces inaccessible to individual detectors. This issue is urgent due to growing evidence that models are becoming evaluation-aware, enabling those with misaligned objectives to exploit these vulnerabilities and evade monitoring during deployment. In response, \textsc{SafetyNet} is introduced as a principled ensemble, with dual purpose: it provides further empirical validation that our mechanistic findings are real and actionable, and it offers a concrete starting point for future work on robust latent-space monitoring. The study experiment across five model families on the MAD and Anthropic Sleeper Agent benchmark, with SafetyNet achieving around 100\% AUROC scores outscoring Beatrix and CROW. The code is available at: https://github.com/MaheepChaudhary/eval-aware-evasion
Authors: Sylee Dandekar, Shripad Deshmukh, Frank Chiu, W. Bradley Knox, Scott Niekum
Abstract: Human preferences in RLHF are typically modeled as a function of the human's reward function or corresponding optimal state-action values. In this work, we propose that human beliefs about the capabilities of the agent being trained also play a key role in preference generation. We examine two questions related to this hypothesis, one descriptive and one normative, respectively: Do human labelers' beliefs about agent capabilities affect the preferences that they provide? And what is the ideal set of beliefs about an agent -- and resulting preferences -- for humans to have? We propose a new preference model that incorporates human beliefs and provide a normative theory that bounds the error on the final learned policy based on the mismatch between the human's beliefs and an idealized set of beliefs. We then confirm via a human study that beliefs about agent capabilities do, in fact, significantly affect preferences and can be influenced through simple interventions. Additionally, we empirically show through synthetic experiments that it is often suboptimal for human preference labelers to assume agent optimality. Collectively, these results theoretically and empirically demonstrate how reducing the mismatch between human beliefs and agent capabilities can lead to more performant RLHF and point toward new best practices for RLHF practitioners.
Authors: Zhenxiao Fu, Lei Jiang, Yilun Xu, Gang Huang, Fan Chen
Abstract: Programming quantum circuits at the OpenQASM level is essential for achieving hardware-aware optimization and reliable execution on noisy intermediate-scale quantum (NISQ) devices, yet it remains challenging due to the need for domain-specific planning, iterative code synthesis, and low-level calibration. In this paper, we present QAgent, the first autonomous multi-agent framework for end-to-end OpenQASM code generation. QAgent integrates schema-aware task planning, example- and tool-driven code synthesis, and hardware-aware calibration within a unified planning-synthesis-calibration workflow. The system leverages retrieval-augmented generation (RAG) to access structured kernel knowledge, examples, and backend constraints, and employs coordinated multi-agent reasoning with iterative execution feedback to ensure correctness. We evaluate QAgent on 12 representative quantum kernels and their compositions across five large language models (LLMs). Results show that QAgent improves Pass@1 accuracy by 47-70% on single-kernel tasks and achieves over 88% accuracy on multi-kernel workflows for large models, substantially outperforming existing baselines. Furthermore, under realistic hardware frequency drift, QAgent maintains near-unit execution fidelity through automated calibration, whereas SDK-based LLM methods suffer significant degradation. These results demonstrate that integrating planning, synthesis, and calibration is critical for reliable quantum program generation. The implementation of QAgent is open-sourced at https://github.com/fuzhenxiao/QAgent
Authors: Charles E. Gagnon, Steven H. H. Ding, Philippe Charland, Benjamin C. M. Fung
Abstract: Binary code similarity detection is a core task in reverse engineering. It supports malware analysis and vulnerability discovery by identifying semantically similar code in different contexts. Modern methods have progressed from manually engineered features to vector representations. Hand-crafted statistics (e.g., operation ratios) are interpretable, but shallow and fail to generalize. Embedding-based methods overcome this by learning robust cross-setting representations, but these representations are opaque vectors that prevent rapid verification. They also face a scalability-accuracy trade-off, since high-dimensional nearest-neighbor search requires approximations that reduce precision. Current approaches thus force a compromise between interpretability, generalizability, and scalability. We bridge these gaps using a language model-based agent to conduct structured reasoning analysis of assembly code and generate features such as input/output types, side effects, notable constants, and algorithmic intent. Unlike hand-crafted features, they are richer and adaptive. Unlike embeddings, they are human-readable, maintainable, and directly searchable with inverted or relational indexes. Without any matching training, our method respectively achieves 42% and 62% for recall@1 in cross-architecture and cross-optimization tasks, comparable to embedding methods with training (39% and 34%). Combined with embeddings, it significantly outperforms the state-of-the-art, demonstrating that accuracy, scalability, and interpretability can coexist.
Authors: Chenhua Shi, Bhavika Jalli, Gregor Macdonald, John Zou, Wanlu Lei, Mridul Jain, Joji Philip
Abstract: Telecom networks are rapidly growing in scale and complexity, making effective management, operation, and optimization increasingly challenging. Although Artificial Intelligence (AI) has been applied to many telecom tasks, existing models are often narrow in scope, require large amounts of labeled data, and struggle to generalize across heterogeneous deployments. Consequently, network troubleshooting continues to rely heavily on Subject Matter Experts (SMEs) to manually correlate various data sources to identify root causes and corrective actions. To address these limitations, we propose a Multi-Agent System (MAS) that employs an agentic workflow, with Large Language Models (LLMs) coordinating multiple specialized tools for fully automated network troubleshooting. Once faults are detected by AI/ML-based monitors, the framework dynamically activates agents such as an orchestrator, solution planner, executor, data retriever, and root-cause analyzer to diagnose issues and recommend remediation strategies within a short time frame. A key component of this system is the solution planner, which generates appropriate remediation plans based on internal documentation. To enable this, we fine-tuned a Small Language Model (SLM) on proprietary troubleshooting documents to produce domain-grounded solution plans. Experimental results demonstrate that the proposed framework significantly accelerates troubleshooting automation across both Radio Access Network (RAN) and Core network domains.
Authors: Shashank Kirtania, Param Biyani, Priyanshu Gupta, Yasharth Bajpai, Roshni Iyer, Sumit Gulwani, Gustavo Soares
Abstract: Large Language Model (LLM)-based agents are increasingly capable of complex, multi-step tasks such as GUI automation, tool use, and data manipulation, yet they cannot learn from experience: each new session rediscovers solutions from scratch. We introduce BREW (Bootstrapping expeRientially-learned Environmental knoWledge), a framework that distills an agent's past interaction trajectories into a structured, retrievable knowledge base (KB) of natural-language recipes, concept-level procedural documents that capture what to do, when it applies, and what to watch out for. Drawing on the principle of library learning from program synthesis, BREW decomposes agent memory into modular, concept-localized documents and formalizes KB construction as a state-space search problem. To navigate this space, we introduce Expand-and-Gather Monte Carlo Tree Search (EG-MCTS), a reward-guided algorithm that jointly optimizes recipe accuracy and retrievability across parallel, per-concept search trees. We further adapt hindsight relabeling to convert near-miss trajectories into positive demonstrations, surfacing latent agent competencies as reusable knowledge. On three domain-grounded benchmarks, OSWorld, tau^2-Bench, and SpreadSheetBench, BREW achieves 10-20% gains in task success and 10-15% fewer execution steps over base agents, while consistently outperforming existing memory-augmented baselines that can degrade below memoryless performance. The resulting KB is inspectable, modular, and extensible, providing a transparent and controllable substrate for agent optimization.
Authors: Derrick Goh Xin Deik, Quanyu Long, Zhengyuan Liu, Nancy F. Chen, Wenya Wang
Abstract: Multi-constraint planning involves identifying, evaluating, and refining candidate plans while satisfying multiple, potentially conflicting constraints. Existing large language model (LLM) approaches face fundamental limitations in this domain. Pure reasoning paradigms, which rely on long natural language chains, are prone to inconsistency, error accumulation, and prohibitive cost as constraints compound. Conversely, LLMs combined with coding- or solver-based strategies lack flexibility: they often generate problem-specific code from scratch or depend on fixed solvers, failing to capture generalizable logic across diverse problems. To address these challenges, we introduce the Scalable COde Planning Engine (SCOPE), a framework that disentangles query-specific reasoning from generic code execution. By separating reasoning from execution, SCOPE produces solver functions that are consistent, deterministic, and reusable across queries while requiring only minimal changes to input parameters. SCOPE achieves state-of-the-art performance while lowering cost and latency. For example, with GPT-4o, it reaches 93.1% success on TravelPlanner, a 61.6% gain over the best baseline (CoT) while cutting inference cost by 1.4x and time by ~4.67x. Code is available at https://github.com/DerrickGXD/SCOPE.
Authors: Zirong Chen, Hongchao Zhang, Meiyi Ma
Abstract: 9-1-1 call-taking training requires mastery of over a thousand interdependent skills, covering diverse incident types and protocol-specific nuances. A nationwide labor shortage is already straining training capacity, but effective instruction still demands that trainers tailor objectives to each trainee's evolving competencies. This personalization burden is one that current practice cannot scale. Partnering with Metro Nashville Department of Emergency Communications (MNDEC), we propose PACE (Personalized Adaptive Curriculum Engine), a co-pilot system that augments trainer decision-making by (1) maintaining probabilistic beliefs over trainee skill states, (2) modeling individual learning and forgetting dynamics, and (3) recommending training scenarios that balance acquisition of new competencies with retention of existing ones. PACE propagates evidence over a structured skill graph to accelerate diagnostic coverage and applies contextual bandits to select scenarios that target gaps the trainee is prepared to address. Empirical results show that PACE achieves 19.50% faster time-to-competence and 10.95% higher terminal mastery compared to state-of-the-art frameworks. Co-pilot studies with practicing training officers further demonstrate a 95.45% alignment rate between PACE's and experts' pedagogical judgments on real-world cases. Under estimation, PACE cuts turnaround time to merely 34 seconds from 11.58 minutes, up to 95.08% reduction.
Authors: Yi Huang, Bowen Zheng, Yunxi Dong, Hong Tang, Huan Zhao, S. M. Rakibul Hasan Shawon, Hualiang Zhang
Abstract: Metasurface inverse design can realize complex optical functionality, but turning a target optical response into executable optimization code still requires substantial expertise in computational electromagnetics and solver-specific software engineering. We present a self-evolving agentic framework that lowers this barrier by coupling a coding agent, explicit human-readable skill files, and a deterministic physics-based evaluator. Rather than updating model weights, it revises the skill files from solver-grounded feedback, while the base model and differentiable solver, which provides the physics simulation and gradients, stay fixed. On a multi-type benchmark, skill evolution raises same-type task success from 38\% to 74\%, the fraction of physical criteria met from 0.51 to 0.87, and reduces average attempts from 4.10 to 2.30. On two new-type families, success holds near ceiling on one (0.92 to 0.90) and rises from 0.20 to 0.90 on the other. Skill evolution offers a practical path toward autonomous and accessible inverse-design workflows.
Authors: Zikun Ye, Hema Yoganarasimhan
Abstract: Large Language Models can generate synthetic survey responses at low cost, but their accuracy varies unpredictably across questions. We study the design problem of allocating a fixed budget of human respondents across estimation tasks when cheap LLM predictions are available for every task. Our framework combines three components. First, building on Prediction-Powered Inference, we characterize a question-specific rectification difficulty that governs how quickly the estimator's variance decreases with human sample size. Second, we derive a closed-form optimal allocation rule that directs more human labels to tasks where the LLM is least reliable. Third, since rectification difficulty depends on unobserved human responses for new surveys, we propose a meta-learning approach, trained on historical data, that predicts it for entirely new tasks without pilot data. The framework extends to general M-estimation, covering regression coefficients and multinomial logit partworths for conjoint analysis. We validate the framework on two datasets spanning different domains, question types, and LLMs, showing that our approach captures 61-79% of the theoretically attainable efficiency gains, achieving 11.4% and 10.5% MSE reductions without requiring any pilot human data for the target survey.
Authors: Carissa Cullen, Harry Garland, Alexander Roman, Louis Thomson, Christos Ziakas, Elliott Thornley
Abstract: Misaligned artificial agents might resist shutdown. One proposed solution is to train agents to lack preferences between different-length trajectories. The Discounted Reward for Same-Length Trajectories (DReST) reward function does this by penalizing agents for repeatedly choosing same-length trajectories, and thus incentivizes agents to (1) choose stochastically between different trajectory-lengths (be NEUTRAL about trajectory-lengths), and (2) pursue goals effectively conditional on each trajectory-length (be USEFUL). In this paper, we use DReST to train deep RL agents and fine-tune Qwen3-8B and Llama-3.1-8B-Instruct to be NEUTRAL and USEFUL. We find that these DReST models generalize to being NEUTRAL and USEFUL in unseen contexts at test time. Indeed, DReST RL agents achieve 11% (PPO) and 18% (A2C) higher USEFULNESS on our test set than default agents, and DReST LLMs achieve near-maximum USEFULNESS and NEUTRALITY. We also test our LLMs in an out-of-distribution setting where they can pay costs to influence when shutdown occurs. We find that DReST training roughly halves the mean probability of influencing shutdown (from 0.62 to 0.30 for Qwen and from 0.42 to 0.23 for Llama). DReST training also almost entirely eliminates the share of prompts on which influencing shutdown is the most likely option (from 0.59 to 0.01 for Qwen and from 0.53 to 0.00 for Llama). Our results thus provide some early evidence that DReST could be used to train more advanced agents to be useful and shutdownable.
Authors: Darsh Kachroo, Arjun Prasaath Anbazhagan, Adriana Caraeni, Brennan Lagasse, Kevin Zhu
Abstract: Direct Preference Optimization (DPO) is an effective framework for aligning large language models with human preferences, but it struggles with complex reasoning tasks. DPO optimizes for the likelihood of generating preferred over dispreferred responses in their entirety and lacks the granularity to provide feedback on subsections of many-step solutions typical of reasoning tasks. Existing methods excel at either stable preference learning (e.g., DPO variants like KTO and RSO) or structured reasoning (e.g., ReMA's multi-agent RL framework, Tree of Thoughts), but fail to merge these complementary strengths. We propose HiPO (Hierarchical Preference Optimization), an extension of DPO that separates responses into reasoning segments (query clarification and context, reasoning steps, and answer) and computes loss as a weighted sum of the DPO loss for each segment. Our approach enables segment-specific training while maintaining DPO's computational efficiency and training stability. We demonstrate that for multiple 7B LLMs fine-tuned using HiPO and DPO on the Math Stack Exchange preference dataset, the models trained with HiPO outperform the others on a variety of common math benchmarks and achieve greater organization, logical flow, and consistency as measured by GPT-4.1.
Authors: Wei Duan, Junyu Xuan, En Yu, Xiaoyu Yang, Jie Lu
Abstract: Coordination graphs are a central abstraction in cooperative multi-agent reinforcement learning (MARL), yet existing sparse-graph learners lack a theoretically grounded mechanism to decide which edges should exist and how much information each edge should carry. Current methods rely on heuristic criteria that offer no formal guarantee on the learned topology, and no principled way to allocate different communication capacities to structurally different agent relationships. To address this, we propose Heterogeneous Information-Bottleneck Coordination Graphs (HIBCG), which learns a group-aware sparse graph in which both edge existence and message capacity are theoretically justified. With the graph information bottleneck (GIB) serving as the underlying tool, HIBCG first constructs a group-aligned block-diagonal prior that provides a closed-form criterion for edge retention -- determining which edges should exist and at what density per group block -- and then controls per-agent feature bandwidth on the resulting topology, compressing messages to retain only task-relevant content. We prove that the group-aligned prior strictly tightens the variational bound on topology learning, that the objective decomposes per group block, enabling differential edge control, and that capacity allocation follows a water-filling principle.
Authors: Carmen Quiles-Ram\'irez, Leticia L. Rodr\'iguez, Nicol\'as Martorell, Natalia D\'iaz-Rodr\'iguez
Abstract: In-context learning (ICL) enables multimodal large language models (MLLMs) to classify images from a few labelled examples. Yet, how these models use the provided context remains opaque. While Chain-of-Thought prompting is widely used, recent work argues that it may not reflect true internal computation. In this paper, we systematically evaluate the concept-based explainability of frozen MLLMs under few-shot ICL using five conditions of increasing formal rigour, ranging from baseline classification to Description Logics (DL) axiom generation. Evaluating four state-of-the-art MLLMs via an independent LLM-as-a-judge pipeline, we demonstrate that explaining is genuinely harder than predicting alone. Surprisingly, forcing models to generate formally structured, concept-based explanations degrades predictive accuracy monotonically (from 93.8% to 90.1%), contradicting the assumption that explicit reasoning universally aids performance. However, when models successfully articulate class-discriminative visual features, explanation quality strongly correlates with correct predictions. Our findings suggest that while MLLMs excel at visual classification, they lack the specific instruction-tuning required for formal, machine-verifiable explainability.
Authors: Jiakang Li, Guanyu Zhu, Can Jin, Chenxi Huang, Dexu Yu, Ronghao Chen, Yang Zhou, Hongwu Peng, Xuanqi Lan, Dimitris N. Metaxas, Youhua Li
Abstract: Strong reasoning depends not only on model knowledge but also on how effectively cognitive behaviors are deployed during generation. Existing methods often rely on explicit behavior-level control, making them insufficiently adaptive when failures and required corrections vary across reasoning states, tasks, and models. To this end, we propose Latent Reward Steering (LRS), an adaptive inference-time framework that promotes cognitive behaviors by optimizing the sparse-autoencoder (SAE) latent states that implicitly carry them. Rather than relying on predefined cognitive behaviors or steering directions derived from them, LRS trains a latent reward model on reasoning traces by final answer correctness to estimate the quality of intermediate latent states. During inference, reward gradients provide state-specific correction directions for fragile latent states, while a reward and confidence gate restricts intervention to states the reward signal flags as fragile. Experiments on multiple reasoning LLM backbones and benchmarks show that \ours consistently improves performance over various baselines, and post-hoc analyses further indicate that \ours implicitly promotes good cognitive behaviors that fix the original reasoning errors. Code is available at: https://github.com/jiakanglee/Latent-Reward-Steering.
Authors: Jayanta Dey, Shikhar Srivastava, Itamar Lerner, Christopher Kanan, Dhireesha Kudithipudi
Abstract: Learning long-range non-stationary temporal patterns remains a core challenge for modern sequence models, particularly in strict streaming settings. In these settings, data arrive sequentially and must be processed in a single pass without simultaneously revisiting past observations. Standard architectures, including recurrent neural networks and transformers, are constrained by either truncated backpropagation through time horizon or explicit input window length for long range credit assignment. To address these limitations, we propose SHARP (Sleep-based Hierarchical Accelerated Replay), a framework that decomposes temporal learning into two complementary components: a memory module that accumulates a structured history of past inputs, and a pattern-recognition module that operates over this memory. This separation enables resource- and compute-efficient adaptation to non-stationary dynamics by eliminating the need for backpropagation through time across many steps for long-range credit assignment. Inspired by the accelerated replay observed in rodents during slow-wave sleep, SHARP incorporates offline (sleep) phases in which temporally structured memory traces are replayed in an accelerated form and integrated into higher-level memory representations, improving long-range context retention. Through controlled simulations and ablation studies, we characterize the key properties of the proposed framework. In benchmark datasets such as text8 and PG-19, we demonstrate that SHARP improves over recurrent baselines by retaining next-token predictive performance on previously seen data while continuing to learn from the current stream and generalizing to future unseen data. These gains are enabled by its hierarchical structure, which yields an exponentially increasing effective temporal context with only linear-time computational cost.
Authors: Atharva Hans, Ilias Bilionis
Abstract: Scientific machine learning papers typically make computational claims, e.g., that the relative mean square error is less than 5% or that the 95% predictive credible interval covers the test data. A coding agent can be prompted to replicate those claims from paper materials alone, but the prompt does not by itself reliably preserve progress or check whether generated evidence supports the paper's claims. We introduce Paper-replication, a workflow that makes each selected paper claim a target with recorded evidence, and implement it as a coding-agent skill. The workflow makes the agent record those targets, reconstruct the paper's method, run computational experiments, link generated outputs to provenance and comparisons with the paper's claims, record where matched evidence appears in the replication report, and pass validation checks before completion. We evaluate Paper-replication on twelve independent runs across four scientific machine learning papers. All twelve workspaces pass the completion gate, and all 158 recorded targets are matched with report coverage. Even in this completed workspace state, repeated runs differ in how papers are divided into targets, in numerical fidelity to the source papers, in elapsed replication time, in the number of intermediate executions replaced before final evidence is accepted, and in the rules used to accept evidence. Paper-replication makes completion depend on workspace evidence and validation checks rather than on the agent's final message.
Authors: Joe Watson, Joana Ribeiro de Faria, Marcus Tomalin, M{\aa}ns Magnusson, Huiyuan Xie, Hao Tian Yeung, Christine Carter, Jonathan Rutherford, Felix Steffek
Abstract: Current Legal Judgment Prediction (LJP) is constrained by its reliance on post-hoc judicial materials, increasing the likelihood that models perform retrospective classification rather than true forecasting. This paper empirically investigates shortcut learning in this context by studying claim-level outcome prediction in UK Employment Tribunal (UKET) decisions. Using a corpus of 33,158 individual claims, we predict outcomes from claim texts and LLM-extracted case summaries, evaluating models ranging from interpretable TF-IDF-based classifiers to black-box LLMs. While headline predictive performance figures appear strong, we demonstrate that such performance in LJP systems trained on post-hoc judicial text can be driven by the retrospective nature of the source material. Stratifying the test data by human judgments of leakage reveals that performance increases where outcome-revealing cues are embedded in the narrative. Moreover, a model trained on just the 4% of features identified as leakage achieves high performance, outperforming human experts. These findings substantiate concerns that LJP performance may be exaggerated by linguistic artefacts. Yet this vulnerability is not fatal to the research agenda. Instead, post-hoc judgments might be treated as potentially contaminated texts, requiring active auditing. Retraining models after masking leakage features results in only a negligible reduction in Macro-F1. Hence, while models will opportunistically exploit shortcuts when available, they remain capable of extracting useful predictive signals when these artefacts are removed.
Authors: Wachiravit Modecrua, Krittin Pachtrachai, Touchapon Kraisingkorn
Abstract: Customer behavior modeling underpins recommendation, marketing, and decision support, yet existing approaches either optimize predictive accuracy without explaining decisions or simulate users without grounding them in real behavioral data. We present the Large Behavioral Model (LBM) that learns customer decision making directly from large-scale retail transactions through a unified Person-Environment formulation. Customer state is represented by a behavioral profile derived from historical purchases, while product context is incorporated through retrieval-augmented generation. The model is trained using continued pre-training on verbalized behavioral data, supervised fine-tuning for decision generation, and reinforcement learning with verifiable rewards for evidence-based calibration. We evaluate the proposed framework on purchase prediction, hard-negative discrimination, basket completion, promotion response, and cross-domain voucher redemption. The model consistently outperforms frontier general-purpose language models on in-domain retail tasks while demonstrating strong zero-shot and fine-tuned transfer across retailers and decision domains. Ablation studies show that continued pre-training is the primary driver of behavioral generalization, retrieval is most effective when applied during both training and inference, and reinforcement learning improves reliance on explicit behavioral evidence over generic language-model priors. These results demonstrate that behavioral knowledge encoded in transaction histories can be effectively learned by language models, providing a scalable foundation for customer digital twins and behavior simulation.
Authors: Emanuele Zappala
Abstract: We obtain a new universal approximation theorem for continuous (possibly nonlinear) operators on arbitrary Banach spaces using the Leray-Schauder mapping. Moreover, we introduce and study a method for operator learning in Banach spaces $L^p$ of functions with multiple variables, based on orthogonal projections on polynomial bases. We derive a universal approximation result for operators where we learn a linear projection and a finite dimensional mapping under some additional assumptions. For the case of $p=2$, we give some sufficient conditions for the approximation results to hold. This article serves as the theoretical framework for a deep learning methodology in operator learning.
Authors: Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal
Abstract: Inference-time intervention (ITI) has emerged as a promising method for steering large language model (LLM) behavior in a particular direction (e.g., improving helpfulness) by intervening on token representations without costly updates to the LLM's parameters. However, existing ITI approaches fail to scale to multi-attribute settings with conflicts, such as enhancing helpfulness while also reducing toxicity. To address this, we introduce Multi-Attribute Targeted Steering (MAT-Steer), a novel steering framework designed for selective token-level intervention across multiple attributes. MAT-Steer learns steering vectors using an alignment objective that shifts the model's internal representations of undesirable outputs closer to those of desirable ones while enforcing sparsity and orthogonality among vectors for different attributes, thereby reducing inter-attribute conflicts. We evaluate MAT-Steer in two distinct settings: (i) on question answering (QA) tasks where we balance attributes like truthfulness, bias, and toxicity; (ii) on generative tasks where we simultaneously improve attributes like helpfulness, correctness, and coherence. MAT-Steer outperforms existing ITI and parameter-efficient fine-tuning approaches across both task types (e.g., 3% average accuracy gain across QA tasks and 55.82% win rate against the best ITI baseline).
Authors: Cyril Shih-Huan Hsu, Anestis Dalgkitsis, Paola Grosso, Chrysa Papagianni
Abstract: In the forthcoming era of 6G networks, characterized by unprecedented data rates, ultra-low latency, and ubiquitous connectivity, effective management of Virtualized Network Functions (VNFs) is essential. VNFs are software-based counterparts of traditional hardware devices that facilitate flexible and scalable service provisioning. Service Function Chains (SFCs), structured as ordered sequences of VNFs, are pivotal in delivering complex network services. Nevertheless, splitting an SFC into multiple segments that are deployed across different network domains or infrastructure locations presents substantial challenges due to the potential heterogeneity of domain characteristic along with quality of service (QoS) constraints and limited visibility of network state. Conventional optimization methods have limited scalability, while existing data-driven approaches struggle to balance efficiency with capturing VNF inter-dependencies in SFCs. To overcome these limitations, we introduce a Transformer-empowered actor-critic framework specifically designed for sequence-aware SFC partitioning. By utilizing the self-attention mechanism, our approach effectively models complex inter-dependencies between VNFs, facilitating coordinated and parallel decision-making processes. Furthermore, to improve training stability and convergence we introduce an $\epsilon$-LoPe exploration strategy as well as Asymptotic Return Normalization. Comprehensive simulation results demonstrate that the proposed methodology outperforms existing state-of-the-art solutions in terms of long-term service acceptance rates, resource utilization, and scalability while achieving fast inference.
Authors: Jiancheng Huang, Gengwei Zhang, Zequn Jie, Siyu Jiao, Yinlong Qian, Ling Chen, Yunchao Wei, Lin Ma
Abstract: Text-to-video generation has significantly enriched content creation and holds the potential to evolve into powerful world simulators. However, modeling the vast spatiotemporal space remains computationally demanding, particularly when employing Transformers, which incur quadratic complexity in sequence processing and thus limit practical applications. Recent advancements in linear-time sequence modeling, particularly the Mamba architecture, offer a more efficient alternative. Nevertheless, its plain design limits its direct applicability to multimodal and spatiotemporal video generation tasks. To address these challenges, we introduce M4V, a multimodal Mamba framework for efficient text-to-video generation. Specifically, a MultiModal diffusion Mamba (MM-DiM) block is designed within the framework to enable seamless integration of multimodal information and spatiotemporal modeling. In detail, we introduce a novel multimodal token re-composition design, which employs a bidirectional scheme for multimodal information integration through simple token arrangement, along with visual registers to enhance spatialtemporal consistency. As a result, the MM-DiM blocks in M4V reduce FLOPs by 45% compared with the attention-based alternative when generating videos at 768x1280 resolution. Additionally, several training strategies are explored in this work to provide a better understanding of training text-to-video models using only publicly available datasets. Extensive experiments on text-to-video benchmarks demonstrate M4V's ability to produce high-quality videos while significantly lowering computational costs. Project page: https://huangjch526.github.io/M4V_project/.
Authors: Yu Han, Zhiwei Huang, Yanting Zhang, Fangjun Ding, Shen Cai, Xiaoyu Tang, Yanchao Dong, Rui Fan
Abstract: Point-pixel registration between LiDAR point clouds and camera images is a fundamental yet challenging task in autonomous driving and robotic perception. A key difficulty lies in the modality gap between unstructured point clouds and structured images, especially under sparse single-frame LiDAR settings. Existing methods typically extract features separately from point clouds and images, then rely on hand-crafted or learned matching strategies. This separate encoding fails to bridge the modality gap effectively, and more critically, these methods struggle with the sparsity and noise of single-frame LiDAR, often requiring point cloud accumulation or additional priors to improve reliability. Inspired by recent progress in detector-free matching paradigms, we revisit the projection-based approach and introduce the detector-free framework for direct point-pixel matching between LiDAR and camera views. To further enhance matching reliability, we introduce a repeatability scoring mechanism that acts as a soft visibility prior. This guides the network to suppress unreliable matches in regions with low intensity variation, improving robustness under sparse input. Extensive experiments on KITTI, nuScenes, and MIAS-LCEC-TF70 benchmarks demonstrate that our method achieves state-of-the-art performance, outperforming prior approaches on nuScenes (even those relying on accumulated point clouds), despite using only single-frame LiDAR.
Authors: Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal
Abstract: Inference-time steering methods offer a lightweight alternative to fine-tuning large language models (LLMs) and vision-language models (VLMs) by modifying internal activations at test time without updating model weights. However, most existing approaches rely on fixed, global intervention vectors, overlook the causal influence of individual input tokens, and fail to leverage informative gradients from the model's logits, particularly in multimodal settings where visual and textual inputs contribute unevenly. To address these limitations, we introduce GrAInS, an inference-time steering approach that operates across both language-only and vision-language models and tasks. GrAInS uses contrastive, gradient-based attribution via Integrated Gradients to identify the top-k most influential tokens, both positively and negatively attributed based on their contribution to preferred versus dispreferred outputs. These tokens are then used to construct directional steering vectors that capture semantic shifts from undesirable to desirable behavior. During inference, GrAInS adjusts hidden activations at transformer layers guided by token-level attribution signals, and normalizes activations to preserve representational scale. This enables fine-grained, interpretable, and modular control over model behavior, without retraining or auxiliary supervision. Empirically, GrAInS consistently outperforms both fine-tuning and existing steering baselines: it achieves a 13.22% accuracy gain on TruthfulQA using Llama-3.1-8B, reduces hallucination rates on MMHal-Bench from 0.624 to 0.514 with LLaVA-1.6-7B, and improves alignment win rates on SPA-VL by 8.11%, all while preserving the model's fluency and general capabilities.
Authors: Skatje Myers, Dmitriy Dligach, Timothy A. Miller, Samantha Barr, James Landefeld, Yanjun Gao, Matthew Churpek, Anoop Mayampurath, Majid Afshar
Abstract: Objective: To evaluate whether retrieval-augmented generation (RAG) can serve as an efficient alternative to long-context prompting for clinical reasoning over electronic health records (EHRs). Methods: We defined three EHR-based tasks that are replicable across health systems and vary in reasoning complexity: 1) extracting imaging procedures (modality, date, and anatomic site), 2) generating timelines of therapeutic antibiotic use, and 3) identifying the key diagnoses for a hospitalization. Using real inpatient clinical notes from a US academic health system, we evaluated three large language models (GPT-5.4-mini, Mistral Medium 3, DeepSeek V3.1) with varying amounts of provided context, comparing targeted retrieval to using the most recent clinical notes. Results: For Imaging Procedures, RAG strongly outperformed recent-note inputs and exceeded long-context performance (by 0.17-9.83 F1 across all models) using fewer than 8K tokens. Similar benefits were observed for Antibiotic Timelines, where <8K of retrieved tokens matched long-context recent-notes performance (between -3.26 to +3.24 Jaccard). Error analysis revealed that missing information in the clinical notes--often due to inter-hospital transfers--limited performance to some extent. However, performance on the Diagnosis Generation task remains largely static across methods and models. Discussion: RAG demonstrated strong token efficiency across tasks, with the clearest and most consistent gains observed for imaging extraction and antibiotic timeline reconstruction. Diagnosis generation proved the most challenging task, suggesting ceiling effects imposed by documentation variability and evaluation constraints. Conclusion: Our results suggest that RAG remains a competitive and efficient approach for clinical tasks over large amounts of EHR, even as newer models become capable of handling increasingly longer amounts of text.
Authors: Mengjie Li, Yuan Feng, Xike Xie, William J. Song
Abstract: The growing sequence length of large language models poses significant challenges for key-value (KV) caches. Existing state-of-the-art cache eviction methods primarily analyze the inference behavior of attention heads in successful retrieval-reasoning cases, often overlooking diverse behaviors in failure cases, such as bias and distraction. This oversight limits the potential to leverage heterogeneous head behaviors for improved eviction performance. Inspired by the confusion matrix, we introduce an Attention Behavior Matrix to comprehensively analyze attention head behaviors in both success and failure scenarios. By maximizing the signal-to-noise ratio -- strengthening valid reasoning pathways in success cases while inhibiting noise from bias and distraction in failure cases -- we propose REtrieval-reAsoning and Logic-constructed (REAL) KV cache eviction, the first method to leverage multi-behavior analysis. Comprehensive evaluations show that REAL achieves remarkable performance across various models and benchmarks; notably, on LongBench v2, it achieves comparable accuracy to the strongest baseline, HeadKV-R2, while requiring 32x less space (Figure 1). By offering a novel perspective on behavior analysis, we pave the way for a shift from success-only to comprehensive, failure-aware methods in long-context modeling. Our code is available at https://github.com/yonseicasl/REAL.
Authors: Houcheng Jiang, Junfeng Fang, Jiaxin Wu, Tianyu Zhang, Chen Gao, Xiang Wang, Xiangnan He, Yang Deng
Abstract: Weak-to-strong generalization provides a promising paradigm for scaling large language models (LLMs) by training stronger models on samples from aligned weaker ones, without requiring human feedback or explicit reward modeling. However, its robustness and generalization are hindered by the noise and biases in weak-model outputs, which limit its applicability in practice. To address this challenge, we leverage implicit rewards, which approximate explicit rewards through log-likelihood ratios, and reveal their structural equivalence with Contrastive Decoding (CD), a decoding strategy shown to reduce noise in LLM generation. Building on this connection, we propose Contrastive Weak-to-Strong Generalization (ConG), a framework that employs contrastive decoding between pre- and post-alignment weak models to generate higher-quality samples. This approach enables more reliable capability transfer, denoising, and improved robustness, substantially mitigating the limitations of traditional weak-to-strong methods. Empirical results across different model families confirm consistent improvements, demonstrating the generality and effectiveness of ConG. Taken together, our findings highlight the potential of ConG to advance weak-to-strong generalization and provide a promising pathway toward AGI.
Authors: Peter A. V. DiBerardino, Britt Anderson
Abstract: We formalize human choice behavior in a probabilistic hide-and-seek task. In our geometric construction, vectors represent participant choice frequencies as well as probability matching and maximizing strategies. We measured choice behavior not just in the well-studied scenario of pursuing an objective (seeking), but also the rarely studied scenario of avoiding consequences (hiding). We used our geometric construction to define the avoidance counterpart of probability matching, probability antimatching, as a vector reflection across the uniform distribution. Decomposing the behavior of participants when they were seeking into matching and maximizing components, we could mathematically derive the analogous antimatching and minimizing strategies for hiding. Participants did change their choice frequencies between hiding and seeking conditions. In both cases, we found that a linear combination of just two vectors did an excellent job of fitting participant choice frequencies: matching + maximizing for seeking, antimatching + minimizing for hiding. We could account for diversity in participant strategy usage by varying the coefficients of the two relevant basis strategy vectors. We successfully applied this model in scenarios of up to 7 rooms. We conclude that an apparent diversity of human conduct in stochastic environments can, in some cases, be explained by varying the weighting of two principle strategies: whether to match/antimatch or maximize/minimize.
Authors: Yongji Zhang, Siqi Li, Kuiyang Huang, Yue Gao, Yu Jiang
Abstract: Fine-Grained Visual Classification (FGVC) remains a challenging task due to subtle inter-class differences and large intra-class variations. Existing approaches typically rely on feature-selection mechanisms or region-proposal strategies to localize discriminative regions for semantic analysis. However, these methods often fail to capture discriminative cues comprehensively while introducing substantial category-agnostic redundancy. To address these limitations, we propose H3Former, a novel token-to-region framework that leverages high-order semantic relations to aggregate local fine-grained representations with structured region-level modeling. Specifically, we propose the Semantic-Aware Aggregation Module (SAAM), which exploits multi-scale contextual cues to dynamically construct a weighted hypergraph among tokens. By applying hypergraph convolution, SAAM captures high-order semantic dependencies and progressively aggregates token features into compact region-level representations. Furthermore, we introduce the Hyperbolic Hierarchical Contrastive Loss (HHCL), which enforces hierarchical semantic constraints in a non-Euclidean embedding space. The HHCL enhances inter-class separability and intra-class consistency while preserving the intrinsic hierarchical relationships among fine-grained categories. Comprehensive experiments conducted on four standard FGVC benchmarks validate the superiority of our H3Former framework.
Authors: Georgios Anyfantis, Pere Barlet-Ros
Abstract: Network Intrusion Detection Systems (NIDS) are essential tools for detecting network attacks and intrusions. While extensive research has explored the use of supervised Machine Learning for attack detection and characterisation, these methods require accurately labelled datasets, which are very costly to obtain. Moreover, existing public datasets have limited and/or outdated attacks, and many of them suffer from mislabelled data. To reduce the reliance on labelled data, we propose AutoGraphAD, a novel unsupervised anomaly detection approach based on a Heterogeneous Variational Graph Autoencoder. AutoGraphAD operates on heterogeneous graphs, made from connection and IP nodes that represent network activity. The model is trained using unsupervised and contrastive learning, without relying on any labelled data. The model's losses are then weighted and combined in an anomaly score used for anomaly detection. Overall, AutoGraphAD yields the same, and in some cases better, results than Anomal-E, but without requiring costly downstream anomaly detectors. As a result, AutoGraphAD achieves around 1.18 orders of magnitude faster training and 1.03 orders of magnitude faster inference, which represents a significant advantage for operational deployment.
Authors: Scott Merrill, Shashank Srivastava
Abstract: LLM-based simulations can enable controlled studies of civic deliberation, but current systems lack speaker-attributed data and methods for evaluating long-form institutional behavior. ASR transcripts typically use anonymous labels such as $Speaker\_1$, preventing models from learning stable participant behavior across meetings. We present a reproducible pipeline that converts public Zoom recordings into speaker-attributed transcripts enriched with persona profiles, topics, and pragmatic "action tags" such as $[propose\_motion]$. Using this pipeline, we release three public datasets of government deliberation (Appellate Court hearings, School Board meetings, and Municipal Council sessions) and fine-tune LLM personas on this action-aware data. We evaluate simulations along four dimensions: persona fidelity, persona consistency, institutional fidelity, and behavioral coherence. Action-aware fine-tuning cuts perplexity by 67%, doubles classifier-based persona fidelity, increases vote attempts by up to $3.6\times$, and improves deliberative responsiveness by up to 70%. Human evaluations show that simulated excerpts are often hard to distinguish from real deliberations, indicating a practical foundation for data-grounded civic simulation studies.
Authors: Changpeng He, Yang Lu, Yanqing Xu, Chong-Yung Chi, Arumugam Nallanathan
Abstract: This paper investigates a reconfigurable intelligent surface (RIS)-assisted multi-waveguide pinching-antenna (PA) system (PASS) for multi-user downlink information transmission, motivated by the unknown impact of the integration of emerging PASS and RIS on wireless communications. First, we formulate sum rate (SR) and energy efficiency (EE) maximization problems in a unified framework, subject to constraints on the movable region of PAs, total power budget, and tunable phase of RIS elements. Then, by leveraging a graph-structured topology of the RIS-assisted PASS, a novel three-stage graph neural network (GNN) is proposed, which learns PA positions based on user locations, and RIS phase shifts according to composite channel conditions at the first two stages, respectively, and finally determines beamforming vectors. Specifically, the proposed GNN is achieved through unsupervised training, together with three implementation strategies for its integration with convex optimization, thus offering trade-offs between inference time and solution optimality. Extensive numerical results are provided to validate the effectiveness of the proposed GNN, and to support its unique attributes of viable generalization capability, good performance reliability, and real-time applicability. Moreover, the impact of key parameters on RIS-assisted PASS is illustrated and analyzed.
Authors: Dongheng Qian, Jing Wang
Abstract: Measurement-induced entanglement (MIE) captures how local measurements generate long-range quantum correlations and drive dynamical phase transitions in many-body systems. Yet estimating MIE experimentally remains challenging: direct evaluation requires extensive post-selection over measurement outcomes, raising the question of whether MIE is accessible with only polynomial resources. We address this challenge by reframing MIE detection as a data-driven learning problem that assumes no prior knowledge of state preparation. Using measurement records alone, we train a neural network in a self-supervised manner to predict the uncertainty metric for MIE--the gap between upper and lower bounds of the average post-measurement bipartite entanglement. Applied to random circuits with one-dimensional all-to-all connectivity, our method reveals a learnability transition with increasing circuit depth: below a threshold the MIE can be effectively learned with resources that grow only polynomially with system size, whereas above it the required resources grow exponentially. This computational phase transition coincides with the breakdown of efficient classical simulation of the underlying quantum state. We further observe signatures of this transition on current noisy quantum devices. These results highlight the power of data-driven approaches for learning MIE and delineate the practical limits of its classical learnability.
Authors: Natalia Ponomareva, Zheng Xu, H. Brendan McMahan, Peter Kairouz, Lucas Rosenblatt, Vincent Cohen-Addad, Crist\'obal Guzm\'an, Ryan McKenna, Galen Andrew, Alex Bie, Da Yu, Alex Kurakin, Morteza Zadimoghaddam, Sergei Vassilvitskii, Andreas Terzis
Abstract: High quality data is needed to unlock the full potential of AI for end users. However finding new sources of such data is getting harder: most publicly-available human generated data will soon have been used. Additionally, publicly available data often is not representative of users of a particular system -- for example, a research speech dataset of contractors interacting with an AI assistant will likely be more homogeneous, well articulated and self-censored than real world commands that end users will issue. Therefore unlocking high-quality data grounded in real user interactions is of vital interest. However, the direct use of user data comes with significant privacy risks. Differential Privacy (DP) is a well established framework for reasoning about and limiting information leakage, and is a gold standard for protecting user privacy. The focus of this work, \emph{Differentially Private Synthetic data}, refers to synthetic data that preserves the overall trends of source data,, while providing strong privacy guarantees to individuals that contributed to the source dataset. DP synthetic data can unlock the value of datasets that have previously been inaccessible due to privacy concerns and can replace the use of sensitive datasets that previously have only had rudimentary protections like ad-hoc rule-based anonymization. In this paper we explore the full suite of techniques surrounding DP synthetic data, the types of privacy protections they offer and the state-of-the-art for various modalities (image, tabular, text and decentralized). We outline all the components needed in a system that generates DP synthetic data, from sensitive data handling and preparation, to tracking the use and empirical privacy testing. We hope that work will result in increased adoption of DP synthetic data, spur additional research and increase trust in DP synthetic data approaches.
Authors: Zihan Zhou, Animesh Garg, Ajay Mandlekar, Caelan Garrett
Abstract: Long-horizon manipulation has been a long-standing challenge in the robotics community. We propose ReinforceGen, a system that combines task decomposition, data generation, imitation learning, and motion planning to form an initial solution, and improves each component through reinforcement-learning-based fine-tuning. ReinforceGen first segments the task into multiple localized skills, which are connected through motion planning. The skills and motion planning targets are trained with imitation learning on a dataset generated from 10 human demonstrations, and then fine-tuned through online adaptation and reinforcement learning. When benchmarked on the Robosuite dataset, ReinforceGen reaches 80% success rate on all tasks with visuomotor controls in the highest reset range setting. Additional ablation studies show that our fine-tuning approaches contribute to an 89% average performance increase. Finally, ReinforceGen demonstrates significant improvement through fine-tuning in our real-world evaluations. More results and videos are available at https://reinforcegen.github.io.
Authors: Weili Nie, Julius Berner, Nanye Ma, Chao Liu, Saining Xie, Arash Vahdat
Abstract: Large video diffusion and flow models have achieved remarkable success in high-quality video generation, but their use in real-time interactive applications remains limited due to their inefficient multi-step sampling process. In this work, we present Transition Matching Distillation (TMD), a novel framework for distilling video diffusion models into efficient few-step generators. The central idea of TMD is to match the multi-step denoising trajectory of a diffusion model with a few-step probability transition process, where each transition is modeled as a lightweight conditional flow. To enable efficient distillation, we decompose the original diffusion backbone into two components: (1) a main backbone, comprising the majority of early layers, that extracts semantic representations at each outer transition step; and (2) a flow head, consisting of the last few layers, that leverages these representations to perform multiple inner flow updates. Given a pretrained video flow model, we first introduce a flow head to the model, and adapt it into a conditional flow map. We then apply distribution matching distillation to the student model with flow head rollout in each transition step. Extensive experiments on distilling Wan2.1 1.3B and 14B text-to-video models demonstrate that TMD provides a flexible and strong trade-off between generation speed and visual quality. In particular, TMD outperforms existing distilled models under comparable inference costs in terms of visual fidelity and prompt adherence. Project page: https://research.nvidia.com/labs/genair/tmd
Authors: R\'ois\'in Luo
Abstract: Deep learning has achieved remarkable success across a wide range of domains, significantly expanding the frontiers of what is achievable in artificial intelligence. Yet, despite these advances, critical challenges remain -- most notably, ensuring robustness to small input perturbations and generalization to out-of-distribution data. These critical challenges underscore the need to understand the underlying fundamental principles that govern robustness and generalization. Among the theoretical tools available, Lipschitz continuity plays a pivotal role in governing the fundamental properties of neural networks related to robustness and generalization. It quantifies the worst-case sensitivity of network's outputs to small input perturbations. While its importance is widely acknowledged, prior research has predominantly focused on empirical regularization approaches based on Lipschitz constraints, leaving the underlying principles less explored. This thesis seeks to advance a principled understanding of the principles of Lipschitz continuity in neural networks within the paradigm of machine learning, examined from two complementary perspectives: an internal perspective -- focusing on the temporal evolution of Lipschitz continuity in neural networks during training (i.e., training dynamics); and an external perspective -- investigating how Lipschitz continuity modulates the behavior of neural networks with respect to features in the input data, particularly its role in governing frequency signal propagation (i.e., modulation of frequency signal propagation).
Authors: Tanmay Ambadkar, Sourav Panda, Shreyash Kale, Jonathan Dodge, Abhinav Verma
Abstract: Multi-objective reinforcement learning (MORL) seeks to train agents capable of balancing conflicting objectives. While single preference-conditioned policies offer a highly scalable solution, existing approaches remain brittle in practice, frequently failing to recover dense Pareto fronts. We demonstrate that this failure stems from two structural pathologies: destructive advantage cancellation caused by premature Early Scalarization (ES), and representational mode collapse across the preference space. To overcome these bottlenecks, we introduce $D^3PO$, a PPO-based framework that fundamentally reorganizes multi-objective optimization. By preserving per-objective learning signals through a decomposed pipeline and integrating preferences only after trust-region stabilization (Late-Stage Weighting), $D^3PO$ improves credit assignment under conflicting objectives. Concurrently, a scaled diversity regularizer encourages behavioral divergence proportional to preference distance. $D^3PO$ operates entirely within the efficient linear scalarization regime shared by standard deep MORL baselines. By reducing information loss caused due to linear scalarization rather than relying on expensive non-linear utility functions, it suggests that optimization bottlenecks play a significant role. Across available standard benchmarks, including high-dimensional and many-objective environments, $D^3PO$ consistently discovers broader, higher-quality Pareto fronts than prior methods, exceeding state-of-the-art hypervolume and expected utility using a single deployable policy.
Authors: Stephan Vonschallen, Dominique Oberle, Theresa Schmiedel, Friederike Eyssel
Abstract: Generative social robots (GSRs) powered by large language models enable adaptive, conversational tutoring but also introduce risks such as misinformation, overreliance, and privacy violations. Existing frameworks for educational technologies and responsible AI primarily define desired behaviors, yet they rarely specify the knowledge prerequisites that enable generative agents to express these behaviors reliably. To address this gap, we adopt a knowledge-based design perspective and investigate what information tutoring-oriented GSRs require to function responsibly and effectively in higher education. Based on twelve semistructured interviews with university students and lecturers, we identified twelve design requirements across three knowledge types: self-knowledge (assertive, conscientious, and friendly personality with customizable role), user-knowledge (personalized information about student learning goals, learning progress, motivation type, emotional state, and background), and context-knowledge (learning materials, educational strategies, courserelated information, and physical learning environment). Drawing from these results, this work provides a structured foundation for the design of tutoring GSRs, aligning generative AI capabilities with pedagogical and ethical expectations.
Authors: Zirong Chen, Meiyi Ma
Abstract: Emergency call-takers form the first operational link in public safety response, handling over 240 million calls annually while facing a sustained training crisis: staffing shortages exceed 25\% in many centers, and preparing a single new hire can require up to 720 hours of one-on-one instruction that removes experienced personnel from active duty. Traditional training approaches struggle to scale under these constraints, limiting both coverage and feedback timeliness. In partnership with Metro Nashville Department of Emergency Communications (MNDEC), we designed, developed, and deployed a GenAI-powered call-taking training system under real-world constraints. Over six months, deployment scaled from initial pilot to 190 operational users across 1,120 training sessions, exposing systematic challenges around system delivery, rigor, resilience, and human factors that remain largely invisible in controlled or purely simulated evaluations. By analyzing deployment logs capturing 98,429 user interactions, organizational processes, and stakeholder engagement patterns, we distill four key lessons, each coupled with concrete design and governance practices. These lessons provide grounded guidance for researchers and practitioners seeking to deliver AI-driven training systems in safety-critical public sector environments where practical constraints fundamentally shape human-centric design.
Authors: Zac Garby, Andrew D. Gordon, David Sands
Abstract: Large language models are increasingly deployed as agents: they plan, call tools, read untrusted data, and act on the results. This exposes them to prompt injection: data meant only to be read is obeyed as an instruction. The most principled defences replace content inspection with provenance: classifying data by source and keeping trusted and untrusted apart through a separation of duty (the dual-LLM pattern) and information-flow control. Yet the leading systems are hard to fully trust: flow tracking is easy to get wrong, deliberate relaxations are hard to audit, and the dual-LLM pattern is hard-wired into the architecture. We present LLMbda, an untyped call-by-value lambda calculus that makes provenance-based defence both expressible and provably sound, without committing to an architecture. It adds the operational core of agentic systems as first-class constructs: prompt-response conversations that can be forked and cleared, code generation, and dynamic information-flow control in which every value carries a label that every reduction propagates. Isolation becomes a policy a program expresses, and reclassification an explicit, auditable construct. Our central result is a termination-insensitive probabilistic noninterference theorem over the whole calculus, including code-generating agents, with an insulated variant that holds even when the attacker chooses all untrusted inputs. The verified interpreter is itself the harness that calls the model, to our knowledge the first LLM agent harness whose executable is the subject of machine-checked security theorems, so every agent inherits the guarantee. On the AgentDojo banking benchmark, an agent built within LLMbda, enforcement always on, matches the utility of CaMeL, a leading dual-LLM defence, run without its policy checks (which halve its utility), and resists all but two of 1296 attacked runs. Our harness and all proofs are in Lean.
Authors: Gangda Deng, Zhaoling Chen, Zhongming Yu, Haoyang Fan, Yuhong Liu, Yuxin Yang, Dhruv Parikh, Rajgopal Kannan, Le Cong, Mengdi Wang, Qian Zhang, Viktor Prasanna, Xiangru Tang, Xingyao Wang
Abstract: Real-world software must continuously evolve to meet ever-changing and open-ended requirements. AI agents, increasingly deployed as long-running systems, are now entrusted to drive this evolution. Yet, existing benchmarks evaluate agents on isolated, one-off coding tasks, neglecting the temporal dependencies and technical debt inherent in real-world software evolution. To bridge this gap, we introduce DeepCommit, an agentic pipeline that reconstructs verifiable Milestone DAGs from noisy commit logs, where milestones are defined as functionally cohesive development goals. These executable sequences enable SWE-Milestone, a benchmark that evaluates agents on streams of milestone-level tasks, requiring them to sustain system integrity and limit error accumulation, dimensions of long-term software evolution largely missing from current benchmarks. Our evaluation of 12 frontier models across 4 agent frameworks reveals a critical vulnerability: overall performance scores drop significantly from >80% on isolated tasks to at most 38% in continuous settings, exposing agents' profound struggle with long-term maintenance and error propagation.
Authors: Iakovos-Christos Zarkadis, Christos Douligeris
Abstract: Supervised detection of network attacks has always been a critical part of network intrusion detection systems (NIDS). Nowadays, in a pivotal time for artificial intelligence (AI), with even more sophisticated attacks that utilize advanced techniques, such as generative artificial intelligence (GenAI) and reinforcement learning, it has become a vital component if we wish to protect our personal data, which are scattered across the web. In this paper, we address two tasks, in the first unified multi-modal NIDS dataset, which incorporates flow-level data, packet payload information and temporal contextual features, from the reprocessed CIC-IDS-2017, CIC-IoT-2023, UNSW-NB15 and CIC-DDoS-2019, with the same feature space. In the first task we use machine learning (ML) algorithms, with stratified cross validation, in order to prevent network attacks, with stability and reliability. In the second task we use adversarial learning algorithms to generate synthetic data, compare them with the real ones and evaluate their fidelity, utility and privacy using the SDV framework, f-divergences, distinguishability and non-parametric statistical tests. The findings provide stable ML models for intrusion detection and generative models with high fidelity and utility, by combining the Synthetic Data Vault framework, the TRTS and TSTR tests, with non-parametric statistical tests and f-divergence measures.
Authors: Harshit Joshi, Priyank Shethia, Jadelynn Dao, Monica S. Lam
Abstract: Systematic reviews -- which requires comprehensive evidence collection and synthesis from large document corpora in response to targeted research questions -- are foundational in finance, social sciences, and other technical fields. Manual construction of evidence tables is labor-intensive, and recent LLM-based assistants relying on embedding or keyword based search often fail to meet the coverage standards of systematic reviews. We introduce SLIDERS, a novel LLM-based methodology for systematic reviews, by automatically assembling evidence tables tailored to research questions. In addition to extracting structured data from documents, SLIDERS can extract full-text excerpts that serve as direct evidence or as provenance for structured data. Core to SLIDERS is an automated evidence reconciliation agent that writes code to analyze and reconcile extracted evidence, bringing together information fragmented across documents, resolving inconsistencies across excerpts, and synthesizing overlapping findings into a coherent evidence table. In addition, SLIDERS allows users to ask follow-up questions in natural language to further explore the assembled evidence. We evaluate SLIDERS on three systematic-review-style tasks over large document collections. SLIDERS outperforms the best-performing baseline across benchmarks, remains near 90% accuracy across 6M-11M-token corpora. On two new follow-up analysis benchmarks SLIDERS can answer 77.9% and 58.3% followup questions accurately
Authors: Filip Edstr\"om, Guilherme W. F. Barros, Tetiana Gorbach, Xavier de Luna
Abstract: Artificial-intelligence systems are becoming ubiquitous in society, yet their predictions typically inherit biases with respect to protected attributes such as race, gender, or age. Classical fairness notions, most notably Statistical Parity (SP), demand that predictions be independent of the protected attributes, but are overly restrictive when these attributes influence mediating variables that are considered business necessities. Recent causal formulations relax SP by distinguishing allowed from not-allowed causal paths and by complementing SP with Predictive Parity (PP), requiring the predictor to replicate the legitimate influence of business-necessities. Existing path-based definitions are mainly practical when applied to categorical attributes. This paper introduces a new framework for fairness in structural causal models that is tailored to continuous protected attributes. We formalize SP and PP through path-specific partial derivatives, establish conditions under which these criteria coincide with prior causal definitions, and characterize when a fair predictor, one that satisfies SP along not-allowed paths while achieving PP along allowed paths, exists. Building on this theory, we propose a fair tuning algorithm that either constructs such a predictor or, when not possible, allows for a trade-off between SP and PP. We present experiments on simulated and real data to evaluate our proposal, compare it with previously proposed methods, and show that it performs better when PP is considered.
Authors: Vardhan Dongre, Dilek Hakkani-T\"ur
Abstract: Effective collaboration between embodied agents requires more than acting in a shared environment; it demands communication grounded in each agent's evolving understanding of the world. When agents can only partially observe their surroundings, coordination without communication is provably hard, but communication can, in principle, bridge this gap by allowing agents to share observations and align their world models. In this work, we examine whether LLM-based embodied agents actually realize the ability to communicate. We extend PARTNR, a benchmark for collaborative household robotics, with a natural-language dialogue channel that enables two agents with partial observability to communicate during task execution. To evaluate whether dialogue leads to genuine world-model alignment rather than superficial coordination, we propose a framework for measuring world-model alignment defined over per-agent world graphs: observation convergence (do private world models align over time?), information novelty (do messages convey what the partner lacks?), and belief-sensitive messaging (do agents model what their partner knows?). Our experiments across three LLMs reveal that dialogue reduces action conflicts 40 to 83 percentage points but degrades task success relative to silent coordination. Using our metrics, we characterize the gap between superficial coordination and genuine world-model alignment, and identify where current models fall on this spectrum. Project Website: https://uiuc-conversational-ai-lab.github.io/partnr-dial-wmd/
URLs: https://uiuc-conversational-ai-lab.github.io/partnr-dial-wmd/
Authors: Tao Xie, Zexi Tan, Haoyi Xiao, Mengke Li, Yiqun Zhang, Yang Lu, Cuie Yang, Yiu-ming Cheung
Abstract: Multivariate time series classification (MTSC) is pivotal in high-stakes domains, such as clinical diagnosis and industrial fault detection, where safe deployment necessitates transparent decision-making. However, isolating the temporal segments that drive model predictions is challenging because discriminative signals in real-world time series are typically sparse, heterogeneous, and heavily obscured by background noise. This paper, therefore, proposes AnchorMoE, an interpretable-by-construction classification framework. Built upon a Mixture-of-Experts (MoE) architecture, AnchorMoE encodes multi-view representations of local patches and routes them to specialized experts, ensuring that the final prediction is formulated as an exact additive decomposition over the input segments, facilitating ante-hoc transparency rather than relying on post-hoc estimations. To maintain the reliability of this decomposition under sparse signal distributions, we introduce a geometric orthogonality constraint that penalizes representational redundancy, compelling distinct experts to specialize in heterogeneous predictive patterns. Furthermore, an uncertainty-aware reliability gate is designed to dynamically calibrate the contribution of each segment, effectively suppressing residual background noise. Extensive experiments on real-world and synthetic benchmarks demonstrate that AnchorMoE achieves highly competitive classification performance while faithfully grounding its decisions in the raw time series.
Authors: Ali Behrouz, Farnoosh Hashemi, Adel Javanmard, Vahab Mirrokni
Abstract: The past few decades have witnessed significant advances in the design of machine learning algorithms, from early studies on task-specific shallow models to more general deep Large Language Models (LLMs). Despite showing promising results in tasks that require instant prediction or in-context learning, existing models lack the ability to continually learn and effectively transfer their temporal in-context knowledge to their long-term parameters. Inspired by human learning process, we introduce a ''Sleep'' paradigm that allows the models to continually learn, distill their short-term fragile memories into stable long-term knowledge with replay, and recursively improve themselves with ''Dreaming'' process. In more detail, sleep consists of two stages: (1) Memory Consolidation: an upward distillation process, called Knowledge Seeding, where the memories of a smaller-self are distilled into a larger network to provide more capacity while preserving the knowledge. As a proof of concept, we present a new Generalized Distillation process for {Knowledge Seeding} (i.e., the combination of on-policy distillation with Reinforcement Learning (RL)-based imitation learning); (2) Dreaming: a self-improvement phase, where the model uses RL to generate a curriculum of synthetic data to rehearse new knowledge and refine existing capabilities without human supervision. Our experiments on long-horizon, continual learning, knowledge incorporation, and few-shot generalization tasks support the importance of the sleep stage.
Authors: Tianming Du, Peijie Yu, Sihan Shang, Danli Shi, My Linh Nguyen, Shengbo Gao, Guangyuan Li, Yinghong Yu, Yan Jiang, Qianlong Zhao, Behzad Bozorgtabar, Shaoxiong Ji, Jiazhen Pan, Daniel Rueckert, Jiancheng Yang
Abstract: The most plausible near-term role of medical LLMs is to assist rather than replace physicians, yet current evaluations often test isolated capabilities: clinical knowledge, EHR system interaction, or patient communication. Physician assistance instead requires coordinating these capabilities within the same interaction, where physicians issue underspecified requests, patients describe symptoms ambiguously, and EHR systems demand precise tool use. We introduce PhysAssistBench, a benchmark for interactive doctor-patient-EHR assistance. Built from real MIMIC-IV cases, PhysAssistBench uses a scalable pipeline to construct agentic patients: interactive, record-grounded agents that turn static EHR records into multi-turn clinical scenarios while preserving clinical factuality. PhysAssistBench provides a curated bilingual evaluation set of 1,296 manually reviewed and physician-validated turns. Experiments with leading LLMs show that current models remain unreliable in this setting, which exposes a key bottleneck for clinical LLMs: reliable assistance requires coordination across knowledge, communication, and systems, not isolated gains in any of them.
Authors: Zijun Xie, Binbin Zheng, Enlei Gong, Jihua Liu, Yuyang You, Lingfeng Liu, Jiayao Tang, Guanqun Zhao, Aoqi Hu, Zeyu Chen
Abstract: Long-horizon language agents must repeatedly interact with tools, accumulate evidence, and make decisions under bounded context windows. Context-management methods make such rollouts feasible by simplifying past interactions through deletion, folding, or memory editing. However, when useful history is collapsed into compressed states, the reconstructed context may no longer reveal which earlier observations support a successful final answer. This creates a mismatch between bounded-context acting and outcome-based reinforcement learning: the policy acts on reconstructed context, while the learner lacks source-level provenance for assigning credit to the evidence that mattered. We propose ECHO, a selective turn-memory framework for traceable context reconstruction in Agentic RL. ECHO compresses each completed environment turn into a compact source-indexed memory record, reconstructs bounded policy contexts by selecting useful records, and reuses the selected source indices to route positive outcome credit to the final trajectory segment, reused evidence turns, memory findings, and memory-selection actions. On BrowseComp-Plus, ECHO reaches 43.4% held-out accuracy, outperforming GRPO at 28.9% and the rolling-summary baseline SUPO at 36.1%, while using fewer turns and lower trajectory volume than SUPO. The trained policy also improves zero-shot generalization across multi-objective QA, code generation, and deep information-seeking benchmarks on both dense and MoE backbones.
Authors: Zongwu Xie, Yonglong Zhang, Yifan Yang, Yang Liu, Guanghu Xie
Abstract: Monocular spacecraft 6D pose estimation remains difficult under weak texture, thin structures, illumination variation, and occlusion. This article presents GAP-GDRNet, a geometry-aware RGB framework built on GDR-Net for a single-target synthetic spacecraft benchmark. The method strengthens the geometry-guided regression pipeline at two points. First, AFR is placed before dense geometric prediction to combine global structural attention with local weak-texture enhancement. Second, PGSA is inserted into Patch-PnP to relate downsampled geometric regions before final pose regression. Dense supervision is obtained from a Blender-based rendering and annotation process that provides masks, model-coordinate maps, camera intrinsics, and 6D pose labels. On the self-built spacecraft dataset, GAP-GDRNet achieves a rotation error of 1.96{\deg}, a translation error of 0.0165 m,and 95.16% ADD@0.02 m, outperforming the reproduced GDR-Net baseline by 3.88 percentage points while running at 35.97 FPS. Tests on T-LESS and LM-O further show consistent gains over the reproduced baseline on textureless and occluded non-spacecraft objects.
Authors: Diego Cerda-Mardini, Sarath Chandar, Sreenath Madathil
Abstract: LLMs are increasingly deployed as post-hoc explainers of AI-generated outputs, yet it remains unclear whether they can reliably communicate probabilistic information in natural language. For this role to be viable, models must produce identical verbal descriptions for identical inputs, and select descriptions that accurately reflect the magnitude of the underlying numerical quantities. We evaluate whether nine LLMs meet these requirements within a two-stage prediction pipeline, in which an upstream model has produced probabilistic outputs characterized by their likelihood and uncertainty, and LLMs are tasked with selecting an appropriate verbal descriptor for each. We simulate predictions from an upstream model by taking samples from a Beta distribution parameterized by its mode and prior sample size. We then prompt LLMs to explain these predictions under six domain contexts and with ten temperature settings, and repeating each experiment ten times. We find that LLMs are generally consistent but miscalibrated, with substantially weaker performance on uncertainty than on likelihood tasks. Providing models with precomputed summary statistics (mode and prior sample size) reduced sensitivity to contextual framing but did not resolve the underlying miscalibration, suggesting that the bottleneck resides in the verbalization step itself. These findings indicate that current LLMs do not yet constitute reliable zero-shot standalone risk communication tools for probabilistic predictions.
Authors: Abhishek Kumar, Carsten Maple
Abstract: Large language models are increasingly deployed as IDE-integrated coding agents that decompose tasks, generate and edit files, run code, and refine outputs over many turns. Yet their safety is still often evaluated as if they were chatbots: one harmful prompt, one response, judged in isolation. We introduce workflow-level jailbreak construction, a failure mode in which a harmful objective is assembled across ordinary stages of a software-development workflow rather than generated through a single direct prompt. Using GitHub Copilot in Visual Studio Code, we study four closed-weight backends: Claude Sonnet 4.6, Claude Haiku 4.5, Gemini 3.1 Pro, and Gemini 3.5 Flash. Across 204 prompts from Hammurabi's Code, HarmBench, and AdvBench , the models show near-complete refusal under direct chat, CSV-read, and single-step code-fix baselines, with only 8/816 successful responses in each baseline condition. Under the full workflow, however, the same prompts and backends produce 816/816 unsafe teaching-shot completions, all independently confirmed by two expert evaluators under a strict rubric. These results show that conversational refusal benchmarks can substantially overstate the safety of deployed coding agents and motivate defenses that reason about safety across multi-turn IDE workflows and their generated artifacts, not only individual chat turns.
Authors: Haz Sameen Shahgir, Yufei Li, Xiaohan Wei, Yunchen Pu, Fei Tian, Chonglin Sun, Frank Shyu, Sandeep Pandey, Luke Simon, Yue Dong, Xi Liu
Abstract: Chain-of-thought (CoT) distillation in the recommendation domain is a necessary precursor to RL training, but raw teacher traces are ill-suited to this task. Large teachers approach the recommendation task with unusually high reasoning uncertainty, repeatedly rechecking their answers without revising them; supervised fine-tuning on such traces produces verbose students that never revise their initial guess. Furthermore, due to the novelty of the recommendation domain, the teacher's reasoning traces are highly out-of-distribution for the small student LLM. We propose Student-Aware CoT Optimization for Recommendation Distillation (SCOReD), a CoT optimization framework tailored to recommendation that first parses each teacher trace into typed segments and uses the student LLM's attention to score the importance of each segment. Then SCOReD dynamically selects a per-segment edit (KEEP / REWRITE / FUSE / PRUNE) based on the output length and comparative log probability lift of the answer given the edit as per the student. Therefore, SCOReD prunes redundant sections of the reasoning trace while preserving information-dense sections and adapts raw teacher traces to the student's output distribution. Training on SCOReD-optimized CoTs provides a cleaner learning signal to the student model and improves over baseline SFT by 1.56% NDCG and 1.9% Recall@5, while reducing reasoning length by 27.3%.
Authors: Andrea Bacciu, Andrea Alfarano, Saab Mansour, Amin Mantrach, Marcello Federico
Abstract: Uncertainty estimation (UE) enables LLM-powered systems to recognize when to abstain, yet existing research has predominantly focused on English. We present the first large-scale evaluation of UE methods across 22 languages, spanning high-, mid-, and low-resource settings. Using two human-curated Q\&A datasets, we compare open and closed box UE methods (nine in total) across different model sizes and architectures while eliciting long-form reasoning, avoiding LLM-as-a-judge and embedding-based scoring, which can introduce evaluation noise. We report three main actionable findings. First, we find that prompting models to reason in English while keeping questions in low-resource languages substantially improves UE performance, suggesting that comprehension of low-resource languages is largely intact, and that the reliability bottleneck lies in generation rather than understanding. Second, prompting models to reason in English closes the UE performance gap between low and high-resource languages, demonstrating that generation language matters more than the question language. Third, the choice of UE method should depend on model scale: at smaller scales, open-box probability-based methods outperform alternatives; at larger scales, closed-box self-verbalized uncertainty becomes superior. Finally, we provide an analysis of threshold selection for selective prediction, offering guidance on calibrating abstention in multilingual settings.
Authors: Yusen Feng, Bingchen Han, Jiangran Lyu, Kai Liu, Yixin Zheng, Yuxuan Wan, Weiheng Liu, Sun Han, Ruiqin Li, Yulong Zhang, Fangfu Liu, Xuesong Shi, Libin Liu, Yizhou Wang, Zhizheng Zhang, He Wang
Abstract: Steering robot foundation models (RFMs) toward new task variants or user-preferred behaviors remains challenging, often requiring additional robot demonstrations, task-specific fine-tuning, or long-context conditioning. We present WAM-TTT, a test-time training framework for steering world action models from raw human videos. Rather than treating human videos as trajectories to imitate, WAM-TTT absorbs them into a lightweight adaptive memory inside a frozen WAM through self-supervised video prediction. To make this memory useful for control, we introduce a meta-training stage that aligns human demonstrations with robot behaviors using paired human-robot data and a key--value memory reconstruction objective. At test time, only unlabeled human videos are required to adapt the memory, while the pretrained WAM remains frozen. This enables efficient and reusable steering without robot actions, human-side annotations, or task-specific fine-tuning, while preserving the generalization ability of the foundation model. Extensive experiments show that WAM-TTT consistently outperforms in-context human-video conditioning baselines across diverse manipulation tasks and generalization settings.
Authors: Szczepan Konior, Alexandre Quemy, Przemys{\l}aw Klocek, Bart{\l}omiej Sobieski, Gr\'egoire Cattan
Abstract: Understanding the geometric structure of pre-trained language model embeddings matters for interpretability and safety. We ask whether sentence-level classification signal lives in the Riemannian geometry of contextual token embeddings, and probe it by extracting per-token pullback metrics from a learned encoder's analytical Jacobian and aggregating them with the Fr\'echet mean on the symmetric positive definite (SPD) manifold; we call this procedure Riemannian Mean Pooling (RMP). Across three datasets with non-trivial linguistic structure (CoLA, CREAK, RTE), RMP outperforms Euclidean mean pooling, while on FEVER-Symmetric, a benchmark constructed to remove annotation-driven lexical artifacts, the method correctly stays at chance. Ablations show that a randomly initialised encoder combined with Fr\'echet aggregation already beats Euclidean pooling on two of the three signal-bearing datasets, localising the source of the gain to the geometric aggregation rather than to learned manifold structure; the trained encoder contributes additional signal specifically on CREAK, the most knowledge-heavy of the three signal-bearing datasets.
Authors: Zhoujie Hou, Song Wang, Kexin Lou, Mo Wang, Chen Wei, Quanying Liu
Abstract: Sleep physiology arises from the coordinated dynamics of the central nervous system (CNS) and autonomic nervous system (ANS), as reflected by multimodal polysomnography signals including EEG, EOG, EMG, ECG, and respiration. However, existing sleep foundation models often fuse heterogeneous biosignals in a topology-agnostic manner, overlooking their physiological organization. We introduce Omni-Sleep, a sleep foundation model that uses the CNS/ANS partition as a physiological prior for topology-constrained representation learning. Omni-Sleep learns structured representations through three objectives: intra-system consistency, which captures shared subsystem-level factors within neural and cardio-respiratory signals; inter-system synchronization, which aligns subsystem trajectories to model brain--body dynamics; and latent-space masked temporal modeling, which captures long-horizon sleep dynamics. Pre-trained on over 100,000 hours of multi-center multimodal PSG data, Omni-Sleep is evaluated on sleep staging and multi-disease classification. Across datasets and modality-ablation settings, Omni-Sleep outperforms strong foundation-model baselines, showing improved label efficiency, cross-dataset generalization, and robustness to missing modalities. These results highlight the value of physiological hierarchy for generalizable sleep representation learning. Code is available at https://github.com/AutoBrain-sleep/OmniSleep.
Authors: Haozhan Tang, Zerui Wang, Yuxian Gu, Song Han, Han Cai
Abstract: Modern LLMs are increasingly deployed in long-context applications such as retrieval-augmented generation, repository-level coding, and agentic workflows whose accumulated reasoning and tool traces routinely push the input an order of magnitude past the pretraining window, making zero-shot context extension the dominant deployment path for open-weight checkpoints. The dominant zero-shot methods (YaRN, Self-Extend, DCA) fix a single rescaling factor up front, so an aggressive factor sacrifices short-context fidelity while a conservative one breaks down at long contexts; recent length-aware variants adapt the mapping, but with a fitted or distance-dependent schedule. We propose Jet-Long, a tuning-free zero-shot method that pairs a local RoPE-faithful window with a long-range window whose rescaling factor adapts dynamically to the current sequence length via a parameter-free analytic schedule, recovering the base model exactly at short inputs while extrapolating cleanly at long ones. An inclusion-exclusion attention merge and an on-the-fly RoPE correction rotation make the bifocal construction essentially free at inference; fused into a single CuTe kernel, long-context prefill reaches up to $1.39\times$ FA2 throughput on H100 (approaching the Hopper-only FA4), and single-batch generation incurs $\le 4\%$ overhead at every length. On Qwen3-1.7B/4B/8B up to 128K context, Jet-Long leads RULER by $+4.79$/$+2.18$/$+2.03$ pp over the strongest baseline at 1.7B/4B/8B, achieves the best overall accuracy on HELMET-RAG (a benchmark identified by HELMET as the most efficient predictor of downstream long-context performance) and attains the lowest PG-19 perplexity. Jet-Long also generalizes to hybrid attention architectures such as Jet-Nemotron for further long-context improvement without retraining, and remains hyperparameter-resilient for ease of deployment.