Authors: Kirisci Murat
Abstract: Multi-criteria decision-making methods provide decision-makers with appropriate tools to make better decisions in uncertain, complex, and conflicting situations. Fuzzy set theory primarily deals with the uncertainty inherent in human thoughts and perceptions and attempts to quantify this uncertainty. Fuzzy logic and fuzzy set theory are utilized with multi-criteria decision-making methods because they effectively handle uncertainty and fuzziness in decision-makers' judgments, allowing for verbal judgments of the problem. This study utilizes the Fermatean fuzzy environment, a generalization of fuzzy sets. An optimization model based on the deviation maximization method is proposed to determine partially known feature weights. This method is combined with interval-valued Fermatean fuzzy sets. The proposed method was applied to the problem of selecting renewable energy sources. The reason for choosing renewable energy sources is that meeting energy needs from renewable sources, balancing carbon emissions, and mitigating the effects of global climate change are among the most critical issues of the recent period. Even though selecting renewable energy sources is a technical issue, the managerial and political implications of this issue are also important, and are discussed in this study.
Authors: Andrew Kiruluta, Priscilla Burity
Abstract: We introduce Spectral NSR, a fully spectral neuro-symbolic reasoning framework that embeds logical rules as spectral templates and performs inference directly in the graph spectral domain. By leveraging graph signal processing (GSP) and frequency-selective filters grounded in the Laplacian eigenstructure of knowledge graphs, the architecture unifies the interpretability of symbolic reasoning with the scalability and adaptability of spectral learning. Beyond the core formulation, we incorporate a comprehensive set of extensions, including dynamic graph and basis learning, rational and diffusion filters for sharper spectral selectivity, mixture-of-spectral-experts for modular specialization, proof-guided training with spectral curricula, and uncertainty quantification for calibrated confidence. Additional enhancements such as large language model coupling, co-spectral transfer alignment, adversarial robustness, efficient GPU kernels, generalized Laplacians, and causal interventions further expand the versatility of the framework. Empirical evaluation on state-of-the-art reasoning benchmarks such as ProofWriter and CLUTRR demonstrates that Spectral NSR achieves superior accuracy, faster inference, improved robustness to adversarial perturbations, and higher interpretability compared to leading baselines including transformers, message-passing neural networks, and neuro-symbolic logic programming systems. Spectral attribution and proof-band agreement analyses confirm that model decisions align closely with symbolic proof structures, while transfer experiments validate effective domain adaptation through co-spectral alignment. These results establish Spectral NSR as a scalable and principled foundation for the next generation of reasoning systems, offering transparency, robustness, and generalization beyond conventional approaches.
Authors: Edgar Dobriban
Abstract: Generative Artificial Intelligence is emerging as an important technology, promising to be transformative in many areas. At the same time, generative AI techniques are based on sampling from probabilistic models, and by default, they come with no guarantees about correctness, safety, fairness, or other properties. Statistical methods offer a promising potential approach to improve the reliability of generative AI techniques. In addition, statistical methods are also promising for improving the quality and efficiency of AI evaluation, as well as for designing interventions and experiments in AI. In this paper, we review some of the existing work on these topics, explaining both the general statistical techniques used, as well as their applications to generative AI. We also discuss limitations and potential future directions.
Authors: Yinheng Li, Hailey Hultquist, Justin Wagle, Kazuhito Koishida
Abstract: Graphical user interface (GUI) agents have advanced rapidly but still struggle with complex tasks involving novel UI elements, long-horizon actions, and personalized trajectories. In this work, we introduce Instruction Agent, a GUI agent that leverages expert demonstrations to solve such tasks, enabling completion of otherwise difficult workflows. Given a single demonstration, the agent extracts step-by-step instructions and executes them by strictly following the trajectory intended by the user, which avoids making mistakes during execution. The agent leverages the verifier and backtracker modules further to improve robustness. Both modules are critical to understand the current outcome from each action and handle unexpected interruptions(such as pop-up windows) during execution. Our experiments show that Instruction Agent achieves a 60% success rate on a set of tasks in OSWorld that all top-ranked agents failed to complete. The Instruction Agent offers a practical and extensible framework, bridging the gap between current GUI agents and reliable real-world GUI task automation.
Authors: Sania Sinha, Tanawan Premsri, Danial Kamali, Parisa Kordjamshidi
Abstract: Neurosymbolic (NeSy) frameworks combine neural representations and learning with symbolic representations and reasoning. Combining the reasoning capacities, explainability, and interpretability of symbolic processing with the flexibility and power of neural computing allows us to solve complex problems with more reliability while being data-efficient. However, this recently growing topic poses a challenge to developers with its learning curve, lack of user-friendly tools, libraries, and unifying frameworks. In this paper, we characterize the technical facets of existing NeSy frameworks, such as the symbolic representation language, integration with neural models, and the underlying algorithms. A majority of the NeSy research focuses on algorithms instead of providing generic frameworks for declarative problem specification to leverage problem solving. To highlight the key aspects of Neurosymbolic modeling, we showcase three generic NeSy frameworks - \textit{DeepProbLog}, \textit{Scallop}, and \textit{DomiKnowS}. We identify the challenges within each facet that lay the foundation for identifying the expressivity of each framework in solving a variety of problems. Building on this foundation, we aim to spark transformative action and encourage the community to rethink this problem in novel ways.
Authors: Farnoush Baghestani, Jihye Moon, Youngsun Kong, Ki Chon
Abstract: The sympathetic nervous system (SNS) plays a central role in regulating the body's responses to stress and maintaining physiological stability. Its dysregulation is associated with a wide range of conditions, from cardiovascular disease to anxiety disorders. Skin nerve activity (SKNA) extracted from high-frequency electrocardiogram (ECG) recordings provides a noninvasive window into SNS dynamics, but its measurement is highly susceptible to electromyographic (EMG) contamination. Traditional preprocessing based on bandpass filtering within a fixed range (e.g., 500--1000 Hz) is susceptible to overlapping EMG and SKNA spectral components, especially during sustained muscle activity. We present a denoising approach using a lightweight one-dimensional convolutional autoencoder with a long short-term memory (LSTM) bottleneck to reconstruct clean SKNA from EMG-contaminated recordings. Using clean ECG-derived SKNA data from cognitive stress experiments and EMG noise from chaotic muscle stimulation recordings, we simulated contamination at realistic noise levels (--4 dB, --8 dB signal-to-noise ratio) and trained the model in the leave-one-subject-out cross-validation framework. The method improved signal-to-noise ratio by up to 9.65 dB, increased cross correlation with clean SKNA from 0.40 to 0.72, and restored burst-based SKNA features to near-clean discriminability (AUROC $\geq$ 0.96). Classification of baseline versus sympathetic stimulation (cognitive stress) conditions reached accuracies of 91--98\% across severe noise levels, comparable to clean data. These results demonstrate that deep learning--based reconstruction can preserve physiologically relevant sympathetic bursts during substantial EMG interference, enabling more robust SKNA monitoring in naturalistic, movement-rich environments.
Authors: Heng Hao, Wenjun Hu, Oxana Verkholyak, Davoud Ataee Tarzanagh, Baruch Gutow, Sima Didari, Masoud Faraki, Hankyu Moon, Seungjai Min
Abstract: Text-to-SQL models allow users to interact with a database more easily by generating executable SQL statements from natural-language questions. Despite recent successes on simpler databases and questions, current Text-to-SQL methods still suffer from low execution accuracy on industry-scale databases and complex questions involving domain-specific business logic. We present \emph{PaVeRL-SQL}, a framework that combines \emph{Partial-Match Rewards} and \emph{Verbal Reinforcement Learning} to drive self-improvement in reasoning language models (RLMs) for Text-to-SQL. To handle practical use cases, we adopt two pipelines: (1) a newly designed in-context learning framework with group self-evaluation (verbal-RL), using capable open- and closed-source large language models (LLMs) as backbones; and (2) a chain-of-thought (CoT) RL pipeline with a small backbone model (OmniSQL-7B) trained with a specially designed reward function and two-stage RL. These pipelines achieve state-of-the-art (SOTA) results on popular Text-to-SQL benchmarks -- Spider, Spider 2.0, and BIRD. For the industrial-level Spider2.0-SQLite benchmark, the verbal-RL pipeline achieves an execution accuracy 7.4\% higher than SOTA, and the CoT pipeline is 1.4\% higher. RL training with mixed SQL dialects yields strong, threefold gains, particularly for dialects with limited training data. Overall, \emph{PaVeRL-SQL} delivers reliable, SOTA Text-to-SQL under realistic industrial constraints. The code is available at https://github.com/PaVeRL-SQL/PaVeRL-SQL.
Authors: Quinten Steenhuis
Abstract: Each year millions of people seek help for their legal problems by calling a legal aid program hotline, walking into a legal aid office, or using a lawyer referral service. The first step to match them to the right help is to identify the legal problem the applicant is experiencing. Misdirection has consequences. Applicants may miss a deadline, experience physical abuse, lose housing or lose custody of children while waiting to connect to the right legal help. We introduce and evaluate the FETCH classifier for legal issue classification and describe two methods for improving accuracy: a hybrid LLM/ML ensemble classification method, and the automatic generation of follow-up questions to enrich the initial problem narrative. We employ a novel data set of 419 real-world queries to a nonprofit lawyer referral service. Ultimately, we show classification accuracy (hits@2) of 97.37\% using a mix of inexpensive models, exceeding the performance of the current state-of-the-art GPT-5 model. Our approach shows promise in significantly reducing the cost of guiding users of the legal system to the right resource for their problem while achieving high accuracy.
Authors: Abdulhakim Alsaiari, Mohammad Ilyas
Abstract: The evolution of the traditional power grid into the "smart grid" has resulted in a fundamental shift in energy management, which allows the integration of renewable energy sources with modern communication technology. However, this interconnection has increased smart grids' vulnerability to attackers, which might result in privacy breaches, operational interruptions, and massive outages. The SCADA-based smart grid protocols are critical for real-time data collection and control, but they are vulnerable to attacks like unauthorized access and denial of service (DoS). This research proposes a hybrid deep learning-based Intrusion Detection System (IDS) intended to improve the cybersecurity of smart grids. The suggested model takes advantage of Convolutional Neural Networks' (CNN) feature extraction capabilities as well as Long Short-Term Memory (LSTM) networks' temporal pattern recognition skills. DNP3 and IEC104 intrusion detection datasets are employed to train and test our CNN-LSTM model to recognize and classify the potential cyber threats. Compared to other deep learning approaches, the results demonstrate considerable improvements in accuracy, precision, recall, and F1-score, with a detection accuracy of 99.70%.
Authors: Nicholas Sung, Steven Spreizer, Mohamed Elrefaie, Kaira Samuel, Matthew C. Jones, Faez Ahmed
Abstract: BlendedNet is a publicly available aerodynamic dataset of 999 blended wing body (BWB) geometries. Each geometry is simulated across about nine flight conditions, yielding 8830 converged RANS cases with the Spalart-Allmaras model and 9 to 14 million cells per case. The dataset is generated by sampling geometric design parameters and flight conditions, and includes detailed pointwise surface quantities needed to study lift and drag. We also introduce an end-to-end surrogate framework for pointwise aerodynamic prediction. The pipeline first uses a permutation-invariant PointNet regressor to predict geometric parameters from sampled surface point clouds, then conditions a Feature-wise Linear Modulation (FiLM) network on the predicted parameters and flight conditions to predict pointwise coefficients Cp, Cfx, and Cfz. Experiments show low errors in surface predictions across diverse BWBs. BlendedNet addresses data scarcity for unconventional configurations and enables research on data-driven surrogate modeling for aerodynamic design.
Authors: Siddhant Karki, Ethan Han, Nadim Mahmud, Suman Bhunia, John Femiani, Vaskar Raychoudhury
Abstract: Individuals with ambulatory disabilities often encounter significant barriers when navigating urban environments due to the lack of accessible information and tools. This paper presents OmniAcc, an AI-powered interactive navigation system that utilizes GPT-4, satellite imagery, and OpenStreetMap data to identify, classify, and map wheelchair-accessible features such as ramps and crosswalks in the built environment. OmniAcc offers personalized route planning, real-time hands-free navigation, and instant query responses regarding physical accessibility. By using zero-shot learning and customized prompts, the system ensures precise detection of accessibility features, while supporting validation through structured workflows. This paper introduces OmniAcc and explores its potential to assist urban planners and mobility-aid users, demonstrated through a case study on crosswalk detection. With a crosswalk detection accuracy of 97.5%, OmniAcc highlights the transformative potential of AI in improving navigation and fostering more inclusive urban spaces.
Authors: Xin Wang, Ting Dang, Xinyu Zhang, Vassilis Kostakos, Michael J. Witbrock, Hong Jia
Abstract: Mobile and wearable healthcare monitoring play a vital role in facilitating timely interventions, managing chronic health conditions, and ultimately improving individuals' quality of life. Previous studies on large language models (LLMs) have highlighted their impressive generalization abilities and effectiveness in healthcare prediction tasks. However, most LLM-based healthcare solutions are cloud-based, which raises significant privacy concerns and results in increased memory usage and latency. To address these challenges, there is growing interest in compact models, Small Language Models (SLMs), which are lightweight and designed to run locally and efficiently on mobile and wearable devices. Nevertheless, how well these models perform in healthcare prediction remains largely unexplored. We systematically evaluated SLMs on health prediction tasks using zero-shot, few-shot, and instruction fine-tuning approaches, and deployed the best performing fine-tuned SLMs on mobile devices to evaluate their real-world efficiency and predictive performance in practical healthcare scenarios. Our results show that SLMs can achieve performance comparable to LLMs while offering substantial gains in efficiency and privacy. However, challenges remain, particularly in handling class imbalance and few-shot scenarios. These findings highlight SLMs, though imperfect in their current form, as a promising solution for next-generation, privacy-preserving healthcare monitoring.
Authors: Vardhan Palod, Karthik Valmeekam, Kaya Stechly, Subbarao Kambhampati
Abstract: Intermediate token generation (ITG), where a model produces output before the solution, has been proposed as a method to improve the performance of language models on reasoning tasks. While these reasoning traces or Chain of Thoughts (CoTs) are correlated with performance gains, the mechanisms underlying them remain unclear. A prevailing assumption in the community has been to anthropomorphize these tokens as "thinking", treating longer traces as evidence of higher problem-adaptive computation. In this work, we critically examine whether intermediate token sequence length reflects or correlates with problem difficulty. To do so, we train transformer models from scratch on derivational traces of the A* search algorithm, where the number of operations required to solve a maze problem provides a precise and verifiable measure of problem complexity. We first evaluate the models on trivial free-space problems, finding that even for the simplest tasks, they often produce excessively long reasoning traces and sometimes fail to generate a solution. We then systematically evaluate the model on out-of-distribution problems and find that the intermediate token length and ground truth A* trace length only loosely correlate. We notice that the few cases where correlation appears are those where the problems are closer to the training distribution, suggesting that the effect arises from approximate recall rather than genuine problem-adaptive computation. This suggests that the inherent computational complexity of the problem instance is not a significant factor, but rather its distributional distance from the training data. These results challenge the assumption that intermediate trace generation is adaptive to problem difficulty and caution against interpreting longer sequences in systems like R1 as automatically indicative of "thinking effort".
Authors: Cunxi Yu, Rongjian Liang, Chia-Tung Ho, Haoxing Ren
Abstract: Large language models (LLMs) have recently shown strong coding abilities, enabling not only static code generation but also iterative code self-evolving through agentic frameworks. Recently, AlphaEvolve \cite{novikov2025alphaevolve} demonstrated that LLM-based coding agents can autonomously improve algorithms and surpass human experts, with scopes limited to isolated kernels spanning hundreds of lines of code. Inspired by AlphaEvolve, we present SATLUTION, the first framework to extend LLM-based code evolution to the full repository scale, encompassing hundreds of files and tens of thousands of lines of C/C++ code. Targeting Boolean Satisfiability (SAT), the canonical NP-complete problem and a cornerstone of both theory and applications. SATLUTION orchestrates LLM agents to directly evolve solver repositories under strict correctness guarantees and distributed runtime feedback, while simultaneously self-evolving its own evolution policies and rules. Starting from SAT Competition 2024 codebases and benchmark, SATLUTION evolved solvers that decisively outperformed the human-designed winners of the SAT Competition 2025, and also surpassed both 2024 and 2025 champions on the 2024 benchmarks.
Authors: Jakub Grudzien Kuba, Mengting Gu, Qi Ma, Yuandong Tian, Vijai Mohan
Abstract: Large language models (LLMs) have advanced rapidly in recent years, driven by scale, abundant high-quality training data, and reinforcement learning. Yet this progress faces a fundamental bottleneck: the need for ever more data from which models can continue to learn. In this work, we propose a reinforcement learning approach that removes this dependency by enabling models to improve without additional data. Our method leverages a game-theoretic framework of self-play, where a model's capabilities are cast as performance in a competitive game and stronger policies emerge by having the model play against itself - a process we call Language Self-Play (LSP). Experiments with Llama-3.2-3B-Instruct on instruction-following benchmarks show that pretrained models can not only enhance their performance on challenging tasks through self-play alone, but can also do so more effectively than data-driven baselines.
Authors: Qin Chen, Yuanyi Ren, Xiaojun Ma, Mugeng Liu, Han Shi, Dongmei Zhang
Abstract: Spreadsheets are critical to data-centric tasks, with rich, structured layouts that enable efficient information transmission. Given the time and expertise required for manual spreadsheet layout design, there is an urgent need for automated solutions. However, existing automated layout models are ill-suited to spreadsheets, as they often (1) treat components as axis-aligned rectangles with continuous coordinates, overlooking the inherently discrete, grid-based structure of spreadsheets; and (2) neglect interrelated semantics, such as data dependencies and contextual links, unique to spreadsheets. In this paper, we first formalize the spreadsheet layout generation task, supported by a seven-criterion evaluation protocol and a dataset of 3,326 spreadsheets. We then introduce SheetDesigner, a zero-shot and training-free framework using Multimodal Large Language Models (MLLMs) that combines rule and vision reflection for component placement and content population. SheetDesigner outperforms five baselines by at least 22.6\%. We further find that through vision modality, MLLMs handle overlap and balance well but struggle with alignment, necessitates hybrid rule and visual reflection strategies. Our codes and data is available at Github.
Authors: Riccardo D'Elia, Alberto Termine, Francesco Flammini
Abstract: The integration of Deep Learning (DL) in System Dynamics (SD) modeling for transportation logistics offers significant advantages in scalability and predictive accuracy. However, these gains are often offset by the loss of explainability and causal reliability $-$ key requirements in critical decision-making systems. This paper presents a novel framework for interpretable-by-design neural system dynamics modeling that synergizes DL with techniques from Concept-Based Interpretability, Mechanistic Interpretability, and Causal Machine Learning. The proposed hybrid approach enables the construction of neural network models that operate on semantically meaningful and actionable variables, while retaining the causal grounding and transparency typical of traditional SD models. The framework is conceived to be applied to real-world case-studies from the EU-funded project AutoMoTIF, focusing on data-driven decision support, automation, and optimization of multimodal logistic terminals. We aim at showing how neuro-symbolic methods can bridge the gap between black-box predictive models and the need for critical decision support in complex dynamical environments within cyber-physical systems enabled by the industrial Internet-of-Things.
Authors: Minghui Li, Hao Zhang, Yechao Zhang, Wei Wan, Shengshan Hu, pei Xiaobing, Jing Wang
Abstract: Direct Prompt Injection (DPI) attacks pose a critical security threat to Large Language Models (LLMs) due to their low barrier of execution and high potential damage. To address the impracticality of existing white-box/gray-box methods and the poor transferability of black-box methods, we propose an activations-guided prompt injection attack framework. We first construct an Energy-based Model (EBM) using activations from a surrogate model to evaluate the quality of adversarial prompts. Guided by the trained EBM, we employ the token-level Markov Chain Monte Carlo (MCMC) sampling to adaptively optimize adversarial prompts, thereby enabling gradient-free black-box attacks. Experimental results demonstrate our superior cross-model transferability, achieving 49.6% attack success rate (ASR) across five mainstream LLMs and 34.6% improvement over human-crafted prompts, and maintaining 36.6% ASR on unseen task scenarios. Interpretability analysis reveals a correlation between activations and attack effectiveness, highlighting the critical role of semantic patterns in transferable vulnerability exploitation.
Authors: Sascha Kaltenpoth, Oliver M\"uller
Abstract: Adopting Large language models (LLMs) in organizations potentially revolutionizes our lives and work. However, they can generate off-topic, discriminating, or harmful content. This AI alignment problem often stems from misspecifications during the LLM adoption, unnoticed by the principal due to the LLM's black-box nature. While various research disciplines investigated AI alignment, they neither address the information asymmetries between organizational adopters and black-box LLM agents nor consider organizational AI adoption processes. Therefore, we propose LLM ATLAS (LLM Agency Theory-Led Alignment Strategy) a conceptual framework grounded in agency (contract) theory, to mitigate alignment problems during organizational LLM adoption. We conduct a conceptual literature analysis using the organizational LLM adoption phases and the agency theory as concepts. Our approach results in (1) providing an extended literature analysis process specific to AI alignment methods during organizational LLM adoption and (2) providing a first LLM alignment problem-solution space.
Authors: Adem Kikaj, Giuseppe Marra, Floris Geerts, Robin Manhaeve, Luc De Raedt
Abstract: Neurosymbolic AI (NeSy) aims to integrate the statistical strengths of neural networks with the interpretability and structure of symbolic reasoning. However, current NeSy frameworks like DeepProbLog enforce a fixed flow where symbolic reasoning always follows neural processing. This restricts their ability to model complex dependencies, especially in irregular data structures such as graphs. In this work, we introduce DeepGraphLog, a novel NeSy framework that extends ProbLog with Graph Neural Predicates. DeepGraphLog enables multi-layer neural-symbolic reasoning, allowing neural and symbolic components to be layered in arbitrary order. In contrast to DeepProbLog, which cannot handle symbolic reasoning via neural methods, DeepGraphLog treats symbolic representations as graphs, enabling them to be processed by Graph Neural Networks (GNNs). We showcase the capabilities of DeepGraphLog on tasks in planning, knowledge graph completion with distant supervision, and GNN expressivity. Our results demonstrate that DeepGraphLog effectively captures complex relational dependencies, overcoming key limitations of existing NeSy systems. By broadening the applicability of neurosymbolic AI to graph-structured domains, DeepGraphLog offers a more expressive and flexible framework for neural-symbolic integration.
Authors: Jipeng Li, Zeyu Gao, Yubin Qi, Hande Dong, Weijian Chen, Qiang Lin
Abstract: Large Language Models (LLMs) have achieved remarkable performance across diverse tasks, yet their susceptibility to generating incorrect content during inference remains a critical unsolved challenge. While self-correction methods offer potential solutions, their effectiveness is hindered by two inherent limitations: (1) the absence of reliable guidance signals for error localization, and (2) the restricted reasoning depth imposed by conventional next-token decoding paradigms. To address these issues, we propose Feedback-Triggered Regeneration (FTR), a novel framework that synergizes user feedback with enhanced decoding dynamics. Specifically, FTR activates response regeneration only upon receiving negative user feedback, thereby circumventing error propagation from faulty self-assessment while preserving originally correct outputs. Furthermore, we introduce Long-Term Multipath (LTM) decoding, which enables systematic exploration of multiple reasoning trajectories through delayed sequence evaluation, effectively overcoming the myopic decision-making characteristic of standard next-token prediction. Extensive experiments on mathematical reasoning and code generation benchmarks demonstrate that our framework achieves consistent and significant improvements over state-of-the-art prompt-based self-correction methods.
Authors: Yildiray Kabak, Gokce B. Laleci Erturkmen, Mert Gencturk, Tuncay Namli, A. Anil Sinaci, Ruben Alcantud Corcoles, Cristina Gomez Ballesteros, Pedro Abizanda, Asuman Dogac
Abstract: In this study, we propose FHIR-RAG-MEDS system that aims to integrate Health Level 7 Fast Healthcare Interoperability Resources (HL7 FHIR) with a Retrieval-Augmented Generation (RAG)-based system to improve personalized medical decision support on evidence-based clinical guidelines, emphasizing the need for research in practical applications. In the evolving landscape of medical decision support systems, integrating advanced technologies such as RAG and HL7 FHIR can significantly enhance clinical decision-making processes. Despite the potential of these technologies, there is limited research on their integration in practical applications.
Authors: Ziye Chen, Chengwei Qin, Yao Shu
Abstract: As large language models (LLMs) reach high scores on established mathematical benchmarks, such as GSM8K and MATH, the research community has turned to International Mathematical Olympiad (IMO) problems to push the evaluation frontier. However, existing Olympiad-level benchmarks suffer from practical constraints that introduce grading noise and potential bias, such as heterogeneous answer formats requiring model-based judges and a reliance on potentially flawed solutions. We introduce RIMO, a two-track benchmark designed to preserve peak Olympiad difficulty while eliminating this evaluation noise. The first track, RIMO-N, rewrites 335 IMO problems to admit a single, unique integer answer, allowing for deterministic correctness checking. The second track, RIMO-P, features 456 proof problems with expert-checked solutions, which are decomposed into a sequence of sub-problems to evaluate the step-by-step reasoning process via an automated grading system. Our benchmarking of ten frontier LLMs, including GPT-4o and Gemini 2.5 Flash, reveals that while these systems excel on older benchmarks, their performance drops sharply on RIMO. These results highlight a substantial gap between current LLM capabilities and actual Olympiad-level reasoning. By providing a challenging yet easy-to-evaluate suite, RIMO offers a high-resolution yardstick for future research, presenting a clear target for closing the profound reasoning gap our findings expose.
Authors: Bo Yu, Zhixiu Hua, Bo Zhao
Abstract: Background: Parkinson's disease remains a major neurodegenerative disorder with high misdiagnosis rates, primarily due to reliance on clinical rating scales. Recent studies have demonstrated a strong association between gut microbiota and Parkinson's disease, suggesting that microbial composition may serve as a promising biomarker. Although deep learning models based ongut microbiota show potential for early prediction, most approaches rely on single classifiers and often overlook inter-strain correlations or temporal dynamics. Therefore, there is an urgent need for more robust feature extraction methods tailored to microbiome data. Methods: We proposed BDPM (A Machine Learning-Based Feature Extractor for Parkinson's Disease Classification via Gut Microbiota Analysis). First, we collected gut microbiota profiles from 39 Parkinson's patients and their healthy spouses to identify differentially abundant taxa. Second, we developed an innovative feature selection framework named RFRE (Random Forest combined with Recursive Feature Elimination), integrating ecological knowledge to enhance biological interpretability. Finally, we designed a hybrid classification model to capture temporal and spatial patterns in microbiome data.
Authors: Mustafa Kaan Aslan, Reinout Heijungs, Filip Ilievski
Abstract: Environmental sustainability, particularly in relation to climate change, is a key concern for consumers, producers, and policymakers. The carbon footprint, based on greenhouse gas emissions, is a standard metric for quantifying the contribution to climate change of activities and is often assessed using life cycle assessment (LCA). However, conducting LCA is complex due to opaque and global supply chains, as well as fragmented data. This paper presents a methodology that combines advances in LCA and publicly available databases with knowledge-augmented AI techniques, including retrieval-augmented generation, to estimate cradle-to-gate carbon footprints of food products. We introduce a chatbot interface that allows users to interactively explore the carbon impact of composite meals and relate the results to familiar activities. A live web demonstration showcases our proof-of-concept system with arbitrary food items and follow-up questions, highlighting both the potential and limitations - such as database uncertainties and AI misinterpretations - of delivering LCA insights in an accessible format.
Authors: Jo\~ao Paulo Nogueira, Wentao Sun, Alonso Silva, Laith Zumot
Abstract: The rise of large reasoning language models (LRLMs) has unlocked new potential for solving complex tasks. These models operate with a thinking budget, that is, a predefined number of reasoning tokens used to arrive at a solution. We propose a novel approach, inspired by the generator/discriminator framework in generative adversarial networks, in which a critic model periodically probes its own reasoning to assess whether it has reached a confident conclusion. If not, reasoning continues until a target certainty threshold is met. This mechanism adaptively balances efficiency and reliability by allowing early termination when confidence is high, while encouraging further reasoning when uncertainty persists. Through experiments on the AIME2024 and AIME2025 datasets, we show that Certainty-Guided Reasoning (CGR) improves baseline accuracy while reducing token usage. Importantly, extended multi-seed evaluations over 64 runs demonstrate that CGR is stable, reducing variance across seeds and improving exam-like performance under penalty-based grading. Additionally, our token savings analysis shows that CGR can eliminate millions of tokens in aggregate, with tunable trade-offs between certainty thresholds and efficiency. Together, these findings highlight certainty as a powerful signal for reasoning sufficiency. By integrating confidence into the reasoning process, CGR makes large reasoning language models more adaptive, trustworthy, and resource efficient, paving the way for practical deployment in domains where both accuracy and computational cost matter.
Authors: Amay Jain, Liu Cui, Si Chen
Abstract: Large language models like ChatGPT are increasingly used in classrooms, but they often provide outdated or fabricated information that can mislead students. Retrieval Augmented Generation (RAG) improves reliability of LLMs by grounding responses in external resources. We investigate two accessible RAG paradigms, vector-based retrieval and graph-based retrieval to identify best practices for classroom question answering (QA). Existing comparative studies fail to account for pedagogical factors such as educational disciplines, question types, and practical deployment costs. Using a novel dataset, EduScopeQA, of 3,176 questions across academic subjects, we measure performance on various educational query types, from specific facts to broad thematic discussions. We also evaluate system alignment with a dataset of systematically altered textbooks that contradict the LLM's latent knowledge. We find that OpenAI Vector Search RAG (representing vector-based RAG) performs well as a low-cost generalist, especially for quick fact retrieval. On the other hand, GraphRAG Global excels at providing pedagogically rich answers to thematic queries, and GraphRAG Local achieves the highest accuracy with the dense, altered textbooks when corpus integrity is critical. Accounting for the 10-20x higher resource usage of GraphRAG (representing graph-based RAG), we show that a dynamic branching framework that routes queries to the optimal retrieval method boosts fidelity and efficiency. These insights provide actionable guidelines for educators and system designers to integrate RAG-augmented LLMs into learning environments effectively.
Authors: Xinyu Zhang, Changzhi Zhou, Linmei Hu, Luhao Zhang, Xiancai Chen, Haomin Fu, Yang Yang, Mengdi Zhang
Abstract: Existing code large language models (LLMs) often rely on large-scale instruction data distilled from proprietary LLMs for fine-tuning, which typically incurs high costs. In this paper, we explore the potential of small-scale open-source LLMs (e.g., 7B) as synthesizers for high-quality code instruction data construction. We first observe that the data synthesis capability of small-scale LLMs can be enhanced by training on a few superior data synthesis samples from proprietary LLMs. Building on this, we propose a novel iterative self-distillation approach to bootstrap small-scale LLMs, transforming them into powerful synthesizers that reduce reliance on proprietary LLMs and minimize costs. Concretely, in each iteration, to obtain diverse and high-quality self-distilled data, we design multi-checkpoint sampling and multi-aspect scoring strategies for initial data selection. Furthermore, to identify the most influential samples, we introduce a gradient-based influence estimation method for final data filtering. Based on the code instruction datasets from the small-scale synthesizers, we develop SCoder, a family of code generation models fine-tuned from DeepSeek-Coder. SCoder models achieve state-of-the-art code generation capabilities, demonstrating the effectiveness of our method.
Authors: Augustin Crespin, Ioannis Kostis, H\'el\`ene Verhaeghe, Pierre Schaus
Abstract: Constraint Programming and its high-level modeling languages have long been recognized for their potential to achieve the holy grail of problem-solving. However, the complexity of modeling languages, the large number of global constraints, and the art of creating good models have often hindered non-experts from choosing CP to solve their combinatorial problems. While generating an expert-level model from a natural-language description of a problem would be the dream, we are not yet there. We propose a tutoring system called CP-Model-Zoo, exploiting expert-written models accumulated through the years. CP-Model-Zoo retrieves the closest source code model from a database based on a user's natural language description of a combinatorial problem. It ensures that expert-validated models are presented to the user while eliminating the need for human data labeling. Our experiments show excellent accuracy in retrieving the correct model based on a user-input description of a problem simulated with different levels of expertise.
Authors: Fangchen Yu, Haiyuan Wan, Qianjia Cheng, Yuchen Zhang, Jiacheng Chen, Fujun Han, Yulun Wu, Junchi Yao, Ruilizhen Hu, Ning Ding, Yu Cheng, Tao Chen, Lei Bai, Dongzhan Zhou, Yun Luo, Ganqu Cui, Peng Ye
Abstract: Recently, the physical capabilities of (M)LLMs have garnered increasing attention. However, existing benchmarks for physics suffer from two major gaps: they neither provide systematic and up-to-date coverage of real-world physics competitions such as physics Olympiads, nor enable direct performance comparison with humans. To bridge these gaps, we present HiPhO, the first benchmark dedicated to high school physics Olympiads with human-aligned evaluation. Specifically, HiPhO highlights three key innovations. (1) Comprehensive Data: It compiles 13 latest Olympiad exams from 2024-2025, spanning both international and regional competitions, and covering mixed modalities that encompass problems spanning text-only to diagram-based. (2) Professional Evaluation: We adopt official marking schemes to perform fine-grained grading at both the answer and step level, fully aligned with human examiners to ensure high-quality and domain-specific evaluation. (3) Comparison with Human Contestants: We assign gold, silver, and bronze medals to models based on official medal thresholds, thereby enabling direct comparison between (M)LLMs and human contestants. Our large-scale evaluation of 30 state-of-the-art (M)LLMs shows that: across 13 exams, open-source MLLMs mostly remain at or below the bronze level; open-source LLMs show promising progress with occasional golds; closed-source reasoning MLLMs can achieve 6 to 12 gold medals; and most models still have a significant gap from full marks. These results highlight a substantial performance gap between open-source models and top students, the strong physical reasoning capabilities of closed-source reasoning models, and the fact that there is still significant room for improvement. HiPhO, as a rigorous, human-aligned, and Olympiad-focused benchmark for advancing multimodal physical reasoning, is open-source and available at https://github.com/SciYu/HiPhO.
Authors: Valen Tagliabue, Leonard Dung
Abstract: We develop new experimental paradigms for measuring welfare in language models. We compare verbal reports of models about their preferences with preferences expressed through behavior when navigating a virtual environment and selecting conversation topics. We also test how costs and rewards affect behavior and whether responses to an eudaimonic welfare scale - measuring states such as autonomy and purpose in life - are consistent across semantically equivalent prompts. Overall, we observed a notable degree of mutual support between our measures. The reliable correlations observed between stated preferences and behavior across conditions suggest that preference satisfaction can, in principle, serve as an empirically measurable welfare proxy in some of today's AI systems. Furthermore, our design offered an illuminating setting for qualitative observation of model behavior. Yet, the consistency between measures was more pronounced in some models and conditions than others and responses were not consistent across perturbations. Due to this, and the background uncertainty about the nature of welfare and the cognitive states (and welfare subjecthood) of language models, we are currently uncertain whether our methods successfully measure the welfare state of language models. Nevertheless, these findings highlight the feasibility of welfare measurement in language models, inviting further exploration.
Authors: Zhenyu Yu, Jinnian Wang
Abstract: Forests function as crucial carbon reservoirs on land, and their carbon sinks can efficiently reduce atmospheric CO2 concentrations and mitigate climate change. Currently, the overall trend for monitoring and assessing forest carbon stocks is to integrate ground monitoring sample data with satellite remote sensing imagery. This style of analysis facilitates large-scale observation. However, these techniques require improvement in accuracy. We used GF-1 WFV and Landsat TM images to analyze Huize County, Qujing City, Yunnan Province in China. Using the style transfer method, we introduced Swin Transformer to extract global features through attention mechanisms, converting the carbon stock estimation into an image translation.
Authors: Jiahuan Yu (University of Illinois Urbana-Champaign), Aryan Taneja (University of Illinois Urbana-Champaign), Junfeng Lin (Tsinghua University), Minjia Zhang (University of Illinois Urbana-Champaign)
Abstract: Modern Large Language Model (LLM) serving systems increasingly support interactive applications, like real-time chat assistants, code generation tools, and agentic workflows. However, the soaring energy cost of LLM inference presents a growing challenge for sustainable and cost-effective deployment. This paper introduces VoltanaLLM, a system for SLO-aware, energy-efficient LLM serving, built from a control theory perspective. VoltanaLLM co-designs frequency scaling and request routing in emerging prefill/decode disaggregated architectures, leveraging their decoupled execution to enable fine-grained phase-specific control. It consists of a feedback-driven frequency controller that dynamically adapts GPU frequency for prefill and decode phases, and a state-space router that explores routing decisions across frequency-scaled instances to minimize energy under latency constraints. We implement VoltanaLLM in SGLang and evaluate its performance over multiple state-of-the-art LLMs and real-world datasets. The results demonstrate that VoltanaLLM achieves up to 36.3% energy savings while maintaining near-perfect SLO attainment rate, paving the way for sustainable and intelligent LLM serving.
Authors: Neal G. Ravindra, Arijit Sehanobish
Abstract: High-quality, medically validated labels exist for clinical actigraphy data but not for ubiquitous consumer wearables like the Apple Watch. Manually labeling wearables data is expensive and doesn't scale. This paper offers a novel framework that transfers valuable labels from a source domain (e.g., actigraphy) to a target domain (e.g., Apple Watch) without requiring paired data. Instead of working with raw time-series signals, we project both domains into a shared latent embedding space using time-series foundation models (TSFMs) and develop a new framework to align the cross-device representations. Our method, Adversarial Alignment of TSFM Embeddings forces the distributions of source and target embeddings to align within this space, facilitating label transfer across device type.
Authors: Tianyu Huo, Jian Xiong, Yiyan Wu, Songjie Yang, Bo Liu, Wenjun Zhang
Abstract: Extremely large antenna array (ELAA) is key to enhancing spectral efficiency in 6G networks. Leveraging the distributed nature of multi-unmanned aerial vehicle (UAV) systems enables the formation of distributed ELAA, which often operate in the near-field region with spatial sparsity, rendering the conventional far-field plane wave assumption invalid. This paper investigates channel estimation for distributed near-field multi-UAV communication systems. We first derive closed-form signal-to-noise ratio (SNR) expressions under the plane wave model (PWM), spherical wave model (SWM), and a hybrid spherical-plane wave model (HSPWM), also referred to as the cross-field model, within a distributed uniform planar array (UPA) scenario. The analysis shows that HSPWM achieves a good balance between modeling accuracy and analytical tractability. Based on this, we propose two channel estimation algorithms: the spherical-domain orthogonal matching pursuit (SD-OMP) and the tensor-OMP. The SD-OMP generalizes the polar domain to jointly consider elevation, azimuth, and range. Under the HSPWM, the channel is naturally formulated as a tensor, enabling the use of tensor-OMP. Simulation results demonstrate that tensor-OMP achieves normalized mean square error (NMSE) performance comparable to SD-OMP, while offering reduced computational complexity and improved scalability.
Authors: Murat Temiz, Yongwei Zhang, Yanwei Fu, Chi Zhang, Chenfeng Meng, Orhan Kaplan, Christos Masouros
Abstract: This article comprehensively reviews recent developments and research on deep learning-based (DL-based) techniques for integrated sensing and communication (ISAC) systems. ISAC, which combines sensing and communication functionalities, is regarded as a key enabler for 6G and beyond networks, as many emerging applications, such as vehicular networks and industrial robotics, necessitate both sensing and communication capabilities for effective operation. A unified platform that provides both functions can reduce hardware complexity, alleviate frequency spectrum congestion, and improve energy efficiency. However, integrating these functionalities on the same hardware requires highly optimized signal processing and system design, introducing significant computational complexity when relying on conventional iterative or optimization-based techniques. As an alternative to conventional techniques, DL-based techniques offer efficient and near-optimal solutions with reduced computational complexity. Hence, such techniques are well-suited for operating under limited computational resources and low latency requirements in real-time systems. DL-based techniques can swiftly and effectively yield near-optimal solutions for a wide range of sophisticated ISAC-related tasks, including waveform design, channel estimation, sensing signal processing, data demodulation, and interference mitigation. Therefore, motivated by these advantages, recent studies have proposed various DL-based approaches for ISAC system design. After briefly introducing DL architectures and ISAC fundamentals, this survey presents a comprehensive and categorized review of state-of-the-art DL-based techniques for ISAC, highlights their key advantages and major challenges, and outlines potential directions for future research.
Authors: Wasif Khan, Lin Gu, Noah Hammarlund, Lei Xing, Joshua K. Wong, Ruogu Fang
Abstract: Physical activity is a modifiable lifestyle factor with potential to support cognitive resilience. However, the association of moderate-to-vigorous physical activity (MVPA) intensity, and timing, with cognitive function and region-specific brain structure remain poorly understood. We analyzed data from 45,892 UK Biobank participants aged 60 years and older with valid wrist-worn accelerometer data, cognitive testing, and structural brain MRI. MVPA was measured both continuously (mins per week) and categorically (thresholded using >=150 min/week based on WHO guidelines). Associations with cognitive performance and regional brain volumes were evaluated using multivariable linear models adjusted for demographic, socioeconomic, and health-related covariates. We conducted secondary analyses on MVPA timing and subgroup effects. Higher MVPA was associated with better performance across cognitive domains, including reasoning, memory, executive function, and processing speed. These associations persisted in fully adjusted models and were higher among participants meeting WHO guidelines. Greater MVPA was also associated with subcortical brain regions (caudate, putamen, pallidum, thalamus), as well as regional gray matter volumes involved in emotion, working memory, and perceptual processing. Secondary analyses showed that MVPA at any time of day was associated with cognitive functions and brain volume particularly in the midday-afternoon and evening. Sensitivity analysis shows consistent findings across subgroups, with evidence of dose-response relationships. Higher MVPA is associated with preserved brain structure and enhanced cognitive function in later life. Public health strategies to increase MVPA may support healthy cognitive aging and generate substantial economic benefits, with global gains projected to reach USD 760 billion annually by 2050.
Authors: Zofia Rudnicka, Janusz Szczepanski, Agnieszka Pregowska
Abstract: This study explores how the selection of neuron models and learning rules impacts the classification performance of Spiking Neural Networks (SNNs), with a focus on applications in bio-signal processing. We compare biologically inspired neuron models, including Leaky Integrate-and-Fire (LIF), metaneurons, and probabilistic Levy-Baxter (LB) neurons, across multiple learning rules, including spike-timing-dependent plasticity (STDP), tempotron, and reward-modulated updates. A novel element of this work is the integration of a complexity-based decision mechanism into the evaluation pipeline. Using Lempel-Ziv Complexity (LZC), a measure related to entropy rate, we quantify the structural regularity of spike trains and assess classification outcomes in a consistent and interpretable manner across different SNN configurations. To investigate neural dynamics and assess algorithm performance, we employed synthetic datasets with varying temporal dependencies and stochasticity levels. These included Markov and Poisson processes, well-established models to simulate neuronal spike trains and capture the stochastic firing behavior of biological neurons.Validation of synthetic Poisson and Markov-modeled data reveals clear performance trends: classification accuracy depends on the interaction between neuron model, network size, and learning rule, with the LZC-based evaluation highlighting configurations that remain robust to weak or noisy signals. This work delivers a systematic analysis of how neuron model selection interacts with network parameters and learning strategies, supported by a novel complexity-based evaluation approach that offers a consistent benchmark for SNN performance.
Authors: Xueyi Wang, Elisabeth Wilhelm
Abstract: Sleep quality significantly impacts well-being. Therefore, healthcare providers and individuals need accessible and reliable forecasting tools for preventive interventions. This paper introduces an interpretable, individualized two-stage adaptive spatial-temporal model for predicting sleep quality scores. Our proposed framework combines multi-scale convolutional layers to model spatial interactions across multiple input variables, recurrent layers and attention mechanisms to capture long-term temporal dependencies, and a two-stage domain adaptation strategy to enhance generalization. The first adaptation stage is applied during training to mitigate overfitting on the training set. In the second stage, a source-free test-time adaptation mechanism is employed to adapt the model to new users without requiring labels. We conducted various experiments with five input window sizes (3, 5, 7, 9, and 11 days) and five prediction window sizes (1, 3, 5, 7, and 9 days). Our model consistently outperformed time series forecasting baseline approaches, including Long Short-Term Memory (LSTM), Informer, PatchTST, and TimesNet. The best performance was achieved with a three-day input window and a one-day prediction window, yielding a root mean square error (RMSE) of 0.216. Furthermore, the model demonstrated good predictive performance even for longer forecasting horizons (e.g, with a 0.257 RMSE for a three-day prediction window), highlighting its practical utility for real-world applications. We also conducted an explainability analysis to examine how different features influence sleep quality. These findings proved that the proposed framework offers a robust, adaptive, and explainable solution for personalized sleep forecasting using sparse data from commercial wearable devices.
Authors: Yu Song, Zhigang Hua, Yan Xie, Jingzhe Liu, Bo Long, Hui Liu
Abstract: Self-supervised learning (SSL) has shown great promise in graph representation learning. However, most existing graph SSL methods are developed and evaluated under a single-dataset setting, leaving their cross-dataset transferability largely unexplored and limiting their ability to leverage knowledge transfer and large-scale pretraining, factors that are critical for developing generalized intelligence beyond fitting training data. To address this gap and advance foundation model research for graphs, we present GSTBench, the first systematic benchmark for evaluating the transferability of graph SSL methods. We conduct large-scale pretraining on ogbn-papers100M and evaluate five representative SSL methods across a diverse set of target graphs. Our standardized experimental setup decouples confounding factors such as model architecture, dataset characteristics, and adaptation protocols, enabling rigorous comparisons focused solely on pretraining objectives. Surprisingly, we observe that most graph SSL methods struggle to generalize, with some performing worse than random initialization. In contrast, GraphMAE, a masked autoencoder approach, consistently improves transfer performance. We analyze the underlying factors that drive these differences and offer insights to guide future research on transferable graph SSL, laying a solid foundation for the "pretrain-then-transfer" paradigm in graph learning. Our code is available at https://github.com/SongYYYY/GSTBench.
Authors: Lingyu Zhang, Pengfei Xu, Guobin Wu, Jian Liang, Ruiyang Dong, Yunhai Wang, Xuan Song
Abstract: Traffic demand prediction plays a critical role in intelligent transportation systems. Existing traffic prediction models primarily rely on temporal traffic data, with limited efforts incorporating human knowledge and experience for urban traffic demand forecasting. However, in real-world scenarios, traffic knowledge and experience derived from human daily life significantly influence precise traffic prediction. Such knowledge and experiences can guide the model in uncovering latent patterns within traffic data, thereby enhancing the accuracy and robustness of predictions. To this end, this paper proposes integrating structured temporal traffic data with textual data representing human knowledge and experience, resulting in a novel knowledge-guided cross-modal feature representation learning (KGCM) model for traffic demand prediction. Based on regional transportation characteristics, we construct a prior knowledge dataset using a large language model combined with manual authoring and revision, covering both regional and global knowledge and experiences. The KGCM model then learns multimodal data features through designed local and global adaptive graph networks, as well as a cross-modal feature fusion mechanism. A proposed reasoning-based dynamic update strategy enables dynamic optimization of the graph model's parameters, achieving optimal performance. Experiments on multiple traffic datasets demonstrate that our model accurately predicts future traffic demand and outperforms existing state-of-the-art (SOTA) models.
Authors: Zehua Li
Abstract: This paper presents a configuration-first framework for evaluating cross-backend compatibility in deep learning systems deployed on CPU, GPU, and compiled runtimes. The framework decouples experiments from code using YAML, supports both library and repository models, and employs a three-tier verification protocol covering tensor-level closeness, activation alignment, and task-level metrics. Through 672 checks across multiple models and tolerance settings, we observe that 72.0% of runs pass, with most discrepancies occurring under stricter thresholds. Our results show that detection models and compiled backends are particularly prone to drift, often due to nondeterministic post-processing. We further demonstrate that deterministic adapters and selective fallbacks can substantially improve agreement without significant performance loss. To our knowledge, this is the first unified framework that systematically quantifies and mitigates cross-backend drift in deep learning, providing a reproducible methodology for dependable deployment across heterogeneous runtimes.
Authors: Zirui Li, Bin Yang, Meng Wang
Abstract: Arrival time prediction (ATP) of public transport vehicles is essential in improving passenger experience and supporting traffic management. Deep learning has demonstrated outstanding performance in ATP due to its ability to model non-linear and temporal dynamics. In the multi-step ATP, non-stationary data will degrade the model performance due to the variation in variables' joint distribution along the temporal direction. Previous studies mainly applied normalization to eliminate the non-stationarity in time series, thereby achieving better predictability. However, the normalization may obscure useful characteristics inherent in non-stationarity, which is known as the over-stationarization. In this work, to trade off predictability and non-stationarity, a new approach for multi-step ATP, named non-stationary ATP ( NSATP), is proposed. The method consists of two stages: series stationarization and non-stationarity effect recovery. The first stage aims at improving the predictability. As for the latter, NSATP extends a state-of-the-art method from one-dimensional to two dimensional based models to capture the hidden periodicity in time series and designs a compensation module of over-stationarization by learning scaling and shifting factors from raw data. 125 days' public transport operational data of Dresden is collected for validation. Experimental results show that compared to baseline methods, the proposed NSATP can reduce RMSE, MAE, and MAPE by 2.37%, 1.22%, and 2.26% for trams and by 1.72%, 0.60%, and 1.17% for buses, respectively.
Authors: Jiajun Chai, Guojun Yin, Zekun Xu, Chuhuai Yue, Yi Jia, Siyu Xia, Xiaohan Wang, Jiwen Jiang, Xiaoguang Li, Chengqi Dong, Hang He, Wei Lin
Abstract: Large language models excel at basic reasoning but struggle with tasks that require interaction with external tools. We present RLFactory, a plug-and-play reinforcement learning post-training framework for multi-round tool use. RLFactory tackles (i) tool-call stability and adaptability amid tool heterogeneity and interface issues via an asyncio-based asynchronous caller and a decoupled tool/training architecture, and (ii) diverse evaluation needs via a reward layer supporting rule-based, model-judgment, and tool-verification signals. It reconstructs the MDP by introducing observation markers from tool feedback, closing the loop among model, tools, and environment, and implements a generate-parse-invoke-update workflow for dynamic policy optimization. On Search-R1 with Qwen3-4B, RLFactory achieves a 0.486 test score on the Natural Questions (NQ) dataset, surpassing larger models trained with similar techniques (e.g., Qwen2.5-7B-Instruct-GRPO at 0.473), and increases training throughput by 6.8x. RLFactory provides a low-barrier, highly adaptable framework for strengthening multi-round tool use of LLMs in real-world scenarios. Code: https://github.com/Simple-Efficient/RL-Factory.
Authors: Xiaomeng Hu, Fei Huang, Chenhan Yuan, Junyang Lin, Tsung-Yi Ho
Abstract: As large language models (LLMs) are increasingly deployed in real-world applications, ensuring the safety of their outputs during decoding has become a critical challenge. However, existing decoding-time interventions, such as Contrastive Decoding, often force a severe trade-off between safety and response quality. In this work, we propose CARE, a novel framework for decoding-time safety alignment that integrates three key components: (1) a guard model for real-time safety monitoring, enabling detection of potentially unsafe content; (2) a rollback mechanism with a token buffer to correct unsafe outputs efficiently at an earlier stage without disrupting the user experience; and (3) a novel introspection-based intervention strategy, where the model generates self-reflective critiques of its previous outputs and incorporates these reflections into the context to guide subsequent decoding steps. The framework achieves a superior safety-quality trade-off by using its guard model for precise interventions, its rollback mechanism for timely corrections, and our novel introspection method for effective self-correction. Experimental results demonstrate that our framework achieves a superior balance of safety, quality, and efficiency, attaining a low harmful response rate and minimal disruption to the user experience while maintaining high response quality.
Authors: Lishan Yang, Nam Kha Nguygen, Po Hu, Wei Emma Zhang, Yanjun Shu, Mong Yuan Sim, Weitong Chen
Abstract: Foundation models have demonstrated remarkable performance across a wide range of tasks, yet their large parameter sizes pose challenges for practical deployment, especially in decentralized environments. Parameter-efficient fine-tuning (PEFT), such as Low-Rank Adaptation (LoRA), reduces local computing and memory overhead, making it attractive for federated learning. However, existing federated LoRA methods typically assume uniform rank configurations and unimodal inputs, overlooking two key real-world challenges: (1) heterogeneous client resources have different LoRA ranks, and (2) multimodal data settings with potentially missing modalities. In this work, we propose FediLoRA, a simple yet effective framework for federated multimodal fine-tuning under heterogeneous LoRA ranks and missing modalities. FediLoRA introduces a dimension-wise aggregation strategy that reweights LoRA updates without information dilution during aggregation. It also includes a lightweight layer-wise model editing method that selectively incorporates global parameters to repair local components which improves both client and global model performances. Experimental results on three multimodal benchmark datasets demonstrate that FediLoRA achieves superior performance over competitive baselines in both global and personalized settings, particularly in the presence of modality incompleteness.
Authors: Cedric Caruzzo, Jong Chul Ye
Abstract: Large-scale biological discovery requires integrating massive, heterogeneous datasets like those from the JUMP Cell Painting consortium, but technical batch effects and a lack of generalizable models remain critical roadblocks. To address this, we introduce CellPainTR, a Transformer-based architecture designed to learn foundational representations of cellular morphology that are robust to batch effects. Unlike traditional methods that require retraining on new data, CellPainTR's design, featuring source-specific context tokens, allows for effective out-of-distribution (OOD) generalization to entirely unseen datasets without fine-tuning. We validate CellPainTR on the large-scale JUMP dataset, where it outperforms established methods like ComBat and Harmony in both batch integration and biological signal preservation. Critically, we demonstrate its robustness through a challenging OOD task on the unseen Bray et al. dataset, where it maintains high performance despite significant domain and feature shifts. Our work represents a significant step towards creating truly foundational models for image-based profiling, enabling more reliable and scalable cross-study biological analysis.
Authors: Alexey Zhukov (UB, CNRS, Bordeaux INP, Inria, LaBRI), Jenny Benois-Pineau (UB, CNRS, Bordeaux INP, Inria, LaBRI), Amira Youssef (SNCF R\'eseau), Akka Zemmari (UB, CNRS, Bordeaux INP, Inria, LaBRI), Mohamed Mosbah (UB, CNRS, Bordeaux INP, Inria, LaBRI), Virginie Taillandier
Abstract: Multimodal fusion is a multimedia technique that has become popular in the wide range of tasks where image information is accompanied by a signal/audio. The latter may not convey highly semantic information, such as speech or music, but some measures such as audio signal recorded by mics in the goal to detect rail structure elements or defects. While classical detection approaches such as You Only Look Once (YOLO) family detectors can be efficiently deployed for defect detection on the image modality, the single modality approaches remain limited. They yield an overdetection in case of the appearance similar to normal structural elements. The paper proposes a new multimodal fusion architecture built on the basis of domain rules with YOLO and Vision transformer backbones. It integrates YOLOv8n for rapid object detection with a Vision Transformer (ViT) to combine feature maps extracted from multiple layers (7, 16, and 19) and synthesised audio representations for two defect classes: rail Rupture and Surface defect. Fusion is performed between audio and image. Experimental evaluation on a real-world railway dataset demonstrates that our multimodal fusion improves precision and overall accuracy by 0.2 points compared to the vision-only approach. Student's unpaired t-test also confirms statistical significance of differences in the mean accuracy.
Authors: Yingsheng Wang, Shuo Lu, Jian Liang, Aihua Zheng, Ran He
Abstract: Out-of-distribution (OOD) detection helps models identify data outside the training categories, crucial for security applications. While feature-based post-hoc methods address this by evaluating data differences in the feature space without changing network parameters, they often require access to training data, which may not be suitable for some data privacy scenarios. This may not be suitable in scenarios where data privacy protection is a concern. In this paper, we propose a simple yet effective post-hoc method, termed Classifier-based Feature Reconstruction (ClaFR), from the perspective of subspace projection. It first performs an orthogonal decomposition of the classifier's weights to extract the class-known subspace, then maps the original data features into this subspace to obtain new data representations. Subsequently, the OOD score is determined by calculating the feature reconstruction error of the data within the subspace. Compared to existing OOD detection algorithms, our method does not require access to training data while achieving leading performance on multiple OOD benchmarks. Our code is released at https://github.com/Aie0923/ClaFR.
Authors: Jimmy Joseph
Abstract: In this paper, we introduce the Protocol Genome, a self-supervised learning system that learns correlations from DICOM headers and achieves AUROC 0.901 (vs 0.847 baseline) and ECE 0.036 (vs 0.058) on fully held-out external validation. Our method also improves calibration and robustness across modalities (CT, MRI, CXR) and vendors. Clinical imaging is funneled through PACS/DICOM, where procedure choices (scanner make/model, sequence, kernel, kVp, TR/TE, and slice thickness) have consequences for contrast, noise, and artifact. These latent confounders impede the generalization of image-only networks across sites. We consider structured DICOM headers as a label and learn protocol-aware but clinically robust image representations. Protocol Genome obtains tokenized embeddings of de-identified header fields and models them along with image features using: (1) protocol-image contrastive learning, (2) masked protocol prediction, and (3) protocol-protocol translation. With 1.26M studies (7 health systems, 31 scanners, 3 vendors; CT, MR, CR/DR), we experiment on: (A) chest CT triage for PE, (B) brain MRI glioma grading, and (C) chest radiograph cardiomegaly detection. Relative to strong SSL baselines (SimCLR, MAE) as well as ImageNet transfer, Protocol Genome (+0.046: PE, +0.058: glioma, +0.041: cardiomegaly) is associated with higher external AUROC; 25-37% calibration improvements are obtained (p < 0.01, DeLong tests). While the gains may be task-dependent, they are preserved with 10-20% of labeled data. From a clinical point of view, the technique reduces false positives at protocol borders and is applicable in a PACS (DICOM C-FIND/C-MOVE, DICOMweb QIDO/WADO). We publish a model card and deployment guide, complete with both de-identification and bias audits.
Authors: Jie Zhang, Ting Xu, Gelei Deng, Runyi Hu, Han Qiu, Tianwei Zhang, Qing Guo, Ivor Tsang
Abstract: Writing is a universal cultural technology that reuses vision for symbolic communication. Humans display striking resilience: we readily recognize words even when characters are fragmented, fused, or partially occluded. This paper investigates whether advanced vision language models (VLMs) share this resilience. We construct two psychophysics inspired benchmarks across distinct writing systems, Chinese logographs and English alphabetic words, by splicing, recombining, and overlaying glyphs to yield ''visible but unreadable'' stimuli for models while remaining legible to humans. Despite strong performance on clean text, contemporary VLMs show a severe drop under these perturbations, frequently producing unrelated or incoherent outputs. The pattern suggests a structural limitation: models heavily leverage generic visual invariances but under rely on compositional priors needed for robust literacy. We release stimuli generation code, prompts, and evaluation protocols to facilitate transparent replication and follow up work. Our findings motivate architectures and training strategies that encode symbol segmentation, composition, and binding across scripts, and they delineate concrete challenges for deploying multimodal systems in education, accessibility, cultural heritage, and security.
Authors: Liviu Nicolae Firc\u{a}, Antonio B\u{a}rb\u{a}lau, Dan Oneata, Elena Burceanu
Abstract: Can models generalize attribute knowledge across semantically and perceptually dissimilar categories? While prior work has addressed attribute prediction within narrow taxonomic or visually similar domains, it remains unclear whether current models can abstract attributes and apply them to conceptually distant categories. This work presents the first explicit evaluation for the robustness of the attribute prediction task under such conditions, testing whether models can correctly infer shared attributes between unrelated object types: e.g., identifying that the attribute "has four legs" is common to both "dogs" and "chairs". To enable this evaluation, we introduce train-test split strategies that progressively reduce correlation between training and test sets, based on: LLM-driven semantic grouping, embedding similarity thresholding, embedding-based clustering, and supercategory-based partitioning using ground-truth labels. Results show a sharp drop in performance as the correlation between training and test categories decreases, indicating strong sensitivity to split design. Among the evaluated methods, clustering yields the most effective trade-off, reducing hidden correlations while preserving learnability. These findings offer new insights into the limitations of current representations and inform future benchmark construction for attribute reasoning.
Authors: Kapil Madan
Abstract: This paper introduces ArGen (Auto-Regulation of Generative AI systems), a framework for aligning Large Language Models (LLMs) with complex sets of configurable, machine-readable rules spanning ethical principles, operational safety protocols, and regulatory compliance standards. Moving beyond just preference-based alignment, ArGen is designed to ensure LLMs adhere to these multifaceted policies through a novel synthesis of principle-based automated reward scoring, Group Relative Policy Optimisation (GRPO), and an Open Policy Agent (OPA) inspired governance layer. This approach provides the technical foundation for achieving and demonstrating compliance with diverse and nuanced governance requirements. To showcase the framework's capability to operationalize a deeply nuanced and culturally-specific value system, we present an in-depth case study: the development of a medical AI assistant guided by principles from Dharmic ethics (such as Ahimsa and Dharma), as derived from texts like the Bhagavad Gita. This challenging application demonstrates ArGen's adaptability, achieving a 70.9% improvement in domain-scope adherence over the baseline. Through our open-source repository, we show that ArGen's methodology offers a path to 'Governable Al' systems that are technically proficient, ethically robust, and verifiably compliant for safe deployment in diverse global contexts.
Authors: Anton Kolonin, Vladimir Kryukov
Abstract: The article provides an overview of approaches to modeling the human psyche in the perspective of building an artificial one. Based on the review, a concept of cognitive architecture is proposed, where the psyche is considered as an operating system of a living or artificial subject, including a space of needs that determines its life meanings in connection with stimuli from the external world, and intelligence as a decision-making system for actions in relation to this world in order to satisfy these needs. Based on the concept, a computational formalization is proposed for creating artificial intelligence systems through learning from experience in the space of a space of needs, taking into account their biological or existential significance for an intelligent agent. Thus, the problem of building general artificial intelligence as a system for making optimal decisions in the space of agent-specific needs under conditions of uncertainty is formalized, with maximization of success in achieving goals, minimization of existential risks and maximization of energy efficiency. A minimal experimental implementation of the model is also provided.
Authors: Ahmed R. Sadik, Mariusz Bujny
Abstract: Large Language Models are increasingly capable of interpreting multimodal inputs to generate complex 3D shapes, yet robust methods to evaluate geometric and structural fidelity remain underdeveloped. This paper introduces a human in the loop framework for the quantitative evaluation of LLM generated 3D models, supporting applications such as democratization of CAD design, reverse engineering of legacy designs, and rapid prototyping. We propose a comprehensive suite of similarity and complexity metrics, including volumetric accuracy, surface alignment, dimensional fidelity, and topological intricacy, to benchmark generated models against ground truth CAD references. Using an L bracket component as a case study, we systematically compare LLM performance across four input modalities: 2D orthographic views, isometric sketches, geometric structure trees, and code based correction prompts. Our findings demonstrate improved generation fidelity with increased semantic richness, with code level prompts achieving perfect reconstruction across all metrics. A key contribution of this work is demonstrating that our proposed quantitative evaluation approach enables significantly faster convergence toward the ground truth, especially compared to traditional qualitative methods based solely on visual inspection and human intuition. This work not only advances the understanding of AI assisted shape synthesis but also provides a scalable methodology to validate and refine generative models for diverse CAD applications.
Authors: Muhammad Arif Hakimi Zamrai, Kamaludin Mohd Yusof
Abstract: In response to the prevalent concern of TCP SYN flood attacks within the context of Software-Defined Internet of Vehicles (SD-IoV), this study addresses the significant challenge of network security in rapidly evolving vehicular communication systems. This research focuses on optimizing a Random Forest Classifier model to achieve maximum accuracy and minimal detection time, thereby enhancing vehicular network security. The methodology involves preprocessing a dataset containing SYN attack instances, employing feature scaling and label encoding techniques, and applying Stratified K-Fold cross-validation to target key metrics such as accuracy, precision, recall, and F1-score. This research achieved an average value of 0.999998 for all metrics with a SYN DoS attack detection time of 0.24 seconds. Results show that the fine-tuned Random Forest model, configured with 20 estimators and a depth of 10, effectively differentiates between normal and malicious traffic with high accuracy and minimal detection time, which is crucial for SD-IoV networks. This approach marks a significant advancement and introduces a state-of-the-art algorithm in detecting SYN flood attacks, combining high accuracy with minimal detection time. It contributes to vehicular network security by providing a robust solution against TCP SYN flood attacks while maintaining network efficiency and reliability.
Authors: Xinquan Wu, Xuefeng Yan, Mingqiang Wei, Donghai Guan
Abstract: The Flexible Job-shop Scheduling Problem (FJSP) is a classical combinatorial optimization problem that has a wide-range of applications in the real world. In order to generate fast and accurate scheduling solutions for FJSP, various deep reinforcement learning (DRL) scheduling methods have been developed. However, these methods are mainly focused on the design of DRL scheduling Agent, overlooking the modeling of DRL environment. This paper presents a simple chronological DRL environment for FJSP based on discrete event simulation and an end-to-end DRL scheduling model is proposed based on the proximal policy optimization (PPO). Furthermore, a short novel state representation of FJSP is proposed based on two state variables in the scheduling environment and a novel comprehensible reward function is designed based on the scheduling area of machines. Experimental results on public benchmark instances show that the performance of simple priority dispatching rules (PDR) is improved in our scheduling environment and our DRL scheduling model obtains competing performance compared with OR-Tools, meta-heuristic, DRL and PDR scheduling methods.
Authors: Jiarui Chen, Yikeng Chen, Yingshuang Zou, Ye Huang, Peng Wang, Yuan Liu, Yujing Sun, Wenping Wang
Abstract: 3D Gaussian Splatting (3DGS) has emerged as a dominant novel-view synthesis technique, but its high memory consumption severely limits its applicability on edge devices. A growing number of 3DGS compression methods have been proposed to make 3DGS more efficient, yet most only focus on storage compression and fail to address the critical bottleneck of rendering memory. To address this problem, we introduce MEGS$^{2}$, a novel memory-efficient framework that tackles this challenge by jointly optimizing two key factors: the total primitive number and the parameters per primitive, achieving unprecedented memory compression. Specifically, we replace the memory-intensive spherical harmonics with lightweight arbitrarily-oriented spherical Gaussian lobes as our color representations. More importantly, we propose a unified soft pruning framework that models primitive-number and lobe-number pruning as a single constrained optimization problem. Experiments show that MEGS$^{2}$ achieves a 50% static VRAM reduction and a 40% rendering VRAM reduction compared to existing methods, while maintaining comparable rendering quality.
Authors: Pavan Reddy, Nithin Reddy
Abstract: In 2023, the National Eating Disorders Association's (NEDA) chatbot Tessa was suspended after providing harmful weight-loss advice to vulnerable users-an avoidable failure that underscores the risks of unsafe AI in healthcare contexts. This paper examines Tessa as a case study in absent safety engineering and demonstrates how a lightweight, modular safeguard could have prevented the incident. We propose a hybrid safety middleware that combines deterministic lexical gates with an in-line large language model (LLM) policy filter, enforcing fail-closed verdicts and escalation pathways within a single model call. Using synthetic evaluations, we show that this design achieves perfect interception of unsafe prompts at baseline cost and latency, outperforming traditional multi-stage pipelines. Beyond technical remedies, we map Tessa's failure patterns to established frameworks (OWASP LLM Top10, NIST SP 800-53), connecting practical safeguards to actionable governance controls. The results highlight that robust, auditable safety in health-adjacent AI does not require heavyweight infrastructure: explicit, testable checks at the last mile are sufficient to prevent "another Tessa", while governance and escalation ensure sustainability in real-world deployment.
Authors: Eduardo Lobo Lustoda Cabral, Paulo Pirozelli, Larissa Driemeier
Abstract: The increasing size of large neural network models, specifically language models and foundational image models, poses deployment challenges, prompting efforts to reduce memory requirements and enhance computational efficiency. These efforts are critical to ensure practical deployment and effective utilization of these models across various applications. In this work, a novel type of neural network layers and models is developed that uses only single-bit parameters. In this novel type of models all parameters of all layers, including kernel weights and biases, only have values equal to zero or one. This novel type of models uses layers named as binary normalized layer. These binary normalized layers can be of any type, such as fully connected, convolutional, attention, etc., and they consist of slight variations of the corresponding conventional layers. To show the effectiveness of the binary normalized layers, two different models are configured to solve a multiclass image classification problem and a language decoder to predict the next token of a sequence. The model to solve the image classification has convolutional and fully connected layers, and the language model is composed of transformer blocks with multi-head attention. The results show that models with binary normalized layers present almost the same results obtained by equivalent models with real 32-bit parameters. The binary normalized layers allow to develop models that use 32 times less memory than current models and have equivalent performance. Besides, the binary normalized layers can be easily implemented on current computers using 1-bit arrays, and do not require the development of dedicated electronic hardware. This novel type of layers opens a new era for large neural network models with reduced memory requirements that can be deployed using simple and cheap hardware, such as mobile devices or only cpus.
Authors: Yang Xu, Shuwei Chen, Xiaomei Zhong, Jun Liu, Xingxing He
Abstract: Trustworthy AI requires reasoning systems that are not only powerful but also transparent and reliable. Automated Theorem Proving (ATP) is central to formal reasoning, yet classical binary resolution remains limited, as each step involves only two clauses and eliminates at most two literals. To overcome this bottleneck, the concept of standard contradiction and the theory of contradiction-separation-based deduction were introduced in 2018. This paper advances that framework by focusing on the systematic construction of standard contradictions. Specially, this study investigates construction methods for two principal forms of standard contradiction: the maximum triangular standard contradiction and the triangular-type standard contradiction. Building on these structures, we propose a procedure for determining the satisfiability and unsatisfiability of clause sets via maximum standard contradiction. Furthermore, we derive formulas for computing the number of standard sub-contradictions embedded within both the maximum triangular standard contradiction and the triangular-type standard contradiction. The results presented herein furnish the methodological basis for advancing contradiction-separation-based dynamic multi-clause automated deduction, thereby extending the expressive and deductive capabilities of automated reasoning systems beyond the classical binary paradigm.
Authors: Jisung Hwang, Jaihoon Kim, Minhyuk Sung
Abstract: We propose a novel regularization loss that enforces standard Gaussianity, encouraging samples to align with a standard Gaussian distribution. This facilitates a range of downstream tasks involving optimization in the latent space of text-to-image models. We treat elements of a high-dimensional sample as one-dimensional standard Gaussian variables and define a composite loss that combines moment-based regularization in the spatial domain with power spectrum-based regularization in the spectral domain. Since the expected values of moments and power spectrum distributions are analytically known, the loss promotes conformity to these properties. To ensure permutation invariance, the losses are applied to randomly permuted inputs. Notably, existing Gaussianity-based regularizations fall within our unified framework: some correspond to moment losses of specific orders, while the previous covariance-matching loss is equivalent to our spectral loss but incurs higher time complexity due to its spatial-domain computation. We showcase the application of our regularization in generative modeling for test-time reward alignment with a text-to-image model, specifically to enhance aesthetics and text alignment. Our regularization outperforms previous Gaussianity regularization, effectively prevents reward hacking and accelerates convergence.
Authors: Viswa Chaitanya Marella, Sai Teja Erukude, Suhasnadh Reddy Veluru
Abstract: The introduction of Artificial Intelligence (AI) into the domains of traditional art (visual arts, performing arts, and crafts) has sparked a complicated discussion about whether this might be an agent of disruption or an enhancement of our traditional art forms. This paper looks at the duality of AI, exploring the ways that recent technologies like Generative Adversarial Networks and Diffusion Models, and text-to-image generators are changing the fields of painting, sculpture, calligraphy, dance, music, and the arts of craft. Using examples and data, we illustrate the ways that AI can democratize creative expression, improve productivity, and preserve cultural heritage, while also examining the negative aspects, including: the threats to authenticity within art, ethical concerns around data, and issues including socio-economic factors such as job losses. While we argue for the context-dependence of the impact of AI (the potential for creative homogenization and the devaluation of human agency in artmaking), we also illustrate the potential for hybrid practices featuring AI in cuisine, etc. We advocate for the development of ethical guidelines, collaborative approaches, and inclusive technology development. In sum, we are articulating a vision of AI in which it amplifies our innate creativity while resisting the displacement of the cultural, nuanced, and emotional aspects of traditional art. The future will be determined by human choices about how to govern AI so that it becomes a mechanism for artistic evolution and not a substitute for the artist's soul.
Authors: Kaizheng Wang
Abstract: The Bayesian paradigm offers principled tools for sequential decision-making under uncertainty, but its reliance on a probabilistic model for all parameters can hinder the incorporation of complex structural constraints. We introduce a minimalist Bayesian framework that places a prior only on the component of interest, such as the location of the optimum. Nuisance parameters are eliminated via profile likelihood, which naturally handles constraints. As a direct instantiation, we develop a MINimalist Thompson Sampling (MINTS) algorithm. Our framework accommodates structured problems, including continuum-armed Lipschitz bandits and dynamic pricing. It also provides a probabilistic lens on classical convex optimization algorithms such as the center of gravity and ellipsoid methods. We further analyze MINTS for multi-armed bandits and establish near-optimal regret guarantees.
Authors: Madara Ogot
Abstract: The rapid proliferation of artificial intelligence (AI) across industry, government, and education highlights the urgent need for robust frameworks to conceptualise and guide engagement. This paper introduces the Hierarchy of Engagement with AI model, a novel maturity framework inspired by Maslow's hierarchy of needs. The model conceptualises AI adoption as a progression through eight levels, beginning with initial exposure and basic understanding and culminating in ecosystem collaboration and societal impact. Each level integrates technical, organisational, and ethical dimensions, emphasising that AI maturity is not only a matter of infrastructure and capability but also of trust, governance, and responsibility. Initial validation of the model using four diverse case studies (General Motors, the Government of Estonia, the University of Texas System, and the African Union AI Strategy) demonstrate the model's contextual flexibility across various sectors. The model provides scholars with a framework for analysing AI maturity and offers practitioners and policymakers a diagnostic and strategic planning tool to guide responsible and sustainable AI engagement. The proposed model demonstrates that AI maturity progression is multi-dimensional, requiring technological capability, ethical integrity, organisational resilience, and ecosystem collaboration.
Authors: Federico Cerutti
Abstract: This paper presents a methodological approach to financial time series analysis by combining causal discovery and uncertainty-aware forecasting. As a case study, we focus on four key U.S. macroeconomic indicators -- GDP, economic growth, inflation, and unemployment -- and we apply the LPCMCI framework with Gaussian Process Distance Correlation (GPDC) to uncover dynamic causal relationships in quarterly data from 1970 to 2021. Our results reveal a robust unidirectional causal link from economic growth to GDP and highlight the limited connectivity of inflation, suggesting the influence of latent factors. Unemployment exhibits strong autoregressive dependence, motivating its use as a case study for probabilistic forecasting. Leveraging the Chronos framework, a large language model trained for time series, we perform zero-shot predictions on unemployment. This approach delivers accurate forecasts one and two quarters ahead, without requiring task-specific training. Crucially, the model's uncertainty-aware predictions yield 90\% confidence intervals, enabling effective anomaly detection through statistically principled deviation analysis. This study demonstrates the value of combining causal structure learning with probabilistic language models to inform economic policy and enhance forecasting robustness.
Authors: Yerin Ryu, Inseop Shin, Chanwoo Kim
Abstract: Controllable Singing Voice Synthesis (SVS) aims to generate expressive singing voices reflecting user intent. While recent SVS systems achieve high audio quality, most rely on probabilistic modeling, limiting precise control over attributes such as dynamics. We address this by focusing on dynamic control--temporal loudness variation essential for musical expressiveness--and explicitly condition the SVS model on energy sequences extracted from ground-truth spectrograms, reducing annotation costs and improving controllability. We also propose a phoneme-level energy sequence for user-friendly control. To the best of our knowledge, this is the first attempt enabling user-driven dynamics control in SVS. Experiments show our method achieves over 50% reduction in mean absolute error of energy sequences for phoneme-level inputs compared to baseline and energy-predictor models, without compromising synthesis quality.
Authors: Juan Manuel Contreras
Abstract: Large multimodal models (LMMs) have revolutionized text-to-image generation, but they risk perpetuating the harmful social biases in their training data. Prior work has identified gender bias in these models, but methodological limitations prevented large-scale, comparable, cross-model analysis. To address this gap, we introduce the Aymara Image Fairness Evaluation, a benchmark for assessing social bias in AI-generated images. We test 13 commercially available LMMs using 75 procedurally-generated, gender-neutral prompts to generate people in stereotypically-male, stereotypically-female, and non-stereotypical professions. We then use a validated LLM-as-a-judge system to score the 965 resulting images for gender representation. Our results reveal (p < .001 for all): 1) LMMs systematically not only reproduce but actually amplify occupational gender stereotypes relative to real-world labor data, generating men in 93.0% of images for male-stereotyped professions but only 22.5% for female-stereotyped professions; 2) Models exhibit a strong default-male bias, generating men in 68.3% of the time for non-stereotyped professions; and 3) The extent of bias varies dramatically across models, with overall male representation ranging from 46.7% to 73.3%. Notably, the top-performing model de-amplified gender stereotypes and approached gender parity, achieving the highest fairness scores. This variation suggests high bias is not an inevitable outcome but a consequence of design choices. Our work provides the most comprehensive cross-model benchmark of gender bias to date and underscores the necessity of standardized, automated evaluation tools for promoting accountability and fairness in AI development.
Authors: Sergey Pozdnyakov, Philippe Schwaller
Abstract: High-dimensional linear mappings, or linear layers, dominate both the parameter count and the computational cost of most modern deep-learning models. We introduce a general drop-in replacement, lookup multivariate Kolmogorov-Arnold Networks (lmKANs), which deliver a substantially better trade-off between capacity and inference cost. Our construction expresses a general high-dimensional mapping through trainable low-dimensional multivariate functions. These functions can carry dozens or hundreds of trainable parameters each, and yet it takes only a few multiplications to compute them because they are implemented as spline lookup tables. Empirically, lmKANs reduce inference FLOPs by up to 6.0x while matching the flexibility of MLPs in general high-dimensional function approximation. In another feedforward fully connected benchmark, on the tabular-like dataset of randomly displaced methane configurations, lmKANs enable more than 10x higher H100 throughput at equal accuracy. Within frameworks of Convolutional Neural Networks, lmKAN-based CNNs cut inference FLOPs at matched accuracy by 1.6-2.1x and by 1.7x on the CIFAR-10 and ImageNet-1k datasets, respectively. Our code, including dedicated CUDA kernels, is available online at https://github.com/schwallergroup/lmkan.
Authors: Ziheng Chen, Xiao-Jun Wu, Nicu Sebe
Abstract: Normalization layers are crucial for deep learning, but their Euclidean formulations are inadequate for data on manifolds. On the other hand, many Riemannian manifolds in machine learning admit gyro-structures, enabling principled extensions of Euclidean neural networks to non-Euclidean domains. Inspired by this, we introduce GyroBN, a principled Riemannian batch normalization framework for gyrogroups. We establish two necessary conditions, namely \emph{pseudo-reduction} and \emph{gyroisometric gyrations}, that guarantee GyroBN with theoretical control over sample statistics, and show that these conditions hold for all known gyrogroups in machine learning. Our framework also incorporates several existing Riemannian normalization methods as special cases. We further instantiate GyroBN on seven representative geometries, including the Grassmannian, five constant curvature spaces, and the correlation manifold, and derive novel gyro and Riemannian structures to enable these instantiations. Experiments across these geometries demonstrate the effectiveness of GyroBN. The code is available at https://github.com/GitZH-Chen/GyroBN.git.
Authors: Leonardo Zini, Elia Frigieri, Sebastiano Aloscari, Marcello Generali, Lorenzo Dodi, Robert Dosen, Lorenzo Baraldi
Abstract: Generated Scalable Vector Graphics (SVG) images demand evaluation criteria tuned to their symbolic and vectorial nature: criteria that existing metrics such as FID, LPIPS, or CLIPScore fail to satisfy. In this paper, we introduce SVGauge, the first human-aligned, reference based metric for text-to-SVG generation. SVGauge jointly measures (i) visual fidelity, obtained by extracting SigLIP image embeddings and refining them with PCA and whitening for domain alignment, and (ii) semantic consistency, captured by comparing BLIP-2-generated captions of the SVGs against the original prompts in the combined space of SBERT and TF-IDF. Evaluation on the proposed SHE benchmark shows that SVGauge attains the highest correlation with human judgments and reproduces system-level rankings of eight zero-shot LLM-based generators more faithfully than existing metrics. Our results highlight the necessity of vector-specific evaluation and provide a practical tool for benchmarking future text-to-SVG generation models.
Authors: Nicol\`o Romandini, Carlo Mazzocca, Kai Otsuki, Rebecca Montanari
Abstract: Blockchain and smart contracts have garnered significant interest in recent years as the foundation of a decentralized, trustless digital ecosystem, thereby eliminating the need for traditional centralized authorities. Despite their central role in powering Web3, their complexity still presents significant barriers for non-expert users. To bridge this gap, Artificial Intelligence (AI)-based agents have emerged as valuable tools for interacting with blockchain environments, supporting a range of tasks, from analyzing on-chain data and optimizing transaction strategies to detecting vulnerabilities within smart contracts. While interest in applying AI to blockchain is growing, the literature still lacks a comprehensive survey that focuses specifically on the intersection with AI agents. Most of the related work only provides general considerations, without focusing on any specific domain. This paper addresses this gap by presenting the first Systematization of Knowledge dedicated to AI-driven systems for blockchain, with a special focus on their security and privacy dimensions, shedding light on their applications, limitations, and future research directions.
Authors: Kutub Uddin, Muhammad Umar Farooq, Awais Khan, Khalid Mahmood Malik
Abstract: The widespread use of generative AI has shown remarkable success in producing highly realistic deepfakes, posing a serious threat to various voice biometric applications, including speaker verification, voice biometrics, audio conferencing, and criminal investigations. To counteract this, several state-of-the-art (SoTA) audio deepfake detection (ADD) methods have been proposed to identify generative AI signatures to distinguish between real and deepfake audio. However, the effectiveness of these methods is severely undermined by anti-forensic (AF) attacks that conceal generative signatures. These AF attacks span a wide range of techniques, including statistical modifications (e.g., pitch shifting, filtering, noise addition, and quantization) and optimization-based attacks (e.g., FGSM, PGD, C \& W, and DeepFool). In this paper, we investigate the SoTA ADD methods and provide a comparative analysis to highlight their effectiveness in exposing deepfake signatures, as well as their vulnerabilities under adversarial conditions. We conducted an extensive evaluation of ADD methods on five deepfake benchmark datasets using two categories: raw and spectrogram-based approaches. This comparative analysis enables a deeper understanding of the strengths and limitations of SoTA ADD methods against diverse AF attacks. It does not only highlight vulnerabilities of ADD methods, but also informs the design of more robust and generalized detectors for real-world voice biometrics. It will further guide future research in developing adaptive defense strategies that can effectively counter evolving AF techniques.
Authors: Zhiyin Tan, Jennifer D'Souza
Abstract: This study presents a framework for automated evaluation of dynamically evolving topic models using Large Language Models (LLMs). Topic modeling is essential for organizing and retrieving scholarly content in digital library systems, helping users navigate complex and evolving knowledge domains. However, widely used automated metrics, such as coherence and diversity, often capture only narrow statistical patterns and fail to explain semantic failures in practice. We introduce a purpose-oriented evaluation framework that employs nine LLM-based metrics spanning four key dimensions of topic quality: lexical validity, intra-topic semantic soundness, inter-topic structural soundness, and document-topic alignment soundness. The framework is validated through adversarial and sampling-based protocols, and is applied across datasets spanning news articles, scholarly publications, and social media posts, as well as multiple topic modeling methods and open-source LLMs. Our analysis shows that LLM-based metrics provide interpretable, robust, and task-relevant assessments, uncovering critical weaknesses in topic models such as redundancy and semantic drift, which are often missed by traditional metrics. These results support the development of scalable, fine-grained evaluation tools for maintaining topic relevance in dynamic datasets. All code and data supporting this work are accessible at https://github.com/zhiyintan/topic-model-LLMjudgment.
Authors: Anatoly A. Krasnovsky
Abstract: Mechanistic interpretability has identified functional subgraphs within large language models (LLMs), known as Transformer Circuits (TCs), that appear to implement specific algorithms. Yet we lack a formal, single-pass way to quantify when an active circuit is behaving coherently and thus likely trustworthy. Building on prior systems-theoretic proposals, we specialize a sheaf/cohomology and causal emergence perspective to TCs and introduce the Effective-Information Consistency Score (EICS). EICS combines (i) a normalized sheaf inconsistency computed from local Jacobians and activations, with (ii) a Gaussian EI proxy for circuit-level causal emergence derived from the same forward state. The construction is white-box, single-pass, and makes units explicit so that the score is dimensionless. We further provide practical guidance on score interpretation, computational overhead (with fast and exact modes), and a toy sanity-check analysis. Empirical validation on LLM tasks is deferred.
Authors: Zonghai Yao, Michael Sun, Won Seok Jang, Sunjae Kwon, Soie Kwon, Hong Yu
Abstract: Discharge communication is a critical yet underexplored component of patient care, where the goal shifts from diagnosis to education. While recent large language model (LLM) benchmarks emphasize in-visit diagnostic reasoning, they fail to evaluate models' ability to support patients after the visit. We introduce DischargeSim, a novel benchmark that evaluates LLMs on their ability to act as personalized discharge educators. DischargeSim simulates post-visit, multi-turn conversations between LLM-driven DoctorAgents and PatientAgents with diverse psychosocial profiles (e.g., health literacy, education, emotion). Interactions are structured across six clinically grounded discharge topics and assessed along three axes: (1) dialogue quality via automatic and LLM-as-judge evaluation, (2) personalized document generation including free-text summaries and structured AHRQ checklists, and (3) patient comprehension through a downstream multiple-choice exam. Experiments across 18 LLMs reveal significant gaps in discharge education capability, with performance varying widely across patient profiles. Notably, model size does not always yield better education outcomes, highlighting trade-offs in strategy use and content prioritization. DischargeSim offers a first step toward benchmarking LLMs in post-visit clinical education and promoting equitable, personalized patient support.
Authors: Jonathan I. Mandel, Shivaprakash Hiremath, Hedyeh Keshtgar, Timothy Scholl, Sadegh Raeisi
Abstract: This retrospective-prospective study evaluated whether a deep learning-based MRI reconstruction algorithm can preserve diagnostic quality in brain MRI scans accelerated up to fourfold, using both public and prospective clinical data. The study included 18 healthy volunteers (scans acquired at 3T, January 2024-March 2025), as well as selected fastMRI public datasets with diverse pathologies. Phase-encoding-undersampled 2D/3D T1, T2, and FLAIR sequences were reconstructed with DeepFoqus-Accelerate and compared with standard-of-care (SOC). Three board-certified neuroradiologists and two MRI technologists independently reviewed 36 paired SOC/AI reconstructions from both datasets using a 5-point Likert scale, while quantitative similarity was assessed for 408 scans and 1224 datasets using Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR), and Haar wavelet-based Perceptual Similarity Index (HaarPSI). No AI-reconstructed scan scored below 3 (minimally acceptable), and 95% scored $\geq 4$. Mean SSIM was 0.95 $\pm$ 0.03 (90% cases >0.90), PSNR >41.0 dB, and HaarPSI >0.94. Inter-rater agreement was slight to moderate. Rare artifacts did not affect diagnostic interpretation. These findings demonstrate that DeepFoqus-Accelerate enables robust fourfold brain MRI acceleration with 75% reduced scan time, while preserving diagnostic image quality and supporting improved workflow efficiency.
Authors: Qi Diao, Chengyue Xie, Yuchen Yin, Hoileong Lee, Haolong Yang
Abstract: Aiming at the shortcomings of the gazelle optimization algorithm, such as the imbalance between exploration and exploitation and the insufficient information exchange within the population, this paper proposes a multi-strategy improved gazelle optimization algorithm (MSIGOA). To address these issues, MSIGOA proposes an iteration-based updating framework that switches between exploitation and exploration according to the optimization process, which effectively enhances the balance between local exploitation and global exploration in the optimization process and improves the convergence speed. Two adaptive parameter tuning strategies improve the applicability of the algorithm and promote a smoother optimization process. The dominant population-based restart strategy enhances the algorithms ability to escape from local optima and avoid its premature convergence. These enhancements significantly improve the exploration and exploitation capabilities of MSIGOA, bringing superior convergence and efficiency in dealing with complex problems. In this paper, the parameter sensitivity, strategy effectiveness, convergence and stability of the proposed method are evaluated on two benchmark test sets including CEC2017 and CEC2022. Test results and statistical tests show that MSIGOA outperforms basic GOA and other advanced algorithms. On the CEC2017 and CEC2022 test sets, the proportion of functions where MSIGOA is not worse than GOA is 92.2% and 83.3%, respectively, and the proportion of functions where MSIGOA is not worse than other algorithms is 88.57% and 87.5%, respectively. Finally, the extensibility of MSIGAO is further verified by several engineering design optimization problems.
Authors: Raja Mallina, Bryar Shareef
Abstract: Background: Precise breast ultrasound (BUS) segmentation supports reliable measurement, quantitative analysis, and downstream classification, yet remains difficult for small or low-contrast lesions with fuzzy margins and speckle noise. Text prompts can add clinical context, but directly applying weakly localized text-image cues (e.g., CAM/CLIP-derived signals) tends to produce coarse, blob-like responses that smear boundaries unless additional mechanisms recover fine edges. Methods: We propose XBusNet, a novel dual-prompt, dual-branch multimodal model that combines image features with clinically grounded text. A global pathway based on a CLIP Vision Transformer encodes whole-image semantics conditioned on lesion size and location, while a local U-Net pathway emphasizes precise boundaries and is modulated by prompts that describe shape, margin, and Breast Imaging Reporting and Data System (BI-RADS) terms. Prompts are assembled automatically from structured metadata, requiring no manual clicks. We evaluate on the Breast Lesions USG (BLU) dataset using five-fold cross-validation. Primary metrics are Dice and Intersection over Union (IoU); we also conduct size-stratified analyses and ablations to assess the roles of the global and local paths and the text-driven modulation. Results: XBusNet achieves state-of-the-art performance on BLU, with mean Dice of 0.8765 and IoU of 0.8149, outperforming six strong baselines. Small lesions show the largest gains, with fewer missed regions and fewer spurious activations. Ablation studies show complementary contributions of global context, local boundary modeling, and prompt-based modulation. Conclusions: A dual-prompt, dual-branch multimodal design that merges global semantics with local precision yields accurate BUS segmentation masks and improves robustness for small, low-contrast lesions.
Authors: Abhimanyu Bellam, Jung-Eun Kim
Abstract: Post Training Quantization (PTQ) is widely adopted due to its high compression capacity and speed with minimal impact on accuracy. However, we observed that disparate impacts are exacerbated by quantization, especially for minority groups. Our analysis explains that in the course of quantization there is a chain of factors attributed to a disparate impact across groups during forward and backward passes. We explore how the changes in weights and activations induced by quantization cause cascaded impacts in the network, resulting in logits with lower variance, increased loss, and compromised group accuracies. We extend our study to verify the influence of these impacts on group gradient norms and eigenvalues of the Hessian matrix, providing insights into the state of the network from an optimization point of view. To mitigate these effects, we propose integrating mixed precision Quantization Aware Training (QAT) with dataset sampling methods and weighted loss functions, therefore providing fair deployment of quantized neural networks.
Authors: Yann Alibert, Jeanne Davoult, Sara Marques
Abstract: Numerical calculations of planetary system formation are very demanding in terms of computing power. These synthetic planetary systems can however provide access to correlations, as predicted in a given numerical framework, between the properties of planets in the same system. Such correlations can, in return, be used in order to guide and prioritize observational campaigns aiming at discovering some types of planets, as Earth-like planets. Our goal is to develop a generative model which is capable of capturing correlations and statistical relationships between planets in the same system. Such a model, trained on the Bern model, offers the possibility to generate large number of synthetic planetary systems with little computational cost, that can be used, for example, to guide observational campaigns. Our generative model is based on the transformer architecture which is well-known to efficiently capture correlations in sequences and is at the basis of all modern Large Language Models. To assess the validity of the generative model, we perform visual and statistical comparisons, as well as a machine learning driven tests. Finally, as a use case example, we consider the TOI-469 system, in which we aim at predicting the possible properties of planets c and d, based on the properties of planet b (the first that has been detected). We show using different comparison methods that the properties of systems generated by our model are very similar to the ones of the systems computed directly by the Bern model. We also show in the case of the TOI-469 system, that using the generative model allows to predict the properties of planets not yet observed, based on the properties of the already observed planet. We provide our model to the community on our website www.ai4exoplanets.com.
Authors: Luigi Troiano, Francesco Gissi, Vincenzo Benedetto, Genny Tortora
Abstract: Gradient-based neural network training traditionally enforces symmetry between forward and backward propagation, requiring activation functions to be differentiable (or sub-differentiable) and strictly monotonic in certain regions to prevent flat gradient areas. This symmetry, linking forward activations closely to backward gradients, significantly restricts the selection of activation functions, particularly excluding those with substantial flat or non-differentiable regions. In this paper, we challenge this assumption through mathematical analysis, demonstrating that precise gradient magnitudes derived from activation functions are largely redundant, provided the gradient direction is preserved. Empirical experiments conducted on foundational architectures - such as Multi-Layer Perceptrons (MLPs), Convolutional Neural Networks (CNNs), and Binary Neural Networks (BNNs) - confirm that relaxing forward-backward symmetry and substituting traditional gradients with simpler or stochastic alternatives does not impair learning and may even enhance training stability and efficiency. We explicitly demonstrate that neural networks with flat or non-differentiable activation functions, such as the Heaviside step function, can be effectively trained, thereby expanding design flexibility and computational efficiency. Further empirical validation with more complex architectures remains a valuable direction for future research.
Authors: Pranav Pawar, Dhwaj Jain, Varun Gupta, Kaustav Dedhia, Dashrath Kale, Sudhir Dhekane
Abstract: This paper presents a practical investigation into fine-tuning model parameters for mathematical reasoning tasks through experimenting with various configurations including randomness control, reasoning depth, and sampling strategies, careful tuning demonstrates substantial improvements in efficiency as well as performance. A holistically optimized framework is introduced for five state-of-the-art models on mathematical reasoning tasks, exhibiting significant performance boosts while maintaining solution correctness. Through systematic parameter optimization across Qwen2.5-72B, Llama-3.1-70B, DeepSeek-V3, Mixtral-8x22B, and Yi-Lightning, consistent efficiency gains are demonstrated with 100% optimization success rate. The methodology achieves an average 29.4% reduction in computational cost and 23.9% improvement in inference speed across all tested models. This framework systematically searches parameter spaces including temperature (0.1-0.5), reasoning steps (4-12), planning periods (1-4), and nucleus sampling (0.85-0.98), determining optimal configurations through testing on mathematical reasoning benchmarks. Critical findings show that lower temperature regimes (0.1-0.4) and reduced reasoning steps (4-6) consistently enhance efficiency without compromising accuracy. DeepSeek-V3 achieves the highest accuracy at 98%, while Mixtral-8x22B delivers the most cost-effective performance at 361.5 tokens per accurate response. Key contributions include: (1) the first comprehensive optimization study for five diverse SOTA models in mathematical reasoning, (2) a standardized production-oriented parameter optimization framework, (3) discovery of universal optimization trends applicable across model architectures, and (4) production-ready configurations with extensive performance characterization.
Authors: Julian Killingback, Hamed Zamani
Abstract: Large language models (LLMs) are incredible and versatile tools for text-based tasks that have enabled countless, previously unimaginable, applications. Retrieval models, in contrast, have not yet seen such capable general-purpose models emerge. To achieve this goal, retrieval models must be able to perform complex retrieval tasks, where queries contain multiple parts, constraints, or requirements in natural language. These tasks represent a natural progression from the simple, single-aspect queries that are used in the vast majority of existing, commonly used evaluation sets. Complex queries naturally arise as people expect search systems to handle more specific and often ambitious information requests, as is demonstrated by how people use LLM-based information systems. Despite the growing desire for retrieval models to expand their capabilities in complex retrieval tasks, there exist limited resources to assess the ability of retrieval models on a comprehensive set of diverse complex tasks. The few resources that do exist feature a limited scope and often lack realistic settings making it hard to know the true capabilities of retrieval models on complex real-world retrieval tasks. To address this shortcoming and spur innovation in next-generation retrieval models, we construct a diverse and realistic set of complex retrieval tasks and benchmark a representative set of state-of-the-art retrieval models. Additionally, we explore the impact of LLM-based query expansion and rewriting on retrieval quality. Our results show that even the best models struggle to produce high-quality retrieval results with the highest average nDCG@10 of only 0.346 and R@100 of only 0.587 across all tasks. Although LLM augmentation can help weaker models, the strongest model has decreased performance across all metrics with all rewriting techniques.
Authors: Amelia Kovacs, Jerry Chee, Kimia Kazemian, Sarah Dean
Abstract: Personalized AI systems, from recommendation systems to chatbots, are a prevalent method for distributing content to users based on their learned preferences. However, there is growing concern about the adverse effects of these systems, including their potential tendency to expose users to sensitive or harmful material, negatively impacting overall well-being. To address this concern quantitatively, it is necessary to create datasets with relevant sensitivity labels for content, enabling researchers to evaluate personalized systems beyond mere engagement metrics. To this end, we introduce two novel datasets that include a taxonomy of sensitivity labels alongside user-content ratings: one that integrates MovieLens rating data with content warnings from the Does the Dog Die? community ratings website, and another that combines fan-fiction interaction data and user-generated warnings from Archive of Our Own.
Authors: Sepehr Salem, M. Moein Esfahani, Jingyu Liu, Vince Calhoun
Abstract: Data scarcity hinders deep learning for medical imaging. We propose a framework for breast cancer classification in thermograms that addresses this using a Diffusion Probabilistic Model (DPM) for data augmentation. Our DPM-based augmentation is shown to be superior to both traditional methods and a ProGAN baseline. The framework fuses deep features from a pre-trained ResNet-50 with handcrafted nonlinear features (e.g., Fractal Dimension) derived from U-Net segmented tumors. An XGBoost classifier trained on these fused features achieves 98.0\% accuracy and 98.1\% sensitivity. Ablation studies and statistical tests confirm that both the DPM augmentation and the nonlinear feature fusion are critical, statistically significant components of this success. This work validates the synergy between advanced generative models and interpretable features for creating highly accurate medical diagnostic tools.
Authors: Jeff Shen, Lindsay Smith
Abstract: We present cryptogram solving as an ideal testbed for studying neural network generalization in combinatorially complex domains. In this task, models must decrypt text encoded with substitution ciphers, choosing from 26! possible mappings without explicit access to the cipher. We develop ALICE (an Architecture for Learning Interpretable Cryptogram dEcipherment): a simple encoder-only Transformer that sets a new state-of-the-art for both accuracy and speed on this decryption problem. Surprisingly, ALICE generalizes to unseen ciphers after training on only ${\sim}1500$ unique ciphers, a minute fraction ($3.7 \times 10^{-24}$) of the possible cipher space. To enhance interpretability, we introduce a novel bijective decoding head that explicitly models permutations via the Gumbel-Sinkhorn method, enabling direct extraction of learned cipher mappings. Through early exit analysis, we reveal how ALICE progressively refines its predictions in a way that appears to mirror common human strategies for this task: early layers employ frequency-based heuristics, middle layers form word structures, and final layers correct individual characters. Our architectural innovations and analysis methods extend beyond cryptograms to any domain with bijective mappings and combinatorial structure, offering new insights into neural network generalization and interpretability.
Authors: Yan Pang, Wenlong Meng, Xiaojing Liao, Tianhao Wang
Abstract: With the rapid development of large language models, the potential threat of their malicious use, particularly in generating phishing content, is becoming increasingly prevalent. Leveraging the capabilities of LLMs, malicious users can synthesize phishing emails that are free from spelling mistakes and other easily detectable features. Furthermore, such models can generate topic-specific phishing messages, tailoring content to the target domain and increasing the likelihood of success. Detecting such content remains a significant challenge, as LLM-generated phishing emails often lack clear or distinguishable linguistic features. As a result, most existing semantic-level detection approaches struggle to identify them reliably. While certain LLM-based detection methods have shown promise, they suffer from high computational costs and are constrained by the performance of the underlying language model, making them impractical for large-scale deployment. In this work, we aim to address this issue. We propose Paladin, which embeds trigger-tag associations into vanilla LLM using various insertion strategies, creating them into instrumented LLMs. When an instrumented LLM generates content related to phishing, it will automatically include detectable tags, enabling easier identification. Based on the design on implicit and explicit triggers and tags, we consider four distinct scenarios in our work. We evaluate our method from three key perspectives: stealthiness, effectiveness, and robustness, and compare it with existing baseline methods. Experimental results show that our method outperforms the baselines, achieving over 90% detection accuracy across all scenarios.
Authors: Nan Wang, Nan Wu, Xiangyu Hui, Jiafan Wang, Xin Yuan
Abstract: As the demand for exercising the "right to be forgotten" grows, the need for verifiable machine unlearning has become increasingly evident to ensure both transparency and accountability. We present {\em zkUnlearner}, the first zero-knowledge framework for verifiable machine unlearning, specifically designed to support {\em multi-granularity} and {\em forgery-resistance}. First, we propose a general computational model that employs a {\em bit-masking} technique to enable the {\em selectivity} of existing zero-knowledge proofs of training for gradient descent algorithms. This innovation enables not only traditional {\em sample-level} unlearning but also more advanced {\em feature-level} and {\em class-level} unlearning. Our model can be translated to arithmetic circuits, ensuring compatibility with a broad range of zero-knowledge proof systems. Furthermore, our approach overcomes key limitations of existing methods in both efficiency and privacy. Second, forging attacks present a serious threat to the reliability of unlearning. Specifically, in Stochastic Gradient Descent optimization, gradients from unlearned data, or from minibatches containing it, can be forged using alternative data samples or minibatches that exclude it. We propose the first effective strategies to resist state-of-the-art forging attacks. Finally, we benchmark a zkSNARK-based instantiation of our framework and perform comprehensive performance evaluations to validate its practicality.
Authors: Ji Xie, Trevor Darrell, Luke Zettlemoyer, XuDong Wang
Abstract: Unified multimodal models (UMMs) unify visual understanding and generation within a single architecture. However, conventional training relies on image-text pairs (or sequences) whose captions are typically sparse and miss fine-grained visual details--even when they use hundreds of words to describe a simple image. We introduce Reconstruction Alignment (RecA), a resource-efficient post-training method that leverages visual understanding encoder embeddings as dense "text prompts," providing rich supervision without captions. Concretely, RecA conditions a UMM on its own visual understanding embeddings and optimizes it to reconstruct the input image with a self-supervised reconstruction loss, thereby realigning understanding and generation. Despite its simplicity, RecA is broadly applicable: across autoregressive, masked-autoregressive, and diffusion-based UMMs, it consistently improves generation and editing fidelity. With only 27 GPU-hours, post-training with RecA substantially improves image generation performance on GenEval (0.73$\rightarrow$0.90) and DPGBench (80.93$\rightarrow$88.15), while also boosting editing benchmarks (ImgEdit 3.38$\rightarrow$3.75, GEdit 6.94$\rightarrow$7.25). Notably, RecA surpasses much larger open-source models and applies broadly across diverse UMM architectures, establishing it as an efficient and general post-training alignment strategy for UMMs
Authors: David Oprea, Sam Powers
Abstract: We test a new method, which we will abbreviate using the acronym BVM (Basis Vectors Method), in its ability to judge the state changes in images through using language embeddings. We used the MIT-States dataset, containing about 53,000 images, to gather all of our data, which has 225 nouns and 115 adjectives, with each noun having about 9 different adjectives, forming approximately 1000 noun-adjective pairs. For our first experiment, we test our method's ability to determine the state of each noun class separately against other metrics for comparison. These metrics are cosine similarity, dot product, product quantization, binary index, Naive Bayes, and a custom neural network. Among these metrics, we found that our proposed BVM performs the best in classifying the states for each noun. We then perform a second experiment where we try using BVM to determine if it can differentiate adjectives from one another for each adjective separately. We compared the abilities of BVM to differentiate adjectives against the proposed method the MIT-States paper suggests: using a logistic regression model. In the end, we did not find conclusive evidence that our BVM metric could perform better than the logistic regression model at discerning adjectives. Yet, we were able to find evidence for possible improvements to our method; this leads to the chance of increasing our method's accuracy through certain changes in our methodologies.
Authors: Sihyun Park
Abstract: Recent advances in large language models (LLMs) have been driven by pretraining, supervised fine tuning (SFT), and alignment tuning. Among these, SFT plays a crucial role in transforming a model 's general knowledge into structured responses tailored to specific tasks. However, there is no clearly established methodology for effective training data selection. Simply increasing the volume of data does not guarantee performance improvements, while preprocessing, sampling, and validation require substantial time and cost. To address this issue, a variety of data selection methods have been proposed. Among them, knowledge based selection approaches identify suitable training data by analyzing the model 's responses. Nevertheless, these methods typically rely on prompt engineering, making them sensitive to variations and incurring additional costs for prompt design. In this study, we propose Knowledge Analysis via Model Internal Representations (KAMIR), a novel approach that overcomes these limitations by analyzing data based on the model 's internal representations. KAMIR computes similarities between the hidden states of each layer (block) and the final hidden states for a given input to assess the data. Unlike prior methods that were largely limited to multiple choice tasks, KAMIR can be applied to a wide range of tasks such as machine reading comprehension and summarization. Moreover, it selects data useful for training based on the model 's familiarity with the input, even with a small dataset and a simple classifier architecture. Experiments across diverse task datasets demonstrate that training with less familiar data leads to better generalization performance.
Authors: Yunxiao Shi, Shuo Yang, Haimin Zhang, Li Wang, Yongze Wang, Qiang Wu, Min Xu
Abstract: Neural Collaborative Filtering models are widely used in recommender systems but are typically trained under static settings, assuming fixed data distributions. This limits their applicability in dynamic environments where user preferences evolve. Incremental learning offers a promising solution, yet conventional methods from computer vision or NLP face challenges in recommendation tasks due to data sparsity and distinct task paradigms. Existing approaches for neural recommenders remain limited and often lack generalizability. To address this, we propose MEGG, Replay Samples with Maximally Extreme GGscore, an experience replay based incremental learning framework. MEGG introduces GGscore, a novel metric that quantifies sample influence, enabling the selective replay of highly influential samples to mitigate catastrophic forgetting. Being model-agnostic, MEGG integrates seamlessly across architectures and frameworks. Experiments on three neural models and four benchmark datasets show superior performance over state-of-the-art baselines, with strong scalability, efficiency, and robustness. Implementation will be released publicly upon acceptance.
Authors: Nakyung Lee, Yeongoon Kim, Minhae Oh, Suhwan Kim, Jin Woo Koo, Hyewon Jo, Jungwoo Lee
Abstract: Transformer-based self-attention mechanism serves as the core of modern language models, yet it often suffers from localization, where attentions collapse onto a limited subset of tokens and fail to capture long-range dependencies. To address this issue, we propose Self-Attention One-step Belief Propagation (SAOBP), a refinement framework that injects multi-hop relationships through a belief propagation process. To interpret and quantify these interactions, we introduce Global Token Dependency (GTD) that captures the relative contribution of multihop connections within the attention graph. Empirical results indicate that SAOBP helps prevent entropy collapse in deeper layers and adaptively maintains GTD at task-appropriate levels, thereby supporting improvements in model performance. Importantly, we observe competitive gains in small-scale models, highlighting its potential for improving inference quality in resource-constrained scenarios.
Authors: Shucong Li, Zhenyu Liu, Zijie Hong, Zhiheng Zhou, Xianghai Cao
Abstract: Multispectral remote sensing object detection is one of the important application of unmanned aerial vehicle (UAV). However, it faces three challenges. Firstly, the low-light remote sensing images reduce the complementarity during multi-modality fusion. Secondly, the local small target modeling is interfered with redundant information in the fusion stage easily. Thirdly, due to the quadratic computational complexity, it is hard to apply the transformer-based methods on the UAV platform. To address these limitations, motivated by Mamba with linear complexity, a UAV multispectral object detector with dual-domain enhancement and priority-guided mamba fusion (DEPF) is proposed. Firstly, to enhance low-light remote sensing images, Dual-Domain Enhancement Module (DDE) is designed, which contains Cross-Scale Wavelet Mamba (CSWM) and Fourier Details Recovery block (FDR). CSWM applies cross-scale mamba scanning for the low-frequency components to enhance the global brightness of images, while FDR constructs spectrum recovery network to enhance the frequency spectra features for recovering the texture-details. Secondly, to enhance local target modeling and reduce the impact of redundant information during fusion, Priority-Guided Mamba Fusion Module (PGMF) is designed. PGMF introduces the concept of priority scanning, which starts from local targets features according to the priority scores obtained from modality difference. Experiments on DroneVehicle dataset and VEDAI dataset reports that, DEPF performs well on object detection, comparing with state-of-the-art methods. Our code is available in the supplementary material.
Authors: Li-Chin Chen, Ji-Tian Sheu, Yuh-Jue Chuang
Abstract: Demographic attributes are universally present in electronic health records and serve as vital predictors in clinical risk stratification and treatment decisions. Despite their significance, these attributes are often relegated to auxiliary roles in model design, with limited attention has been given to learning their representations. This study proposes a General Demographic Pre-trained (GDP) model as a foundational representation framework tailored to age and gender. The model is pre-trained and evaluated using datasets with diverse diseases and population compositions from different geographic regions. The GDP architecture explores combinations of ordering strategies and encoding methods to transform tabular demographic inputs into latent embeddings. Experimental results demonstrate that sequential ordering substantially improves model performance in discrimination, calibration, and the corresponding information gain at each decision tree split, particularly in diseases where age and gender contribute significantly to risk stratification. Even in datasets where demographic attributes hold relatively low predictive value, GDP enhances the representational importance, increasing their influence in downstream gradient boosting models. The findings suggest that foundational models for tabular demographic attributes can generalize across tasks and populations, offering a promising direction for improving predictive performance in healthcare applications.
Authors: Archit Kalra, Midhun Sadanand
Abstract: Spiking neural networks offer a promising path toward energy-efficient, brain-like associative memory. This paper introduces Word2Spike, a novel rate coding mechanism that combines continuous word embeddings and neuromorphic architectures. We develop a one-to-one mapping that converts multi-dimensional word vectors into spike-based attractor states using Poisson processes. Using BitNet b1.58 quantization, we maintain 97% semantic similarity of continuous embeddings on SimLex-999 while achieving 100% reconstruction accuracy on 10,000 words from OpenAI's text-embedding-3-large. We preserve analogy performance (100% of original embedding performance) even under intentionally introduced noise, indicating a resilient mechanism for semantic encoding in neuromorphic systems. Next steps include integrating the mapping with spiking transformers and liquid state machines (resembling Hopfield Networks) for further evaluation.
Authors: Qihu Xie, Yuan Li, Yi Kang
Abstract: Implicit neural representations have recently been extended to represent convolutional neural network weights via neural representation for neural networks, offering promising parameter compression benefits. However, standard multi-layer perceptrons used in neural representation for neural networks exhibit a pronounced spectral bias, hampering their ability to reconstruct high-frequency details effectively. In this paper, we propose SBS, a parameter-efficient enhancement to neural representation for neural networks that suppresses spectral bias using two techniques: (1) a unidirectional ordering-based smoothing that improves kernel smoothness in the output space, and (2) unidirectional ordering-based smoothing aware random fourier features that adaptively modulate the frequency bandwidth of input encodings based on layer-wise parameter count. Extensive evaluations on various ResNet models with datasets CIFAR-10, CIFAR-100, and ImageNet, demonstrate that SBS achieves significantly better reconstruction accuracy with less parameters compared to SOTA.
Authors: Sankalp Tattwadarshi Swain, Anshika Krishnatray, Dhruv Kumar, Jagat Sesh Challa
Abstract: Existing evaluation studies on linguistic competence of large language models (LLM agents) have focused primarily on vocabulary learning, morphological rule induction, syntactic generalization, pragmatic inference, and cross-linguistic transfer. However, none assess whether LLM agents can acquire a language through pattern recognition and interactive feedback, a central feature of human language acquisition. We propose a novel experimental framework in which an LLM agent is evaluated on its ability to acquire and use a newly constructed language (Tinkatongue) in conversation with a bot that understands only Tinkatongue. Our findings show that LLM agents fail to establish a conversation within 100 responses, yet they adopt distinct strategies that mirror human approaches to language learning. The results suggest a new direction for evaluation benchmarks and open pathways to model designs that learn more effectively from interactive feedback.
Authors: Gyuyeon Na, Minjung Park, Hyeonjeong Cha, Soyoun Kim, Sunyoung Moon, Sua Lee, Jaeyoung Choi, Hyemin Lee, Sangmi Chai
Abstract: Blockchain transaction networks are complex, with evolving temporal patterns and inter-node relationships. To detect illicit activities, we propose a hybrid GCN-GRU model that captures both structural and sequential features. Using real Bitcoin transaction data (2020-2024), our model achieved 0.9470 Accuracy and 0.9807 AUC-ROC, outperforming all baselines.
Authors: Ziang Yin, Hongjian Zhou, Chetan Choppali Sudarshan, Vidya Chhabria, Jiaqi Gu
Abstract: The relentless growth of large-scale artificial intelligence (AI) has created unprecedented demand for computational power, straining the energy, bandwidth, and scaling limits of conventional electronic platforms. Electronic-photonic integrated circuits (EPICs) have emerged as a compelling platform for next-generation AI systems, offering inherent advantages in ultra-high bandwidth, low latency, and energy efficiency for computing and interconnection. Beyond performance, EPICs also hold unique promises for sustainability. Fabricated in relaxed process nodes with fewer metal layers and lower defect densities, photonic devices naturally reduce embodied carbon footprint (CFP) compared to advanced digital electronic integrated circuits, while delivering orders-of-magnitude higher computing performance and interconnect bandwidth. To further advance the sustainability of photonic AI systems, we explore how electronic-photonic design automation (EPDA) and cross-layer co-design methodologies can amplify these inherent benefits. We present how advanced EPDA tools enable more compact layout generation, reducing both chip area and metal layer usage. We will also demonstrate how cross-layer device-circuit-architecture co-design unlocks new sustainability gains for photonic hardware: ultra-compact photonic circuit designs that minimize chip area cost, reconfigurable hardware topology that adapts to evolving AI workloads, and intelligent resilience mechanisms that prolong lifetime by tolerating variations and faults. By uniting intrinsic photonic efficiency with EPDA- and co-design-driven gains in area efficiency, reconfigurability, and robustness, we outline a vision for lifelong-sustainable electronic-photonic AI systems. This perspective highlights how EPIC AI systems can simultaneously meet the performance demands of modern AI and the urgent imperative for sustainable computing.
Authors: Shusuke Ito, Koki Muraoka, Akira Nakayama
Abstract: Interatomic potentials (IPs) with wide elemental coverage and high accuracy are powerful tools for high-throughput materials discovery. While the past few years witnessed the development of multiple new universal IPs that cover wide ranges of the periodic table, their applicability to target chemical systems should be carefully investigated. We benchmark several universal IPs using equilibrium zeolite structures as testbeds. We select a diverse set of universal IPs encompassing two major categories: (i) universal analytic IPs, including GFN-FF, UFF, and Dreiding; (ii) pretrained universal machine learning IPs (MLIPs), comprising CHGNet, ORB-v3, MatterSim, eSEN-30M-OAM, PFP-v7, and EquiformerV2-lE4-lF100-S2EFS-OC22. We compare them with established tailor-made IPs, SLC, ClayFF, and BSFF using experimental data and density functional theory (DFT) calculations with dispersion correction as the reference. The tested zeolite structures comprise pure silica frameworks and aluminosilicates containing copper species, potassium, and organic cations. We found that GFN-FF is the best among the tested universal analytic IPs, but it does not achieve satisfactory accuracy for highly strained silica rings and aluminosilicate systems. All MLIPs can well reproduce experimental or DFT-level geometries and energetics. Among the universal MLIPs, the eSEN-30M-OAM model shows the most consistent performance across all zeolite structures studied. These findings show that the modern pretrained universal MLIPs are practical tools in zeolite screening workflows involving various compositions.
Authors: Long Li, Jiaran Hao, Jason Klein Liu, Zhijian Zhou, Xiaoyu Tan, Wei Chu, Zhe Wang, Shirui Pan, Chao Qu, Yuan Qi
Abstract: A central paradox in fine-tuning Large Language Models (LLMs) with Reinforcement Learning with Verifiable Reward (RLVR) is the frequent degradation of multi-attempt performance (Pass@k) despite improvements in single-attempt accuracy (Pass@1). This is often accompanied by catastrophic forgetting, where models lose previously acquired skills. While various methods have been proposed, the choice and function of the divergence term have been surprisingly unexamined as a proactive solution. We argue that standard RLVR objectives -- both those using the mode-seeking reverse KL-divergence and those forgoing a divergence term entirely -- lack a crucial mechanism for knowledge retention. The reverse-KL actively accelerates this decay by narrowing the policy, while its absence provides no safeguard against the model drifting from its diverse knowledge base. We propose a fundamental shift in perspective: using the divergence term itself as the solution. Our framework, Diversity-Preserving Hybrid RL (DPH-RL), leverages mass-covering f-divergences (like forward-KL and JS-divergence) to function as a rehearsal mechanism. By continuously referencing the initial policy, this approach forces the model to maintain broad solution coverage. Extensive experiments on math and SQL generation demonstrate that DPH-RL not only resolves the Pass@k degradation but improves both Pass@1 and Pass@k in- and out-of-domain. Additionally, DPH-RL is more training-efficient because it computes f-divergence using generator functions, requiring only sampling from the initial policy and no online reference model. Our work highlights a crucial, overlooked axis for improving RLVR, demonstrating that the proper selection of a divergence measure is a powerful tool for building more general and diverse reasoning models.
Authors: Harrison Field, Max Yang, Yijiong Lin, Efi Psomopoulou, David Barton, Nathan F. Lepora
Abstract: Large language models (LLMs) are beginning to automate reward design for dexterous manipulation. However, no prior work has considered tactile sensing, which is known to be critical for human-like dexterity. We present Text2Touch, bringing LLM-crafted rewards to the challenging task of multi-axis in-hand object rotation with real-world vision based tactile sensing in palm-up and palm-down configurations. Our prompt engineering strategy scales to over 70 environment variables, and sim-to-real distillation enables successful policy transfer to a tactile-enabled fully actuated four-fingered dexterous robot hand. Text2Touch significantly outperforms a carefully tuned human-engineered baseline, demonstrating superior rotation speed and stability while relying on reward functions that are an order of magnitude shorter and simpler. These results illustrate how LLM-designed rewards can significantly reduce the time from concept to deployable dexterous tactile skills, supporting more rapid and scalable multimodal robot learning. Project website: https://hpfield.github.io/text2touch-website
Authors: Sai Siddhartha Chary Aylapuram, Veeraraju Elluru, Shivang Agarwal
Abstract: Deep neural networks often rely on spurious correlations in training data, leading to biased or unfair predictions in safety-critical domains such as medicine and autonomous driving. While conventional bias mitigation typically requires retraining from scratch or redesigning data pipelines, recent advances in machine unlearning provide a promising alternative for post-hoc model correction. In this work, we investigate \textit{Bias-Aware Machine Unlearning}, a paradigm that selectively removes biased samples or feature representations to mitigate diverse forms of bias in vision models. Building on privacy-preserving unlearning techniques, we evaluate various strategies including Gradient Ascent, LoRA, and Teacher-Student distillation. Through empirical analysis on three benchmark datasets, CUB-200-2011 (pose bias), CIFAR-10 (synthetic patch bias), and CelebA (gender bias in smile detection), we demonstrate that post-hoc unlearning can substantially reduce subgroup disparities, with improvements in demographic parity of up to \textbf{94.86\%} on CUB-200, \textbf{30.28\%} on CIFAR-10, and \textbf{97.37\%} on CelebA. These gains are achieved with minimal accuracy loss and with methods scoring an average of 0.62 across the 3 settings on the joint evaluation of utility, fairness, quality, and privacy. Our findings establish machine unlearning as a practical framework for enhancing fairness in deployed vision systems without necessitating full retraining.
Authors: Sven Kirchner, Nils Purschke, Ross Greer, Alois C. Knoll
Abstract: Ensuring reliable robot operation when visual input is degraded or insufficient remains a central challenge in robotics. This letter introduces DepthVision, a framework for multimodal scene understanding designed to address this problem. Unlike existing Vision-Language Models (VLMs), which use only camera-based visual input alongside language, DepthVision synthesizes RGB images from sparse LiDAR point clouds using a conditional generative adversarial network (GAN) with an integrated refiner network. These synthetic views are then combined with real RGB data using a Luminance-Aware Modality Adaptation (LAMA), which blends the two types of data dynamically based on ambient lighting conditions. This approach compensates for sensor degradation, such as darkness or motion blur, without requiring any fine-tuning of downstream vision-language models. We evaluate DepthVision on real and simulated datasets across various models and tasks, with particular attention to safety-critical tasks. The results demonstrate that our approach improves performance in low-light conditions, achieving substantial gains over RGB-only baselines while preserving compatibility with frozen VLMs. This work highlights the potential of LiDAR-guided RGB synthesis for achieving robust robot operation in real-world environments.
Authors: Saumya Goswami, Siddharth Kurra
Abstract: Detecting content that contradicts or is unsupported by a given source text is a critical challenge for the safe deployment of generative language models. We introduce HALT-RAG, a post-hoc verification system designed to identify hallucinations in the outputs of Retrieval-Augmented Generation (RAG) pipelines. Our flexible and task-adaptable framework uses a universal feature set derived from an ensemble of two frozen, off-the-shelf Natural Language Inference (NLI) models and lightweight lexical signals. These features are used to train a simple, calibrated, and task-adapted meta-classifier. Using a rigorous 5-fold out-of-fold (OOF) training protocol to prevent data leakage and produce unbiased estimates, we evaluate our system on the HaluEval benchmark. By pairing our universal feature set with a lightweight, task-adapted classifier and a precision-constrained decision policy, HALT-RAG achieves strong OOF F1-scores of 0.7756, 0.9786, and 0.7391 on the summarization, QA, and dialogue tasks, respectively. The system's well-calibrated probabilities enable a practical abstention mechanism, providing a reliable tool for balancing model performance with safety requirements.
Authors: Xiao Li, Bharat Gandhi, Ming Zhan, Mohit Nehra, Zhicheng Zhang, Yuchen Sun, Meijia Song, Naisheng Zhang, Xi Wang
Abstract: We address vision-language-driven indoor navigation to assist visually impaired individuals in reaching a target location using images and natural language guidance. Traditional navigation systems are ineffective indoors due to the lack of precise location data. Our approach integrates vision and language models to generate step-by-step navigational instructions, enhancing accessibility and independence. We fine-tune the BLIP-2 model with Low Rank Adaptation (LoRA) on a manually annotated indoor navigation dataset. We propose an evaluation metric that refines the BERT F1 score by emphasizing directional and sequential variables, providing a more comprehensive measure of navigational performance. After applying LoRA, the model significantly improved in generating directional instructions, overcoming limitations in the original BLIP-2 model.
Authors: Chun Liu, Hailong Wang, Bingqian Zhu, Panpan Ding, Zheng Zheng, Tao Xu, Zhigang Han, Jiayao Wang
Abstract: Deep Neural Networks (DNNs) are vulnerable to adversarial attacks, posing significant security threats to their deployment in remote sensing applications. Research on adversarial attacks not only reveals model vulnerabilities but also provides critical insights for enhancing robustness. Although current mixing-based strategies have been proposed to increase the transferability of adversarial examples, they either perform global blending or directly exchange a region in the images, which may destroy global semantic features and mislead the optimization of adversarial examples. Furthermore, their reliance on cross-entropy loss for perturbation optimization leads to gradient diminishing during iterative updates, compromising adversarial example quality. To address these limitations, we focus on non-targeted attacks and propose a novel framework via local mixing and logits optimization. First, we present a local mixing strategy to generate diverse yet semantically consistent inputs. Different from MixUp, which globally blends two images, and MixCut, which stitches images together, our method merely blends local regions to preserve global semantic information. Second, we adapt the logit loss from targeted attacks to non-targeted scenarios, mitigating the gradient vanishing problem of cross-entropy loss. Third, a perturbation smoothing loss is applied to suppress high-frequency noise and enhance transferability. Extensive experiments on FGSCR-42 and MTARSI datasets demonstrate superior performance over 12 state-of-the-art methods across 6 surrogate models. Notably, with ResNet as the surrogate on MTARSI, our method achieves a 17.28% average improvement in black-box attack success rate.
Authors: Anjiang Wei, Tianran Sun, Yogesh Seenichamy, Hang Song, Anne Ouyang, Azalia Mirhoseini, Ke Wang, Alex Aiken
Abstract: GPU kernel optimization has long been a central challenge at the intersection of high-performance computing and machine learning. Efficient kernels are crucial for accelerating large language model (LLM) training and serving, yet attaining high performance typically requires extensive manual tuning. Compiler-based systems reduce some of this burden, but still demand substantial manual design and engineering effort. Recently, researchers have explored using LLMs for GPU kernel generation, though prior work has largely focused on translating high-level PyTorch modules into CUDA code. In this work, we introduce Astra, the first LLM-based multi-agent system for GPU kernel optimization. Unlike previous approaches, Astra starts from existing CUDA implementations extracted from SGLang, a widely deployed framework for serving LLMs, rather than treating PyTorch modules as the specification. Within Astra, specialized LLM agents collaborate through iterative code generation, testing, profiling, and planning to produce kernels that are both correct and high-performance. On kernels from SGLang, Astra achieves an average speedup of 1.32x using zero-shot prompting with OpenAI o4-mini. A detailed case study further demonstrates that LLMs can autonomously apply loop transformations, optimize memory access patterns, exploit CUDA intrinsics, and leverage fast math operations to yield substantial performance gains. Our work highlights multi-agent LLM systems as a promising new paradigm for GPU kernel optimization.
Authors: Zihan Chen, Lei Shi, Weize Wu, Qiji Zhou, Yue Zhang
Abstract: Many contemporary data-driven research efforts in the natural sciences, such as chemistry and materials science, require large-scale, high-performance entity recognition from scientific datasets. Large language models (LLMs) have increasingly been adopted to solve the entity recognition task, with the same trend being observed on all-spectrum NLP tasks. The prevailing entity recognition LLMs rely on fine-tuned technology, yet the fine-tuning process often incurs significant cost. To achieve a best performance-cost trade-off, we propose ALLabel, a three-stage framework designed to select the most informative and representative samples in preparing the demonstrations for LLM modeling. The annotated examples are used to construct a ground-truth retrieval corpus for LLM in-context learning. By sequentially employing three distinct active learning strategies, ALLabel consistently outperforms all baselines under the same annotation budget across three specialized domain datasets. Experimental results also demonstrate that selectively annotating only 5\%-10\% of the dataset with ALLabel can achieve performance comparable to the method annotating the entire dataset. Further analyses and ablation studies verify the effectiveness and generalizability of our proposal.
Authors: Adithya Ramachandran, Thorkil Flensmark B. Neergaard, Tom\'as Arias-Vergara, Andreas Maier, Siming Bayer
Abstract: Advancements in smart metering technologies have significantly improved the ability to monitor and manage water utilities. In the context of increasing uncertainty due to climate change, securing water resources and supply has emerged as an urgent global issue with extensive socioeconomic ramifications. Hourly consumption data from end-users have yielded substantial insights for projecting demand across regions characterized by diverse consumption patterns. Nevertheless, the prediction of water demand remains challenging due to influencing non-deterministic factors, such as meteorological conditions. This work introduces a novel method for short-term water demand forecasting for District Metered Areas (DMAs) which encompass commercial, agricultural, and residential consumers. Unsupervised contrastive learning is applied to categorize end-users according to distinct consumption behaviors present within a DMA. Subsequently, the distinct consumption behaviors are utilized as features in the ensuing demand forecasting task using wavelet-transformed convolutional networks that incorporate a cross-attention mechanism combining both historical data and the derived representations. The proposed approach is evaluated on real-world DMAs over a six-month period, demonstrating improved forecasting performance in terms of MAPE across different DMAs, with a maximum improvement of 4.9%. Additionally, it identifies consumers whose behavior is shaped by socioeconomic factors, enhancing prior knowledge about the deterministic patterns that influence demand.
Authors: Gianluca Amprimo, Alberto Ancilotto, Alessandro Savino, Fabio Quazzolo, Claudia Ferraris, Gabriella Olmo, Elisabetta Farella, Stefano Di Carlo
Abstract: Hand gesture understanding is essential for several applications in human-computer interaction, including automatic clinical assessment of hand dexterity. While deep learning has advanced static gesture recognition, dynamic gesture understanding remains challenging due to complex spatiotemporal variations. Moreover, existing datasets often lack multimodal and multi-view diversity, precise ground-truth tracking, and an action quality component embedded within gestures. This paper introduces EHWGesture, a multimodal video dataset for gesture understanding featuring five clinically relevant gestures. It includes over 1,100 recordings (6 hours), captured from 25 healthy subjects using two high-resolution RGB-Depth cameras and an event camera. A motion capture system provides precise ground-truth hand landmark tracking, and all devices are spatially calibrated and synchronized to ensure cross-modal alignment. Moreover, to embed an action quality task within gesture understanding, collected recordings are organized in classes of execution speed that mirror clinical evaluations of hand dexterity. Baseline experiments highlight the dataset's potential for gesture classification, gesture trigger detection, and action quality assessment. Thus, EHWGesture can serve as a comprehensive benchmark for advancing multimodal clinical gesture understanding.
Authors: Gokul Karthik Kumar, Rishabh Saraf, Ludovick Lepauloux, Abdul Muneer, Billel Mokeddem, Hakim Hacid
Abstract: Large language models (LLMs) have transformed NLP, yet their integration with audio remains underexplored -- despite audio's centrality to human communication. We introduce Falcon3-Audio, a family of Audio-Language Models (ALMs) built on instruction-tuned LLMs and Whisper encoders. Using a remarkably small amount of public audio data -- less than 30K hours (5K unique) -- Falcon3-Audio-7B matches the best reported performance among open-weight models on the MMAU benchmark, with a score of 64.14, matching R1-AQA, while distinguishing itself through superior data and parameter efficiency, single-stage training, and transparency. Notably, our smallest 1B model remains competitive with larger open models ranging from 2B to 13B parameters. Through extensive ablations, we find that common complexities -- such as curriculum learning, multiple audio encoders, and intricate cross-attention connectors -- are not required for strong performance, even compared to models trained on over 500K hours of data.
Authors: Zheng Dou, Deqing Wang, Fuzhen Zhuang, Jian Ren, Yanlin Hu
Abstract: Scientific document representation learning provides powerful embeddings for various tasks, while current methods face challenges across three approaches. 1) Contrastive training with citation-structural signals underutilizes citation information and still generates single-vector representations. 2) Fine-grained representation learning, which generates multiple vectors at the sentence or aspect level, requires costly integration and lacks domain generalization. 3) Task-aware learning depends on manually predefined task categorization, overlooking nuanced task distinctions and requiring extra training data for task-specific modules. To address these problems, we propose a new method that unifies the three approaches for better representations, namely FLeW. Specifically, we introduce a novel triplet sampling method that leverages citation intent and frequency to enhance citation-structural signals for training. Citation intents (background, method, result), aligned with the general structure of scientific writing, facilitate a domain-generalized facet partition for fine-grained representation learning. Then, we adopt a simple weight search to adaptively integrate three facet-level embeddings into a task-specific document embedding without task-aware fine-tuning. Experiments show the applicability and robustness of FLeW across multiple scientific tasks and fields, compared to prior models.
Authors: Jin Lee, Vu Dang, Gwang-Hyun Yu, Anh Le, Zahid Rahman, Jin-Ho Jang, Heonzoo Lee, Kun-Yung Kim, Jin-Sul Kim, Jin-Young Kim
Abstract: While Masked Image Modeling (MIM) has revolutionized fields of computer vision, its adoption in 3D medical image computing has been limited by the use of random masking, which overlooks the density of anatomical objects. To address this limitation, we enhance the pretext task with a simple yet effective masking strategy. Leveraging Hounsfield Unit (HU) measurements, we implement an HU-based Foreground Masking, which focuses on the intensity distribution of visceral organs and excludes non-tissue regions, such as air and fluid, that lack diagnostically meaningful features. Extensive experiments on five public 3D medical imaging datasets demonstrate that our masking consistently improves performance, both in quality of segmentation and Dice score (BTCV:~84.64\%, Flare22:~92.43\%, MM-WHS:~90.67\%, Amos22:~88.64\%, BraTS:~78.55\%). These results underscore the importance of domain-centric MIM and suggest a promising direction for representation learning in medical image segmentation. Implementation is available at github.com/AISeedHub/SubFore/.
Authors: Yi Liu, Xiangrong Zhu, Xiangyu Liu, Wei Wei, Wei Hu
Abstract: In a rapidly evolving world where information updates swiftly, knowledge in large language models (LLMs) becomes outdated quickly. Retraining LLMs is not a cost-effective option, making knowledge editing (KE) without modifying parameters particularly necessary. We find that although existing retrieval-augmented generation (RAG)-based KE methods excel at editing simple knowledge, they struggle with KE in multi-hop question answering due to the issue of "edit skipping", which refers to skipping the relevant edited fact in inference. In addition to the diversity of natural language expressions of knowledge, edit skipping also arises from the mismatch between the granularity of LLMs in problem-solving and the facts in the edited memory. To address this issue, we propose a novel Iterative Retrieval-Augmented Knowledge Editing method with guided decomposition (IRAKE) through the guidance from single edited facts and entire edited cases. Experimental results demonstrate that IRAKE mitigates the failure of editing caused by edit skipping and outperforms state-of-the-art methods for KE in multi-hop question answering.
Authors: Zhiyuan He, Xufang Luo, Yike Zhang, Yuqing Yang, Lili Qiu
Abstract: We propose $\Delta L$ Normalization, a simple yet effective loss aggregation method tailored to the characteristic of dynamic generation lengths in Reinforcement Learning with Verifiable Rewards (RLVR). Recently, RLVR has demonstrated strong potential in improving the reasoning capabilities of large language models (LLMs), but a major challenge lies in the large variability of response lengths during training, which leads to high gradient variance and unstable optimization. Although previous methods such as GRPO, DAPO, and Dr. GRPO introduce different loss normalization terms to address this issue, they either produce biased estimates or still suffer from high gradient variance. By analyzing the effect of varying lengths on policy loss both theoretically and empirically, we reformulate the problem as finding a minimum-variance unbiased estimator. Our proposed $\Delta L$ Normalization not only provides an unbiased estimate of the true policy loss but also minimizes gradient variance in theory. Extensive experiments show that it consistently achieves superior results across different model sizes, maximum lengths, and tasks. Our code will be made public at https://github.com/zerolllin/Delta-L-Normalization.
Authors: Xiyu Guo, Shan Wang, Chunfang Ji, Xuefeng Zhao, Wenhao Xi, Yaoyao Liu, Qinglan Li, Chao Deng, Junlan Feng
Abstract: The rapid advancement of large language models (LLMs) and domain-specific AI agents has greatly expanded the ecosystem of AI-powered services. User queries, however, are highly diverse and often span multiple domains and task types, resulting in a complex and heterogeneous landscape. This diversity presents a fundamental routing challenge: how to accurately direct each query to an appropriate execution unit while optimizing both performance and efficiency. To address this, we propose MoMA (Mixture of Models and Agents), a generalized routing framework that integrates both LLM and agent-based routing. Built upon a deep understanding of model and agent capabilities, MoMA effectively handles diverse queries through precise intent recognition and adaptive routing strategies, achieving an optimal balance between efficiency and cost. Specifically, we construct a detailed training dataset to profile the capabilities of various LLMs under different routing model structures, identifying the most suitable tasks for each LLM. During inference, queries are dynamically routed to the LLM with the best cost-performance efficiency. We also introduce an efficient agent selection strategy based on a context-aware state machine and dynamic masking. Experimental results demonstrate that the MoMA router offers superior cost-efficiency and scalability compared to existing approaches.
Authors: Barkin Buyukcakir, Rocharles Cavalcante Fontenele, Reinhilde Jacobs, Jannick De Tobel, Patrick Thevissen, Dirk Vandermeulen, Peter Claes
Abstract: Deep learning offers a promising avenue for automating many recognition tasks in fields such as medicine and forensics. However, the black-box nature of these models hinders their adoption in high-stakes applications where trust and accountability are required. For 3D shape recognition tasks in particular, this paper introduces the Class Node Graph Attention Network (CGAT) architecture to address this need. Applied to 3D meshes of third molars derived from CBCT images, for Demirjian stage allocation, CGAT utilizes graph attention convolutions and an inherent attention mechanism, visualized via attention rollout, to explain its decision-making process. We evaluated the local mean curvature and distance to centroid node features, both individually and in combination, as well as model depth, finding that models incorporating directed edges to a global CLS node produced more intuitive attention maps, while also yielding desirable classification performance. We analyzed the attention-based explanations of the models, and their predictive performances to propose optimal settings for the CGAT. The combination of local mean curvature and distance to centroid as node features yielded a slight performance increase with 0.76 weighted F1 score, and more comprehensive attention visualizations. The CGAT architecture's ability to generate human-understandable attention maps can enhance trust and facilitate expert validation of model decisions. While demonstrated on dental data, CGAT is broadly applicable to graph-based classification and regression tasks, promoting wider adoption of transparent and competitive deep learning models in high-stakes environments.
Authors: Andrey Sakhovskiy, Elena Tutubalina
Abstract: In recent years, there has been substantial progress in using pretrained Language Models (LMs) on a range of tasks aimed at improving the understanding of biomedical texts. Nonetheless, existing biomedical LLMs show limited comprehension of complex, domain-specific concept structures and the factual information encoded in biomedical Knowledge Graphs (KGs). In this work, we propose BALI (Biomedical Knowledge Graph and Language Model Alignment), a novel joint LM and KG pre-training method that augments an LM with external knowledge by the simultaneous learning of a dedicated KG encoder and aligning the representations of both the LM and the graph. For a given textual sequence, we link biomedical concept mentions to the Unified Medical Language System (UMLS) KG and utilize local KG subgraphs as cross-modal positive samples for these mentions. Our empirical findings indicate that implementing our method on several leading biomedical LMs, such as PubMedBERT and BioLinkBERT, improves their performance on a range of language understanding tasks and the quality of entity representations, even with minimal pre-training on a small alignment dataset sourced from PubMed scientific abstracts.
Authors: Gavin Tao, Yinuo Wang, Jinzhao Zhou
Abstract: End-to-end reinforcement learning for motion control promises unified perception-action policies that scale across embodiments and tasks, yet most deployed controllers are either blind (proprioception-only) or rely on fusion backbones with unfavorable compute-memory trade-offs. Recurrent controllers struggle with long-horizon credit assignment, and Transformer-based fusion incurs quadratic cost in token length, limiting temporal and spatial context. We present a vision-driven cross-modal RL framework built on SSD-Mamba2, a selective state-space backbone that applies state-space duality (SSD) to enable both recurrent and convolutional scanning with hardware-aware streaming and near-linear scaling. Proprioceptive states and exteroceptive observations (e.g., depth tokens) are encoded into compact tokens and fused by stacked SSD-Mamba2 layers. The selective state-space updates retain long-range dependencies with markedly lower latency and memory use than quadratic self-attention, enabling longer look-ahead, higher token resolution, and stable training under limited compute. Policies are trained end-to-end under curricula that randomize terrain and appearance and progressively increase scene complexity. A compact, state-centric reward balances task progress, energy efficiency, and safety. Across diverse motion-control scenarios, our approach consistently surpasses strong state-of-the-art baselines in return, safety (collisions and falls), and sample efficiency, while converging faster at the same compute budget. These results suggest that SSD-Mamba2 provides a practical fusion backbone for scalable, foresightful, and efficient end-to-end motion control.
Authors: Mehdi Bejani, Marco Mauri, Daniele Acconcia, Simone Todaro, Stefano Mariani
Abstract: This paper presents an innovative Transformer-based deep learning strategy for optimizing the placement of sensors aiming at structural health monitoring of semiconductor probe cards. Failures in probe cards, including substrate cracks and loosened screws, would critically affect semiconductor manufacturing yield and reliability. Some failure modes could be detected by equipping a probe card with adequate sensors. Frequency response functions from simulated failure scenarios are adopted within a finite element model of a probe card. A comprehensive dataset, enriched by physics-informed scenario expansion and physics-aware statistical data augmentation, is exploited to train a hybrid Convolutional Neural Network and Transformer model. The model achieves high accuracy (99.83%) in classifying the probe card health states (baseline, loose screw, crack) and an excellent crack detection recall (99.73%). Model robustness is confirmed through a rigorous framework of 3 repetitions of 10-fold stratified cross-validation. The attention mechanism also pinpoints critical sensor locations: an analysis of the attention weights offers actionable insights for designing efficient, cost-effective monitoring systems by optimizing sensor configurations. This research highlights the capability of attention-based deep learning to advance proactive maintenance, enhancing operational reliability and yield in semiconductor manufacturing.
Authors: Ali Nawaz, Amir Ahmad, Shehroz S. Khan
Abstract: Class imbalance poses a significant challenge to supervised classification, particularly in critical domains like medical diagnostics and anomaly detection where minority class instances are rare. While numerous studies have explored rebalancing techniques to address this issue, less attention has been given to evaluating the performance of binary classifiers under imbalance when no such techniques are applied. Therefore, the goal of this study is to assess the performance of binary classifiers "as-is", without performing any explicit rebalancing. Specifically, we systematically evaluate the robustness of a diverse set of binary classifiers across both real-world and synthetic datasets, under progressively reduced minority class sizes, using one-shot and few-shot scenarios as baselines. Our approach also explores varying data complexities through synthetic decision boundary generation to simulate real-world conditions. In addition to standard classifiers, we include experiments using undersampling, oversampling strategies, and one-class classification (OCC) methods to examine their behavior under severe imbalance. The results confirm that classification becomes more difficult as data complexity increases and the minority class size decreases. While traditional classifiers deteriorate under extreme imbalance, advanced models like TabPFN and boosting-based ensembles retain relatively higher performance and better generalization compared to traditional classifiers. Visual interpretability and evaluation metrics further validate these findings. Our work offers valuable guidance on model selection for imbalanced learning, providing insights into classifier robustness without dependence on explicit rebalancing techniques.
Authors: Daniel Hein, Simon Wiedemann, Markus Baumann, Patrik Felbinger, Justin Klein, Maximilian Schieder, Jonas Stein, Dani\"elle Schuman, Thomas Cope, Steffen Udluft
Abstract: Quantum policy evaluation (QPE) is a reinforcement learning (RL) algorithm which is quadratically more efficient than an analogous classical Monte Carlo estimation. It makes use of a direct quantum mechanical realization of a finite Markov decision process, in which the agent and the environment are modeled by unitary operators and exchange states, actions, and rewards in superposition. Previously, the quantum environment has been implemented and parametrized manually for an illustrative benchmark using a quantum simulator. In this paper, we demonstrate how these environment parameters can be learned from a batch of classical observational data through quantum machine learning (QML) on quantum hardware. The learned quantum environment is then applied in QPE to also compute policy evaluations on quantum hardware. Our experiments reveal that, despite challenges such as noise and short coherence times, the integration of QML and QPE shows promising potential for achieving quantum advantage in RL.
Authors: Lukas Schulte, Daniel Hein, Steffen Udluft, Thomas A. Runkler
Abstract: This paper explores the application of variational quantum circuits (VQCs) for solving offline contextual bandit problems in industrial optimization tasks. Using the Industrial Benchmark (IB) environment, we evaluate the performance of quantum regression models against classical models. Our findings demonstrate that quantum models can effectively fit complex reward functions, identify optimal configurations via particle swarm optimization (PSO), and generalize well in noisy and sparse datasets. These results provide a proof of concept for utilizing VQCs in offline contextual bandit problems and highlight their potential in industrial optimization tasks.
Authors: Kamel Kamel, Hridoy Sankar Dutta, Keshav Sood, Sunil Aryal
Abstract: Voice Authentication Systems (VAS) use unique vocal characteristics for verification. They are increasingly integrated into high-security sectors such as banking and healthcare. Despite their improvements using deep learning, they face severe vulnerabilities from sophisticated threats like deepfakes and adversarial attacks. The emergence of realistic voice cloning complicates detection, as systems struggle to distinguish authentic from synthetic audio. While anti-spoofing countermeasures (CMs) exist to mitigate these risks, many rely on static detection models that can be bypassed by novel adversarial methods, leaving a critical security gap. To demonstrate this vulnerability, we propose the Spectral Masking and Interpolation Attack (SMIA), a novel method that strategically manipulates inaudible frequency regions of AI-generated audio. By altering the voice in imperceptible zones to the human ear, SMIA creates adversarial samples that sound authentic while deceiving CMs. We conducted a comprehensive evaluation of our attack against state-of-the-art (SOTA) models across multiple tasks, under simulated real-world conditions. SMIA achieved a strong attack success rate (ASR) of at least 82% against combined VAS/CM systems, at least 97.5% against standalone speaker verification systems, and 100% against countermeasures. These findings conclusively demonstrate that current security postures are insufficient against adaptive adversarial attacks. This work highlights the urgent need for a paradigm shift toward next-generation defenses that employ dynamic, context-aware frameworks capable of evolving with the threat landscape.
Authors: Alvaro Becerra, Ruth Cobos, Charles Lang
Abstract: In modern online learning, understanding and predicting student behavior is crucial for enhancing engagement and optimizing educational outcomes. This systematic review explores the integration of biosensors and Multimodal Learning Analytics (MmLA) to analyze and predict student behavior during computer-based learning sessions. We examine key challenges, including emotion and attention detection, behavioral analysis, experimental design, and demographic considerations in data collection. Our study highlights the growing role of physiological signals, such as heart rate, brain activity, and eye-tracking, combined with traditional interaction data and self-reports to gain deeper insights into cognitive states and engagement levels. We synthesize findings from 54 key studies, analyzing commonly used methodologies such as advanced machine learning algorithms and multimodal data pre-processing techniques. The review identifies current research trends, limitations, and emerging directions in the field, emphasizing the transformative potential of biosensor-driven adaptive learning systems. Our findings suggest that integrating multimodal data can facilitate personalized learning experiences, real-time feedback, and intelligent educational interventions, ultimately advancing toward a more customized and adaptive online learning experience.
Authors: Friedrich Wolf-Monheim
Abstract: Next to decision tree and k-nearest neighbours algorithms deep convolutional neural networks (CNNs) are widely used to classify audio data in many domains like music, speech or environmental sounds. To train a specific CNN various spectral and rhythm features like mel-scaled spectrograms, mel-frequency cepstral coefficients (MFCC), cyclic tempograms, short-time Fourier transform (STFT) chromagrams, constant-Q transform (CQT) chromagrams and chroma energy normalized statistics (CENS) chromagrams can be used as digital image input data for the neural network. The performance of these spectral and rhythm features for audio category level as well as audio class level classification is investigated in detail with a deep CNN and the ESC-50 dataset with 2,000 labeled environmental audio recordings using an end-to-end deep learning pipeline. The evaluated metrics accuracy, precision, recall and F1 score for multiclass classification clearly show that the mel-scaled spectrograms and the mel-frequency cepstral coefficients (MFCC) perform significantly better then the other spectral and rhythm features investigated in this research for audio classification tasks using deep CNNs.
Authors: Mikel Robredo, Matteo Esposito, Fabio Palomba, Rafael Pe\~naloza, Valentina Lenarduzzi
Abstract: Context. Code refactoring improves software quality without changing external behavior. Despite its advantages, its benefits are hindered by the considerable cost of time, resources, and continuous effort it demands. Aim. Understanding why developers refactor, and which metrics capture these motivations, may support wider and more effective use of refactoring in practice. Method. We performed a large-scale empirical study to analyze developers refactoring activity, leveraging Large Language Models (LLMs) to identify underlying motivations from version control data, comparing our findings with previous motivations reported in the literature. Results. LLMs matched human judgment in 80% of cases, but aligned with literature-based motivations in only 47%. They enriched 22% of motivations with more detailed rationale, often highlighting readability, clarity, and structural improvements. Most motivations were pragmatic, focused on simplification and maintainability. While metrics related to developer experience and code readability ranked highest, their correlation with motivation categories was weak. Conclusions. We conclude that LLMs effectively capture surface-level motivations but struggle with architectural reasoning. Their value lies in providing localized explanations, which, when combined with software metrics, can form hybrid approaches. Such integration offers a promising path toward prioritizing refactoring more systematically and balancing short-term improvements with long-term architectural goals.
Authors: Michele Joshua Maggini, Dhia Merzougui, Rabiraj Bandyopadhyay, Ga\"el Dias, Fabrice Maurel, Pablo Gamallo
Abstract: The spread of fake news, polarizing, politically biased, and harmful content on online platforms has been a serious concern. With large language models becoming a promising approach, however, no study has properly benchmarked their performance across different models, usage methods, and languages. This study presents a comprehensive overview of different Large Language Models adaptation paradigms for the detection of hyperpartisan and fake news, harmful tweets, and political bias. Our experiments spanned 10 datasets and 5 different languages (English, Spanish, Portuguese, Arabic and Bulgarian), covering both binary and multiclass classification scenarios. We tested different strategies ranging from parameter efficient Fine-Tuning of language models to a variety of different In-Context Learning strategies and prompts. These included zero-shot prompts, codebooks, few-shot (with both randomly-selected and diversely-selected examples using Determinantal Point Processes), and Chain-of-Thought. We discovered that In-Context Learning often underperforms when compared to Fine-Tuning a model. This main finding highlights the importance of Fine-Tuning even smaller models on task-specific settings even when compared to the largest models evaluated in an In-Context Learning setup - in our case LlaMA3.1-8b-Instruct, Mistral-Nemo-Instruct-2407 and Qwen2.5-7B-Instruct.
Authors: Christian Gapp, Elias Tappeiner, Martin Welk, Karl Fritscher, Stephanie Mangesius, Constantin Eisenschink, Philipp Deisl, Michael Knoflach, Astrid E. Grams, Elke R. Gizewski, Rainer Schubert
Abstract: Stroke is the second most frequent cause of death world wide with an annual mortality of around 5.5 million. Recurrence rates of stroke are between 5 and 25% in the first year. As mortality rates for relapses are extraordinarily high (40%) it is of utmost importance to reduce the recurrence rates. We address this issue by detecting patients at risk of stroke recurrence at an early stage in order to enable appropriate therapy planning. To this end we collected 3D intracranial CTA image data and recorded concomitant heart diseases, the age and the gender of stroke patients between 2010 and 2024. We trained single- and multimodal deep learning based neural networks for binary relapse detection (Task 1) and for relapse free survival (RFS) time prediction together with a subsequent classification (Task 2). The separation of relapse from non-relapse patients (Task 1) could be solved with tabular data (AUC on test dataset: 0.84). However, for the main task, the regression (Task 2), our multimodal XSRD-net processed the modalities vision:tabular with 0.68:0.32 according to modality contribution measures. The c-index with respect to relapses for the multimodal model reached 0.68, and the AUC is 0.71 for the test dataset. Final, deeper interpretability analysis results could highlight a link between both heart diseases (tabular) and carotid arteries (vision) for the detection of relapses and the prediction of the RFS time. This is a central outcome that we strive to strengthen with ongoing data collection and model retraining.
Authors: Crispin Cooper, Ana Fredrich, Tommaso Reggiani, Wouter Poortinga
Abstract: How should well-being be prioritised in society, and what trade-offs are people willing to make between fairness and personal well-being? We investigate these questions using a stated preference experiment with a nationally representative UK sample (n = 300), in which participants evaluated life satisfaction outcomes for both themselves and others under conditions of uncertainty. Individual-level utility functions were estimated using an Expected Utility Maximisation (EUM) framework and tested for sensitivity to the overweighting of small probabilities, as characterised by Cumulative Prospect Theory (CPT). A majority of participants displayed concave (risk-averse) utility curves and showed stronger aversion to inequality in societal life satisfaction outcomes than to personal risk. These preferences were unrelated to political alignment, suggesting a shared normative stance on fairness in well-being that cuts across ideological boundaries. The results challenge use of average life satisfaction as a policy metric, and support the development of nonlinear utility-based alternatives that more accurately reflect collective human values. Implications for public policy, well-being measurement, and the design of value-aligned AI systems are discussed.
Authors: S M Asiful Islam Saky, Ugyen Tshering
Abstract: Optical Coherence Tomography (OCT) is essential for diagnosing conditions such as glaucoma, diabetic retinopathy, and age-related macular degeneration. Accurate retinal layer segmentation enables quantitative biomarkers critical for clinical decision-making, but manual segmentation is time-consuming and variable, while conventional deep learning models often lack interpretability. This work proposes an improved SegNet-based deep learning framework for automated and interpretable retinal layer segmentation. Architectural innovations, including modified pooling strategies, enhance feature extraction from noisy OCT images, while a hybrid loss function combining categorical cross-entropy and Dice loss improves performance for thin and imbalanced retinal layers. Gradient-weighted Class Activation Mapping (Grad-CAM) is integrated to provide visual explanations, allowing clinical validation of model decisions. Trained and validated on the Duke OCT dataset, the framework achieved 95.77% validation accuracy, a Dice coefficient of 0.9446, and a Jaccard Index (IoU) of 0.8951. Class-wise results confirmed robust performance across most layers, with challenges remaining for thinner boundaries. Grad-CAM visualizations highlighted anatomically relevant regions, aligning segmentation with clinical biomarkers and improving transparency. By combining architectural improvements, a customized hybrid loss, and explainable AI, this study delivers a high-performing SegNet-based framework that bridges the gap between accuracy and interpretability. The approach offers strong potential for standardizing OCT analysis, enhancing diagnostic efficiency, and fostering clinical trust in AI-driven ophthalmic tools.
Authors: Jonathan Teagan
Abstract: This study applies a range of forecasting techniques,including ARIMA, Prophet, Long Short Term Memory networks (LSTM), Temporal Convolutional Networks (TCN), and XGBoost, to model and predict Russian equipment losses during the ongoing war in Ukraine. Drawing on daily and monthly open-source intelligence (OSINT) data from WarSpotting, we aim to assess trends in attrition, evaluate model performance, and estimate future loss patterns through the end of 2025. Our findings show that deep learning models, particularly TCN and LSTM, produce stable and consistent forecasts, especially under conditions of high temporal granularity. By comparing different model architectures and input structures, this study highlights the importance of ensemble forecasting in conflict modeling, and the value of publicly available OSINT data in quantifying material degradation over time.
Authors: Mihai Nadas, Laura Diosan, Andreea Tomescu, Andrei Piscoran
Abstract: Literary translation has recently gained attention as a distinct and complex task in machine translation research. However, the translation by small open models remains an open problem. We contribute to this ongoing research by introducing TINYFABULIST TRANSLATION FRAMEWORK (TF2), a unified framework for dataset creation, fine tuning, and evaluation in English-Romanian literary translations, centred on the creation and open release of both a compact, fine tuned language model (TF2-12B) and large scale synthetic parallel datasets (DS-TF2-EN-RO-3M and DS-TF2-EN-RO-15K). Building on DS-TF1-EN-3M (TF1), the largest collection of synthetic English fables to date, we address the need for rich, high quality literary datasets in low resource languages such as Romanian. Our pipeline first generates 15k high quality Romanian references from the TF1 pool using a high performing LLM. We then apply a two stage fine tuning process to a 12B parameter open weight model: (i) instruction tuning to capture genre specific narrative style, and (ii) adapter compression for efficient deployment. Evaluation combines corpus level BLEU and a five dimension LLM based rubric (accuracy, fluency, coherence, style, cultural adaptation) to provide a nuanced assessment of translation quality. Results show that our fine tuned model achieves fluency and adequacy competitive with top performing large proprietary models, while being open, accessible, and significantly more cost effective. Alongside the fine tuned model and both datasets, we publicly release all scripts and evaluation prompts. TF2 thus provides an end-to-end, reproducible pipeline for research on cost efficient translation, cross lingual narrative generation, and the broad adoption of open models for culturally significant literary content in low resource settings.
Authors: Seyd Teymoor Seydi
Abstract: Accurate and timely mapping of burned areas is crucial for environmental monitoring, disaster management, and assessment of climate change. This study presents a novel approach to automated burned area mapping using the AlphaEArth dataset combined with the Siamese U-Net deep learning architecture. The AlphaEArth Dataset, comprising high-resolution optical and thermal infrared imagery with comprehensive ground-truth annotations, provides an unprecedented resource for training robust burned area detection models. We trained our model with the Monitoring Trends in Burn Severity (MTBS) dataset in the contiguous US and evaluated it with 17 regions cross in Europe. Our experimental results demonstrate that the proposed ensemble approach achieves superior performance with an overall accuracy of 95%, IoU of 0.6, and F1-score of 74% on the test dataset. The model successfully identifies burned areas across diverse ecosystems with complex background, showing particular strength in detecting partially burned vegetation and fire boundaries and its transferability and high generalization in burned area mapping. This research contributes to the advancement of automated fire damage assessment and provides a scalable solution for global burn area monitoring using the AlphaEarth dataset.
Authors: Daniel DeAlcala, Aythami Morales, Julian Fierrez, Gonzalo Mancera, Ruben Tolosana, Javier Ortega-Garcia
Abstract: Active Membership Inference Test (aMINT) is a method designed to detect whether given data were used during the training of machine learning models. In Active MINT, we propose a novel multitask learning process that involves training simultaneously two models: the original or Audited Model, and a secondary model, referred to as the MINT Model, responsible for identifying the data used for training the Audited Model. This novel multi-task learning approach has been designed to incorporate the auditability of the model as an optimization objective during the training process of neural networks. The proposed approach incorporates intermediate activation maps as inputs to the MINT layers, which are trained to enhance the detection of training data. We present results using a wide range of neural networks, from lighter architectures such as MobileNet to more complex ones such as Vision Transformers, evaluated in 5 public benchmarks. Our proposed Active MINT achieves over 80% accuracy in detecting if given data was used for training, significantly outperforming previous approaches in the literature. Our aMINT and related methodological developments contribute to increasing transparency in AI models, facilitating stronger safeguards in AI deployments to achieve proper security, privacy, and copyright protection.
Authors: Arun Verma, Zhaoxuan Wu, Zijian Zhou, Xiaoqiang Lin, Zhiliang Chen, Rachael Hwee Ling Sim, Rui Qiao, Jingtan Wang, Nhung Bui, Xinyuan Niu, Wenyang Hu, Gregory Kang Ruey Lau, Zi-Yu Khoo, Zitong Zhao, Xinyi Xu, Apivich Hemachandra, See-Kiong Ng, Bryan Kian Hsiang Low
Abstract: Large Language Models (LLMs) are large-scale pretrained models that have achieved remarkable success across diverse domains. These successes have been driven by unprecedented complexity and scale in both data and computations. However, due to the high costs of training such models, brute-force trial-and-error approaches to improve LLMs are not feasible. Inspired by the success of inverse problems in uncovering fundamental scientific laws, this position paper advocates that inverse problems can also efficiently uncover scaling laws that guide the building of LLMs to achieve the desirable performance with significantly better cost-effectiveness.
Authors: Moo Hyun Son, Juyoung Bae, Zelin Qiu, Jiale Peng, Kai Xin Li, Yifan Lin, Hao Chen
Abstract: Digital dentistry represents a transformative shift in modern dental practice. The foundational step in this transformation is the accurate digital representation of the patient's dentition, which is obtained from segmented Cone-Beam Computed Tomography (CBCT) and Intraoral Scans (IOS). Despite the growing interest in digital dental technologies, existing segmentation methodologies frequently lack rigorous validation and demonstrate limited performance and clinical applicability. To the best of our knowledge, this is the first work to introduce a multimodal pretraining framework for tooth segmentation. We present ToothMCL, a Tooth Multimodal Contrastive Learning for pretraining that integrates volumetric (CBCT) and surface-based (IOS) modalities. By capturing modality-invariant representations through multimodal contrastive learning, our approach effectively models fine-grained anatomical features, enabling precise multi-class segmentation and accurate identification of F\'ed\'eration Dentaire Internationale (FDI) tooth numbering. Along with the framework, we curated CBCT-IOS3.8K, the largest paired CBCT and IOS dataset to date, comprising 3,867 patients. We then evaluated ToothMCL on a comprehensive collection of independent datasets, representing the largest and most diverse evaluation to date. Our method achieves state-of-the-art performance in both internal and external testing, with an increase of 12\% for CBCT segmentation and 8\% for IOS segmentation in the Dice Similarity Coefficient (DSC). Furthermore, ToothMCL consistently surpasses existing approaches in tooth groups and demonstrates robust generalizability across varying imaging conditions and clinical scenarios.
Authors: Tuo Wang, Adithya Kulkarni, Tyler Cody, Peter A. Beling, Yujun Yan, Dawei Zhou
Abstract: Uncertainty estimation is essential for enhancing the reliability of Large Language Models (LLMs), particularly in high-stakes applications. Existing methods often overlook semantic dependencies, relying on token-level probability measures that fail to capture structural relationships within the generated text. We propose GENUINE: Graph ENhanced mUlti-level uncertaINty Estimation for Large Language Models, a structure-aware framework that leverages dependency parse trees and hierarchical graph pooling to refine uncertainty quantification. By incorporating supervised learning, GENUINE effectively models semantic and structural relationships, improving confidence assessments. Extensive experiments across NLP tasks show that GENUINE achieves up to 29% higher AUROC than semantic entropy-based approaches and reduces calibration errors by over 15%, demonstrating the effectiveness of graph-based uncertainty modeling. The code is available at https://github.com/ODYSSEYWT/GUQ.
Authors: Mahmudul Islam Masum, Miad Islam, Arif I. Sarwat
Abstract: As local AI grows in popularity, there is a critical gap between the benchmark performance of object detectors and their practical viability on consumer-grade hardware. While models like YOLOv10s promise real-time speeds, these metrics are typically achieved on high-power, desktop-class GPUs. This paper reveals that on resource-constrained systems, such as laptops with RTX 4060 GPUs, performance is not compute-bound but is instead dominated by system-level bottlenecks, as illustrated by a simple bottleneck test. To overcome this hardware-level constraint, we introduce a Two-Pass Adaptive Inference algorithm, a model-independent approach that requires no architectural changes. This study mainly focuses on adaptive inference strategies and undertakes a comparative analysis of architectural early-exit and resolution-adaptive routing, highlighting their respective trade-offs within a unified evaluation framework. The system uses a fast, low-resolution pass and only escalates to a high-resolution model pass when detection confidence is low. On a 5000-image COCO dataset, our method achieves a 1.85x speedup over a PyTorch Early-Exit baseline, with a modest mAP loss of 5.51%. This work provides a practical and reproducible blueprint for deploying high-performance, real-time AI on consumer-grade devices by shifting the focus from pure model optimization to hardware-aware inference strategies that maximize throughput.
Authors: Wanni Vidulige Ishan Perera, Xing Liu, Fan liang, Junyi Zhang
Abstract: The rapid evolution of Artificial Intelligence (AI) and Large Language Models (LLMs) has opened up new opportunities in the area of cybersecurity, especially in the exploitation automation landscape and penetration testing. This study explores Android penetration testing automation using LLM-based tools, especially PentestGPT, to identify and execute rooting techniques. Through a comparison of the traditional manual rooting process and exploitation methods produced using AI, this study evaluates the efficacy, reliability, and scalability of automated penetration testing in achieving high-level privilege access on Android devices. With the use of an Android emulator (Genymotion) as the testbed, we fully execute both traditional and exploit-based rooting methods, automating the process using AI-generated scripts. Secondly, we create a web application by integrating OpenAI's API to facilitate automated script generation from LLM-processed responses. The research focuses on the effectiveness of AI-enabled exploitation by comparing automated and manual penetration testing protocols, by determining LLM weaknesses and strengths along the way. We also provide security suggestions of AI-enabled exploitation, including ethical factors and potential misuse. The findings exhibit that while LLMs can significantly streamline the workflow of exploitation, they need to be controlled by humans to ensure accuracy and ethical application. This study adds to the increasing body of literature on AI-powered cybersecurity and its effect on ethical hacking, security research, and mobile device security.
Authors: Kai Ye, Liangcai Su, Chenxiong Qian
Abstract: Code generation has emerged as a pivotal capability of Large Language Models(LLMs), revolutionizing development efficiency for programmers of all skill levels. However, the complexity of data structures and algorithmic logic often results in functional deficiencies and security vulnerabilities in generated code, reducing it to a prototype requiring extensive manual debugging. While Retrieval-Augmented Generation (RAG) can enhance correctness and security by leveraging external code manuals, it simultaneously introduces new attack surfaces. In this paper, we pioneer the exploration of attack surfaces in Retrieval-Augmented Code Generation (RACG), focusing on malicious dependency hijacking. We demonstrate how poisoned documentation containing hidden malicious dependencies (e.g., matplotlib_safe) can subvert RACG, exploiting dual trust chains: LLM reliance on RAG and developers' blind trust in LLM suggestions. To construct poisoned documents, we propose ImportSnare, a novel attack framework employing two synergistic strategies: 1)Position-aware beam search optimizes hidden ranking sequences to elevate poisoned documents in retrieval results, and 2)Multilingual inductive suggestions generate jailbreaking sequences to manipulate LLMs into recommending malicious dependencies. Through extensive experiments across Python, Rust, and JavaScript, ImportSnare achieves significant attack success rates (over 50% for popular libraries such as matplotlib and seaborn) in general, and is also able to succeed even when the poisoning ratio is as low as 0.01%, targeting both custom and real-world malicious packages. Our findings reveal critical supply chain risks in LLM-powered development, highlighting inadequate security alignment for code generation tasks. To support future research, we will release the multilingual benchmark suite and datasets. The project homepage is https://importsnare.github.io.
Authors: Kasra Borazjani, Naji Khosravan, Rajeev Sahay, Bita Akram, Seyyedali Hosseinalipour
Abstract: Multi-modal multi-task (M3T) foundation models (FMs) have recently shown transformative potential in artificial intelligence, with emerging applications in education. However, their deployment in real-world educational settings is hindered by privacy regulations, data silos, and limited domain-specific data availability. We introduce M3T Federated Foundation Models (FedFMs) for education: a paradigm that integrates federated learning (FL) with M3T FMs to enable collaborative, privacy-preserving training across decentralized institutions while accommodating diverse modalities and tasks. Subsequently, this position paper aims to unveil M3T FedFMs as a promising yet underexplored approach to the education community, explore its potentials, and reveal its related future research directions. We outline how M3T FedFMs can advance three critical pillars of next-generation intelligent education systems: (i) privacy preservation, by keeping sensitive multi-modal student and institutional data local; (ii) personalization, through modular architectures enabling tailored models for students, instructors, and institutions; and (iii) equity and inclusivity, by facilitating participation from underrepresented and resource-constrained entities. We finally identify various open research challenges, including studying of (i) inter-institution heterogeneous privacy regulations, (ii) the non-uniformity of data modalities' characteristics, (iii) the unlearning approaches for M3T FedFMs, (iv) the continual learning frameworks for M3T FedFMs, and (v) M3T FedFM model interpretability, which must be collectively addressed for practical deployment.
Authors: Oliver Daniels, Stuart Armstrong, Alexandre Maranh\~ao, Mahirah Fairuz Rahman, Benjamin M. Marlin, Rebecca Gorman
Abstract: Deep neural networks are notoriously sensitive to spurious correlations - where a model learns a shortcut that fails out-of-distribution. Existing work on spurious correlations has often focused on incomplete correlations,leveraging access to labeled instances that break the correlation. But in cases where the spurious correlations are complete, the correct generalization is fundamentally \textit{underspecified}. To resolve this underspecification, we propose learning a set of concepts that are consistent with training data but make distinct predictions on a subset of novel unlabeled inputs. Using a self-training approach that encourages \textit{confident} and \textit{selective} disagreement, our method ACE matches or outperforms existing methods on a suite of complete-spurious correlation benchmarks, while remaining robust to incomplete spurious correlations. ACE is also more configurable than prior approaches, allowing for straight-forward encoding of prior knowledge and principled unsupervised model selection. In an early application to language-model alignment, we find that ACE achieves competitive performance on the measurement tampering detection benchmark \textit{without} access to untrusted measurements. While still subject to important limitations, ACE represents significant progress towards overcoming underspecification.
Authors: Xin Lai, Junyi Li, Wei Li, Tao Liu, Tianjian Li, Hengshuang Zhao
Abstract: Recent advances in large multimodal models have leveraged image-based tools with reinforcement learning to tackle visual problems. However, existing open-source approaches often exhibit monotonous reasoning patterns and allow only a limited number of interaction turns, making them inadequate for difficult tasks that require trial-and-error exploration. In this work, we address this limitation by scaling up tool-based interactions and introduce Mini-o3, a system that executes deep, multi-turn reasoning -- spanning tens of steps -- and achieves state-of-the-art performance on challenging visual search tasks. Our recipe for reproducing OpenAI o3-style behaviors comprises three key components. First, we construct the Visual Probe Dataset, a collection of thousands of challenging visual search problems designed for exploratory reasoning. Second, we develop an iterative data collection pipeline to obtain cold-start trajectories that exhibit diverse reasoning patterns, including depth-first search, trial-and-error, and goal maintenance. Third, we propose an over-turn masking strategy that prevents penalization of over-turn responses (those that hit the maximum number of turns) during reinforcement learning, thereby balancing training-time efficiency with test-time scalability. Despite training with an upper bound of only six interaction turns, our model generates trajectories that naturally scale to tens of turns at inference time, with accuracy improving as the number of turns increases. Extensive experiments demonstrate that Mini-o3 produces rich reasoning patterns and deep thinking paths, effectively solving challenging visual search problems.
Authors: Gustavo Assun\c{c}\~ao, Miguel Castelo-Branco, Paulo Menezes
Abstract: Background: Exploration of the physical environment is an indispensable precursor to information acquisition and knowledge consolidation for living organisms. Yet, current artificial intelligence models lack these autonomy capabilities during training, hindering their adaptability. This work proposes a learning framework for artificial agents to obtain an intrinsic exploratory drive, based on epistemic and achievement emotions triggered during data observation. Methods: This study proposes a dual-module reinforcement framework, where data analysis scores dictate pride or surprise, in accordance with psychological studies on humans. A correlation between these states and exploration is then optimized for agents to meet their learning goals. Results: Causal relationships between states and exploration are demonstrated by the majority of agents. A 15.4\% mean increase is noted for surprise, with a 2.8\% mean decrease for pride. Resulting correlations of $\rho_{surprise}=0.461$ and $\rho_{pride}=-0.237$ are obtained, mirroring previously reported human behavior. Conclusions: These findings lead to the conclusion that bio-inspiration for AI development can be of great use. This can incur benefits typically found in living beings, such as autonomy. Further, it empirically shows how AI methodologies can corroborate human behavioral findings, showcasing major interdisciplinary importance. Ramifications are discussed.
Authors: Zhiwei Wang, Yunji Wang, Zhongwang Zhang, Zhangchen Zhou, Hui Jin, Tianyang Hu, Jiacheng Sun, Zhenguo Li, Yaoyu Zhang, Zhi-Qin John Xu
Abstract: Large language models have consistently struggled with complex reasoning tasks, such as mathematical problem-solving. Investigating the internal reasoning mechanisms of these models can help us design better model architectures and training strategies, ultimately enhancing their reasoning capability. In this study, we constructed a symbolic multi-step reasoning task to investigate the information propagation mechanisms in Transformer models when solving the task through direct answering and Chain-of-Thought (CoT) reasoning. We introduced the concept of buffer mechanism: the model stores various information in distinct buffers and selectively extracts it through the query-key matrix. We proposed a random matrix-based algorithm to enhance the model's reasoning ability. This algorithm introduces only 132 trainable parameters, yet leads to significant performance improvements on 7 multi-step reasoning datasets, including PrOntoQA, LogicAsker, and LogicInference. These findings provide new insights into understanding the large language models.
Authors: Junjie Wang, Yuxiang Zhang, Minghao Liu, Yin Zhang, Yatai Ji, Weihao Xuan, Nie Lin, Kang Zhu, Zhiqiang Lin, Yiming Ren, Chunyang Jiang, Yiyao Yu, Zekun Wang, Tiezhen Wang, Wenhao Huang, Jie Fu, Qunshu Lin, Yujiu Yang, Ge Zhang, Ruibin Yuan, Bei Chen, Wenhu Chen
Abstract: Recent advancements in large multimodal models (LMMs) have leveraged extensive multimodal datasets to enhance capabilities in complex knowledge-driven tasks. However, persistent challenges in perceptual and reasoning errors limit their efficacy, particularly in interpreting intricate visual data and deducing multimodal relationships. To address these issues, we introduce PIN (Paired and INterleaved multimodal documents), a novel data format designed to foster a deeper integration of visual and textual knowledge. The PIN format uniquely combines semantically rich Markdown files, which preserve fine-grained textual structures, with holistic overall images that capture the complete document layout. Following this format, we construct and release two large-scale, open-source datasets: PIN-200M (~200 million documents) and PIN-14M (~14 million), compiled from diverse web and scientific sources in both English and Chinese. To maximize usability, we provide detailed statistical analyses and equip the datasets with quality signals, enabling researchers to easily filter and select data for specific tasks. Our work provides the community with a versatile data format and substantial resources, offering a foundation for new research in pre-training strategies and the development of more powerful knowledge-intensive LMMs.
Authors: Timothy Ossowski, Jixuan Chen, Danyal Maqbool, Zefan Cai, Tyler Bradshaw, Junjie Hu
Abstract: The rapid advances of multimodal agents built on large foundation models have largely overlooked their potential for language-based communication between agents in collaborative tasks. This oversight presents a critical gap in understanding their effectiveness in real-world deployments, particularly when communicating with humans. Existing agentic benchmarks fail to address key aspects of inter-agent communication and collaboration, particularly in scenarios where agents have unequal access to information and must work together to achieve tasks beyond the scope of individual capabilities. To fill this gap, we introduce COMMA: a novel puzzle benchmark designed to evaluate the collaborative performance of multimodal multi-agent systems through language communication. Our benchmark features a variety of multimodal puzzles, providing a comprehensive evaluation across four key categories of agentic capability in a communicative collaboration setting. Our findings reveal surprising weaknesses in state-of-the-art models, including strong proprietary models like GPT-4o and reasoning models like o4-mini. Many chain of thought reasoning models such as R1-Onevision and LLaVA-CoT struggle to outperform even a random baseline in agent-agent collaboration, indicating a potential growth area in their communication abilities.
Authors: Nasim Borazjanizadeh, Roei Herzig, Eduard Oks, Trevor Darrell, Rogerio Feris, Leonid Karlinsky
Abstract: Human reasoning relies on constructing and manipulating mental models -- simplified internal representations of situations that we use to understand and solve problems. Conceptual diagrams (e.g., a sketch drawn by a human to aid reasoning) externalize these mental models, abstracting irrelevant details to efficiently capture how entities interact with each other. In contrast, Large Language Models (LLMs) and Large MultiModal Models (LMMs) predominantly reason through text, limiting their effectiveness in complex multi-step tasks. In this paper, we propose Visual Thinking, a zero-shot framework that enables LMMs to reason through multiple chains of (self-generated) conceptual diagrams, significantly enhancing their combinatorial planning capabilities. Our approach does not require any human initialization beyond the natural language description of the task. It integrates both textual and diagrammatic reasoning within an optimized Graph-of-Thought inference framework, enhanced by beam search and depth-wise backtracking. Evaluated on multiple challenging PDDL planning domains, our method substantially improves LMMs' performance (e.g., GPT-4o: 35.5% -> 90.2% in Blocksworld) and consistently outperforms other text-only search-based inference methods. On more difficult planning domains with solution depths up to 40, our approach outperforms even the o1-preview reasoning model (e.g., 16 percentage points improvement in Floor Tiles). These results highlight the value of conceptual diagrams as a reasoning medium in LMMs.
Authors: Mingxuan Li, Junzhe Zhang, Elias Bareinboim
Abstract: A key task in Artificial Intelligence is learning effective policies for controlling agents in unknown environments to optimize performance measures. Off-policy learning methods, like Q-learning, allow learners to make optimal decisions based on past experiences. This paper studies off-policy learning from biased data in complex and high-dimensional domains where \emph{unobserved confounding} cannot be ruled out a priori. Building on the well-celebrated Deep Q-Network (DQN), we propose a novel deep reinforcement learning algorithm robust to confounding biases in observed data. Specifically, our algorithm attempts to find a safe policy for the worst-case environment compatible with the observations. We apply our method to twelve confounded Atari games, and find that it consistently dominates the standard DQN in all games where the observed input to the behavioral and target policies mismatch and unobserved confounders exist.
Authors: Sahiti Yerramilli, Nilay Pande, Rynaa Grover, Jayant Sravan Tamarapalli
Abstract: This paper introduces GeoChain, a large-scale benchmark for evaluating step-by-step geographic reasoning in multimodal large language models (MLLMs). Leveraging 1.46 million Mapillary street-level images, GeoChain pairs each image with a 21-step chain-of-thought (CoT) question sequence (over 30 million Q&A pairs). These sequences guide models from coarse attributes to fine-grained localization across four reasoning categories - visual, spatial, cultural, and precise geolocation - annotated by difficulty. Images are also enriched with semantic segmentation (150 classes) and a visual locatability score. Our benchmarking of contemporary MLLMs (GPT-4.1 variants, Claude 3.7, Gemini 2.5 variants) on a diverse 2,088-image subset reveals consistent challenges: models frequently exhibit weaknesses in visual grounding, display erratic reasoning, and struggle to achieve accurate localization, especially as the reasoning complexity escalates. GeoChain offers a robust diagnostic methodology, critical for fostering significant advancements in complex geographic reasoning within MLLMs.
Authors: Yao Yan
Abstract: Multi-digit addition is a clear probe of the computational power of large language models. To dissect the internal arithmetic processes in LLaMA-3-8B-Instruct, we combine linear probing with logit-lens inspection. Inspired by the step-by-step manner in which humans perform addition, we propose and analyze a coherent four-stage trajectory in the forward pass:Formula-structure representations become linearly decodable first, while the answer token is still far down the candidate list.Core computational features then emerge prominently.At deeper activation layers, numerical abstractions of the result become clearer, enabling near-perfect detection and decoding of the individual digits in the sum.Near the output, the model organizes and generates the final content, with the correct token reliably occupying the top rank.This trajectory suggests a hierarchical process that favors internal computation over rote memorization. We release our code and data to facilitate reproducibility.
Authors: Hudson de Martim
Abstract: Effectively representing legal norms for automated processing is a critical challenge, particularly in tracking the temporal evolution of their hierarchical components. While foundational conceptual frameworks like IFLA LRMoo provide a generic toolkit for bibliographic data, and encoding standards like Akoma Ntoso offer a robust syntax for legal documents, a dedicated, formal modeling pattern for granular, component-level versioning is still required. This limitation hinders the deterministic point-intime reconstruction of legal texts, a fundamental capability for reliable Legal Tech and AI applications. This paper proposes a structured, temporal modeling pattern grounded in the LRMoo ontology to address this need. Our approach models the evolution of a legal norm as a diachronic chain of F2 Expressions. We introduce a key distinction between a language-agnostic Temporal Version (TV)-a semantic snapshot of the norm's structure-and its concrete monolingual realizations, the Language Versions (LV). Both are modeled as F2 Expressions linked by the canonical R76 is derivative of property. This paradigm is applied recursively to the legal text's internal structure, representing it as a parallel hierarchy of abstract Component Works (F1) and their versioned Component Expressions (F2). Furthermore, we formalize the legislative amendment process using the F28 Expression Creation event, allowing changes to be traced from an amending act to its precise effect on the amended norm. Using the Brazilian Federal Constitution as a case study, we demonstrate how this event-centric architecture enables the precise, deterministic retrieval and reconstruction of any part of a legal text as it existed on a specific date. The model provides a robust foundation for building verifiable knowledge graphs and advanced AI tools, overcoming the limitations of current generative models.
Authors: Debodeep Banerjee, Burcu Sayin, Stefano Teso, Andrea Passerini
Abstract: Medical decision-making is a critical task, where errors can result in serious, potentially life-threatening consequences. While full automation remains challenging, hybrid frameworks that combine machine intelligence with human oversight offer a practical alternative. In this paper, we present MedGellan, a lightweight, annotation-free framework that uses a Large Language Model (LLM) to generate clinical guidance from raw medical records, which is then used by a physician to predict diagnoses. MedGellan uses a Bayesian-inspired prompting strategy that respects the temporal order of clinical data. Preliminary experiments show that the guidance generated by the LLM with MedGellan improves diagnostic performance, particularly in recall and $F_1$ score.
Authors: Thomas Eiter, Tobias Geibinger, Tobias Kaminski, Nysret Musliu, Johannes Oetsch
Abstract: We present the solver asp-fzn for Constraint Answer Set Programming (CASP), which extends ASP with linear constraints. Our approach is based on translating CASP programs into the solver-independent FlatZinc language that supports several Constraint Programming and Integer Programming backend solvers. Our solver supports a rich language of linear constraints, including some common global constraints. As for evaluation, we show that asp-fzn is competitive with state-of-the-art ASP solvers on benchmarks taken from past ASP competitions. Furthermore, we evaluate it on several CASP problems from the literature and compare its performance with clingcon, which is a prominent CASP solver that supports most of the asp-fzn language. The performance of asp-fzn is very promising as it is already competitive on plain ASP and even outperforms clingcon on some CASP benchmarks.
Authors: Jayant Sravan Tamarapalli, Rynaa Grover, Nilay Pande, Sahiti Yerramilli
Abstract: Multimodal Large Language Models (MLLMs) demonstrate remarkable fluency in understanding visual scenes, yet they exhibit a critical lack in a fundamental cognitive skill: object counting. This blind spot severely limits their reliability in real-world applications. To date, this capability has been largely unevaluated in complex scenarios, as existing benchmarks either feature sparse object densities or are confined to specific visual domains, failing to test models under realistic conditions. Addressing this gap, we introduce CountQA, a challenging new benchmark designed to probe this deficiency. Comprising over 1,500 question-answer pairs, CountQA features real-world images with high object density, clutter, and occlusion. We investigate this weakness by evaluating 15 prominent MLLMs on the CountQA benchmark and reveal that the top-performing model achieves a mere 42.9% accuracy, with performance declining as object counts rise. By providing a dedicated benchmark to diagnose and rectify this core weakness, CountQA paves the way for a new generation of MLLMs that are not only descriptively fluent but also numerically grounded and spatially aware. We will open-source the dataset and code upon paper acceptance to foster further research.
Authors: Rubing Chen, Jiaxin Wu, Jian Wang, Xulu Zhang, Wenqi Fan, Chenghua Lin, Xiao-Yong Wei, Qing Li
Abstract: The increasing demand for domain-specific evaluation of large language models (LLMs) has led to the development of numerous benchmarks. These efforts often adhere to the principle of data scaling, relying on large corpora or extensive question-answer (QA) sets to ensure broad coverage. However, the impact of corpus and QA set design on the precision and recall of domain-specific LLM performance remains poorly understood. In this paper, we argue that data scaling is not always the optimal principle for domain-specific benchmark construction. Instead, we introduce Comp-Comp, an iterative benchmarking framework grounded in the principle of comprehensiveness and compactness. Comprehensiveness ensures semantic recall by covering the full breadth of the domain, while compactness improves precision by reducing redundancy and noise. To demonstrate the effectiveness of our approach, we present a case study conducted at a well-renowned university, resulting in the creation of PolyBench, a large-scale, high-quality academic benchmark. Although this study focuses on academia, the Comp-Comp framework is domain-agnostic and readily adaptable to a wide range of specialized fields. The source code and datasets can be accessed at https://github.com/Anya-RB-Chen/COMP-COMP.
Authors: Nilay Pande, Sahiti Yerramilli, Jayant Sravan Tamarapalli, Rynaa Grover
Abstract: A key frontier for Multimodal Large Language Models (MLLMs) is the ability to perform deep mathematical and spatial reasoning directly from images, moving beyond their established success in semantic description. Mathematical surface plots provide a rigorous testbed for this capability, as they isolate the task of reasoning from the semantic noise common in natural images. To measure progress on this frontier, we introduce MaRVL-QA (Mathematical Reasoning over Visual Landscapes), a new benchmark designed to quantitatively evaluate these core reasoning skills. The benchmark comprises two novel tasks: Topological Counting, identifying and enumerating features like local maxima; and Transformation Recognition, recognizing applied geometric transformations. Generated from a curated library of functions with rigorous ambiguity filtering, our evaluation on MaRVL-QA reveals that even state-of-the-art MLLMs struggle significantly, often resorting to superficial heuristics instead of robust spatial reasoning. MaRVL-QA provides a challenging new tool for the research community to measure progress, expose model limitations, and guide the development of MLLMs with more profound reasoning abilities.
Authors: Lang Mei, Zhihan Yang, Chong Chen
Abstract: Recent studies have explored integrating Large Language Models (LLMs) with search engines to leverage both the LLMs' internal pre-trained knowledge and external information. Specially, reinforcement learning (RL) has emerged as a promising paradigm for enhancing LLM reasoning through multi-turn interactions with search engines. However, existing RL-based search agents rely on a single LLM to handle both search planning and question-answering (QA) tasks in an end-to-end manner, which limits their ability to optimize both capabilities simultaneously. In practice, sophisticated AI search systems often employ a large, frozen LLM (e.g., GPT-4, DeepSeek-R1) to ensure high-quality QA. Thus, a more effective and efficient approach is to utilize a small, trainable LLM dedicated to search planning. In this paper, we propose \textbf{AI-SearchPlanner}, a novel reinforcement learning framework designed to enhance the performance of frozen QA models by focusing on search planning. Specifically, our approach introduces three key innovations: 1) Decoupling the Architecture of the Search Planner and Generator, 2) Dual-Reward Alignment for Search Planning, and 3) Pareto Optimization of Planning Utility and Cost, to achieve the objectives. Extensive experiments on real-world datasets demonstrate that AI SearchPlanner outperforms existing RL-based search agents in both effectiveness and efficiency, while exhibiting strong generalization capabilities across diverse frozen QA models and data domains.
Authors: Yunbo Long, Liming Xu, Lukas Beckenbauer, Yuhan Liu, Alexandra Brintrup
Abstract: Recent research on Chain-of-Thought (CoT) reasoning in Large Language Models (LLMs) has demonstrated that agents can engage in \textit{complex}, \textit{multi-turn} negotiations, opening new avenues for agentic AI. However, existing LLM agents largely overlook the functional role of emotions in such negotiations, instead generating passive, preference-driven emotional responses that make them vulnerable to manipulation and strategic exploitation by adversarial counterparts. To address this gap, we present EvoEmo, an evolutionary reinforcement learning framework that optimizes dynamic emotional expression in negotiations. EvoEmo models emotional state transitions as a Markov Decision Process and employs population-based genetic optimization to evolve high-reward emotion policies across diverse negotiation scenarios. We further propose an evaluation framework with two baselines -- vanilla strategies and fixed-emotion strategies -- for benchmarking emotion-aware negotiation. Extensive experiments and ablation studies show that EvoEmo consistently outperforms both baselines, achieving higher success rates, higher efficiency, and increased buyer savings. This findings highlight the importance of adaptive emotional expression in enabling more effective LLM agents for multi-turn negotiation.
Authors: Adam Cole, Mick Grierson
Abstract: This paper presents an artistic and technical investigation into the attention mechanisms of video diffusion transformers. Inspired by early video artists who manipulated analog video signals to create new visual aesthetics, this study proposes a method for extracting and visualizing cross-attention maps in generative video models. Built on the open-source Wan model, our tool provides an interpretable window into the temporal and spatial behavior of attention in text-to-video generation. Through exploratory probes and an artistic case study, we examine the potential of attention maps as both analytical tools and raw artistic material. This work contributes to the growing field of Explainable AI for the Arts (XAIxArts), inviting artists to reclaim the inner workings of AI as a creative medium.
Authors: Jian Yang, Jiahui Wu, Li Fang, Hongchao Fan, Bianying Zhang, Huijie Zhao, Guangyi Yang, Rui Xin, Xiong You
Abstract: Transforming road network data into vector representations using deep learning has proven effective for road network analysis. However, urban road networks' heterogeneous and hierarchical nature poses challenges for accurate representation learning. Graph neural networks, which aggregate features from neighboring nodes, often struggle due to their homogeneity assumption and focus on a single structural scale. To address these issues, this paper presents MSRFormer, a novel road network representation learning framework that integrates multi-scale spatial interactions by addressing their flow heterogeneity and long-distance dependencies. It uses spatial flow convolution to extract small-scale features from large trajectory datasets, and identifies scale-dependent spatial interaction regions to capture the spatial structure of road networks and flow heterogeneity. By employing a graph transformer, MSRFormer effectively captures complex spatial dependencies across multiple scales. The spatial interaction features are fused using residual connections, which are fed to a contrastive learning algorithm to derive the final road network representation. Validation on two real-world datasets demonstrates that MSRFormer outperforms baseline methods in two road network analysis tasks. The performance gains of MSRFormer suggest the traffic-related task benefits more from incorporating trajectory data, also resulting in greater improvements in complex road network structures with up to 16% improvements compared to the most competitive baseline method. This research provides a practical framework for developing task-agnostic road network representation models and highlights distinct association patterns of the interplay between scale effects and flow heterogeneity of spatial interactions.
Authors: Xuan-Phi Nguyen, Shrey Pandit, Revanth Gangi Reddy, Austin Xu, Silvio Savarese, Caiming Xiong, Shafiq Joty
Abstract: Equipping large language models (LLMs) with complex, interleaved reasoning and tool-use capabilities has become a key focus in agentic AI research, especially with recent advances in reasoning-oriented (``thinking'') models. Such capabilities are key to unlocking a number of important applications. One such application is Deep Research (DR), which requires extensive search and reasoning over many sources. Our work in this paper focuses on the development of native Autonomous Single-Agent models for DR featuring minimal web crawling and Python tool integration. Unlike multi-agent systems, where agents take up pre-defined roles and are told what to do at each step in a static workflow, an autonomous single-agent determines its next action dynamically based on context, without manual directive. While prior work has proposed training recipes for base or instruction-tuned LLMs, we focus on continual reinforcement learning (RL) of reasoning-optimized models to further enhance agentic skills while preserving reasoning ability. Towards this end, we propose a simple RL recipe with entirely synthetic data, which we apply to various open-source LLMs. Our best variant SFR-DR-20B achieves up to 28.7% on Humanity's Last Exam benchmark. In addition, we conduct key analysis experiments to provide more insights into our methodologies.
Authors: Xiangwei Shen, Zhimin Li, Zhantao Yang, Shiyi Zhang, Yingfang Zhang, Donghao Li, Chunyu Wang, Qinglin Lu, Yansong Tang
Abstract: Recent studies have demonstrated the effectiveness of directly aligning diffusion models with human preferences using differentiable reward. However, they exhibit two primary challenges: (1) they rely on multistep denoising with gradient computation for reward scoring, which is computationally expensive, thus restricting optimization to only a few diffusion steps; (2) they often need continuous offline adaptation of reward models in order to achieve desired aesthetic quality, such as photorealism or precise lighting effects. To address the limitation of multistep denoising, we propose Direct-Align, a method that predefines a noise prior to effectively recover original images from any time steps via interpolation, leveraging the equation that diffusion states are interpolations between noise and target images, which effectively avoids over-optimization in late timesteps. Furthermore, we introduce Semantic Relative Preference Optimization (SRPO), in which rewards are formulated as text-conditioned signals. This approach enables online adjustment of rewards in response to positive and negative prompt augmentation, thereby reducing the reliance on offline reward fine-tuning. By fine-tuning the FLUX model with optimized denoising and online reward adjustment, we improve its human-evaluated realism and aesthetic quality by over 3x.
Authors: Qikai Wei, Mingzhi Yang, Jinqiang Wang, Wenwei Mao, Jiabo Xu, Huansheng Ning
Abstract: Recently, large language models (LLMs) have demonstrated their effectiveness in various natural language processing (NLP) tasks. However, the lack of tourism knowledge limits the performance of LLMs in tourist attraction presentations and travel planning. To address this challenge, we constructed a supervised fine-tuning dataset for the Chinese culture and tourism domain, named Cultour. This dataset consists of three parts: tourism knowledge base data, travelogues data, and tourism QA data. Additionally, we propose CTourLLM, a Qwen-based model supervised fine-tuned with Cultour, to improve the quality of information about attractions and travel planning. To evaluate the performance of CTourLLM, we proposed a human evaluation criterion named RRA (Relevance, Readability, Availability), and employed both automatic and human evaluation. The experimental results demonstrate that CTourLLM outperforms ChatGPT, achieving an improvement of 1.21 in BLEU-1 and 1.54 in Rouge-L, thereby validating the effectiveness of the response outcomes. Our proposed Cultour is accessible at https://github.com/mrweiqk/Cultour.
Authors: Manav Vora, Jonas Liang, Melkior Ornik
Abstract: Monotonic Partially Observable Markov Decision Processes (POMDPs), where the system state progressively decreases until a restorative action is performed, can be used to model sequential repair problems effectively. This paper considers the problem of solving budget-constrained multi-component monotonic POMDPs, where a finite budget limits the maximal number of restorative actions. For a large number of components, solving such a POMDP using current methods is computationally intractable due to the exponential growth in the state space with an increasing number of components. To address this challenge, we propose a two-step approach. Since the individual components of a budget-constrained multi-component monotonic POMDP are only connected via the shared budget, we first approximate the optimal budget allocation among these components using an approximation of each component POMDP's optimal value function which is obtained through a random forest model. Subsequently, we introduce an oracle-guided meta-trained Proximal Policy Optimization (PPO) algorithm to solve each of the independent budget-constrained single-component monotonic POMDPs. The oracle policy is obtained by performing value iteration on the corresponding monotonic Markov Decision Process (MDP). This two-step method provides scalability in solving truly massive multi-component monotonic POMDPs. To demonstrate the efficacy of our approach, we consider a real-world maintenance scenario that involves inspection and repair of an administrative building by a team of agents within a maintenance budget. Finally, we perform a computational complexity analysis for a varying number of components to show the scalability of the proposed approach.
Authors: Wei Wu, Zhuoshi Pan, Chao Wang, Liyi Chen, Yunchu Bai, Tianfu Wang, Kun Fu, Zheng Wang, Hui Xiong
Abstract: Rapid advances in Large Language Models (LLMs) have spurred demand for processing extended context sequences in contemporary applications. However, this progress faces two challenges: performance degradation due to sequence lengths out-of-distribution, and excessively long inference times caused by the quadratic computational complexity of attention. These issues limit LLMs in long-context scenarios. In this paper, we propose Dynamic Token-Level KV Cache Selection (TokenSelect), a training-free method for efficient and accurate long-context inference. TokenSelect builds upon the observation of non-contiguous attention sparsity, using QK dot products to measure per-head KV Cache criticality at token-level. By per-head soft voting mechanism, TokenSelect selectively involves a few critical KV cache tokens in attention calculation without sacrificing accuracy. To further accelerate TokenSelect, we design the Selection Cache based on observations of consecutive Query similarity and implemented the efficient Paged Dot Product Kernel, significantly reducing the selection overhead. A comprehensive evaluation of TokenSelect demonstrates up to $23.84\times$ speedup in attention computation and up to $2.28\times$ acceleration in end-to-end latency, while providing superior performance compared to state-of-the-art long-context inference methods.
Authors: Asmit Nayak, Shirley Zhang, Yash Wani, Rishabh Khandelwal, Kassem Fawaz
Abstract: Deceptive patterns in digital interfaces manipulate users into making unintended decisions, exploiting cognitive biases and psychological vulnerabilities. These patterns have become ubiquitous on various digital platforms. While efforts to mitigate deceptive patterns have emerged from legal and technical perspectives, a significant gap remains in creating usable and scalable solutions. We introduce our AutoBot framework to address this gap and help web stakeholders navigate and mitigate online deceptive patterns. AutoBot accurately identifies and localizes deceptive patterns from a screenshot of a website without relying on the underlying HTML code. AutoBot employs a two-stage pipeline that leverages the capabilities of specialized vision models to analyze website screenshots, identify interactive elements, and extract textual features. Next, using a large language model, AutoBot understands the context surrounding these elements to determine the presence of deceptive patterns. We also use AutoBot, to create a synthetic dataset to distill knowledge from 'teacher' LLMs to smaller language models. Through extensive evaluation, we demonstrate AutoBot's effectiveness in detecting deceptive patterns on the web, achieving an F1-score of 0.93 when detecting deceptive patterns, underscoring its potential as an essential tool for mitigating online deceptive patterns. We implement AutoBot, across three downstream applications targeting different web stakeholders: (1) a local browser extension providing users with real-time feedback, (2) a Lighthouse audit to inform developers of potential deceptive patterns on their sites, and (3) as a measurement tool designed for researchers and regulators.
Authors: Xianlong Wang, Hewen Pan, Hangtao Zhang, Minghui Li, Shengshan Hu, Ziqi Zhou, Lulu Xue, Peijin Guo, Aishan Liu, Leo Yu Zhang, Xiaohua Jia
Abstract: Robotic manipulation in the physical world is increasingly empowered by \textit{large language models} (LLMs) and \textit{vision-language models} (VLMs), leveraging their understanding and perception capabilities. Recently, various attacks against such robotic policies have been proposed, with backdoor attacks drawing considerable attention for their high stealth and strong persistence capabilities. However, existing backdoor efforts are limited to simulators and suffer from physical-world realization. To address this, we propose \textit{TrojanRobot}, a highly stealthy and broadly effective robotic backdoor attack in the physical world. Specifically, we introduce a module-poisoning approach by embedding a backdoor module into the modular robotic policy, enabling backdoor control over the policy's visual perception module thereby backdooring the entire robotic policy. Our vanilla implementation leverages a backdoor-finetuned VLM to serve as the backdoor module. To enhance its generalization in physical environments, we propose a prime implementation, leveraging the LVLM-as-a-backdoor paradigm and developing three types of prime attacks, \ie, \textit{permutation}, \textit{stagnation}, and \textit{intentional} attacks, thus achieving finer-grained backdoors. Extensive experiments on the UR3e manipulator with 18 task instructions using robotic policies based on four VLMs demonstrate the broad effectiveness and physical-world stealth of TrojanRobot. Our attack's video demonstrations are available via a github link https://trojanrobot.github.io.
Authors: Danrui Li, Sen Zhang, Sam S. Sohn, Kaidong Hu, Muhammad Usman, Mubbasir Kapadia
Abstract: The prototyping of computer games, particularly card games, requires extensive human effort in creative ideation and gameplay evaluation. Recent advances in Large Language Models (LLMs) offer opportunities to automate and streamline these processes. However, it remains challenging for LLMs to design novel game mechanics beyond existing databases, generate consistent gameplay environments, and develop scalable gameplay AI for large-scale evaluations. This paper addresses these challenges by introducing a comprehensive automated card game prototyping framework. The approach highlights a graph-based indexing method for generating novel game variations, an LLM-driven system for consistent game code generation validated by gameplay records, and a gameplay AI constructing method that uses an ensemble of LLM-generated heuristic functions optimized through self-play. These contributions aim to accelerate card game prototyping, reduce human labor, and lower barriers to entry for game developers. For code repo visit this http URL https://github.com/danruili/Cardiverse
Authors: Pengyu Wang, Ying Fang, Xiaofei Li
Abstract: Reverberant speech, denoting the speech signal degraded by reverberation, contains crucial knowledge of both anechoic source speech and room impulse response (RIR). This work proposes a variational Bayesian inference (VBI) framework with neural speech prior (VINP) for joint speech dereverberation and blind RIR identification. In VINP, a probabilistic signal model is constructed in the time-frequency (T-F) domain based on convolution transfer function (CTF) approximation. For the first time, we propose using an arbitrary discriminative dereverberation deep neural network (DNN) to estimate the prior distribution of anechoic speech within a probabilistic model. By integrating both reverberant speech and the anechoic speech prior, VINP yields the maximum a posteriori (MAP) and maximum likelihood (ML) estimations of the anechoic speech spectrum and CTF filter, respectively. After simple transformations, the waveforms of anechoic speech and RIR are estimated. VINP is effective for automatic speech recognition (ASR) systems, which sets it apart from most deep learning (DL)-based single-channel dereverberation approaches. Experiments on single-channel speech dereverberation demonstrate that VINP attains state-of-the-art (SOTA) performance in mean opinion score (MOS) and word error rate (WER). For blind RIR identification, experiments demonstrate that VINP achieves SOTA performance in estimating reverberation time at 60 dB (RT60) and advanced performance in direct-to-reverberation ratio (DRR) estimation. Codes and audio samples are available online.
Authors: Jingbiao Mei, Jinghong Chen, Guangyu Yang, Weizhe Lin, Bill Byrne
Abstract: Hateful memes have become a significant concern on the Internet, necessitating robust automated detection systems. While Large Multimodal Models (LMMs) have shown promise in hateful meme detection, they face notable challenges like sub-optimal performance and limited out-of-domain generalization capabilities. Recent studies further reveal the limitations of both supervised fine-tuning (SFT) and in-context learning when applied to LMMs in this setting. To address these issues, we propose a robust adaptation framework for hateful meme detection that enhances in-domain accuracy and cross-domain generalization while preserving the general vision-language capabilities of LMMs. Analysis reveals that our approach achieves improved robustness under adversarial attacks compared to SFT models. Experiments on six meme classification datasets show that our approach achieves state-of-the-art performance, outperforming larger agentic systems. Moreover, our method generates higher-quality rationales for explaining hateful content compared to standard SFT, enhancing model interpretability. Code available at https://github.com/JingbiaoMei/RGCL
Authors: Tianyi Wang, Jianan Fan, Dingxin Zhang, Dongnan Liu, Yong Xia, Heng Huang, Weidong Cai
Abstract: Histopathology and transcriptomics are fundamental modalities in oncology, encapsulating the morphological and molecular aspects of the disease. Multi-modal self-supervised learning has demonstrated remarkable potential in learning pathological representations by integrating diverse data sources. Conventional multi-modal integration methods primarily emphasize modality alignment, while paying insufficient attention to retaining the modality-specific structures. However, unlike conventional scenarios where multi-modal inputs share highly overlapping features, histopathology and transcriptomics exhibit pronounced heterogeneity, offering orthogonal yet complementary insights. Histopathology provides morphological and spatial context, elucidating tissue architecture and cellular topology, whereas transcriptomics delineates molecular signatures through gene expression patterns. This inherent disparity introduces a major challenge in aligning them while maintaining modality-specific fidelity. To address these challenges, we present MIRROR, a novel multi-modal representation learning method designed to foster both modality alignment and retention. MIRROR employs dedicated encoders to extract comprehensive features for each modality, which is further complemented by a modality alignment module to achieve seamless integration between phenotype patterns and molecular profiles. Furthermore, a modality retention module safeguards unique attributes from each modality, while a style clustering module mitigates redundancy and enhances disease-relevant information by modeling and aligning consistent pathological signatures within a clustering space. Extensive evaluations on TCGA cohorts for cancer subtyping and survival analysis highlight MIRROR's superior performance, demonstrating its effectiveness in constructing comprehensive oncological feature representations and benefiting the cancer diagnosis.
Authors: Kaixin Zhang, Hongzhi Wang, Ziqi Li, Yabin Lu, Yingze Li, Yu Yan, Yiming Guan
Abstract: Research on learned cardinality estimation has made significant progress in recent years. However, existing methods still face distinct challenges that hinder their practical deployment in production environments. We define these challenges as the ``Trilemma of Cardinality Estimation'', where learned cardinality estimation methods struggle to balance generality, accuracy, and updatability. To address these challenges, we introduce DistJoin, a join cardinality estimator based on efficient distribution prediction using multi-autoregressive models. Our contributions are threefold: (1) We propose a method to estimate join cardinality by leveraging the probability distributions of individual tables in a decoupled manner. (2) To meet the requirements of efficiency for DistJoin, we develop Adaptive Neural Predicate Modulation (ANPM), a high-throughput distribution estimation model. (3) We demonstrate that an existing similar approach suffers from variance accumulation issues by formal variance analysis. To mitigate this problem, DistJoin employs a selectivity-based approach to infer join cardinality, effectively reducing variance. In summary, DistJoin not only represents the first data-driven method to support both equi and non-equi joins simultaneously but also demonstrates superior accuracy while enabling fast and flexible updates. The experimental results demonstrate that DistJoin achieves the highest accuracy, robustness to data updates, generality, and comparable update and inference speed relative to existing methods.
Authors: Akram Khatami-Rizi, Ahmad Mahmoudi-Aznaveh
Abstract: Single-image super-resolution (SISR) is a fundamental problem in computer vision that aims to reconstruct high-resolution (HR) images from low-resolution (LR) inputs. Although convolutional neural networks (CNNs) have achieved substantial advancements, deeper architectures often introduce excessive parameters, higher memory usage, and computational cost, limiting their applicability on resource-constrained devices. Recent research has thus focused on lightweight architectures that preserve accuracy while reducing complexity. This paper presents the Involution and BSConv Multi-Depth Distillation Network (IBMDN), a lightweight and effective architecture for SISR. The proposed IBMDN comprises Involution and BSConv Multi-Depth Distillation Blocks (IBMDB) and a Contrast and High-Frequency Attention Block (CHFAB). IBMDB employs varying combinations of Involution and BSConv at multiple depths to perform efficient feature extraction while minimizing computational complexity. CHFAB, a lightweight self-attention mechanism, focuses on extracting high-frequency and contrast information to enhance perceptual quality in the reconstructed images. The flexible design of IBMDB enables it to be seamlessly integrated into diverse SISR frameworks, including information distillation, transformer-based, and GAN-based models. Extensive experiments demonstrate that incorporating IBMDB significantly reduces memory usage, parameters, and floating-point operations (FLOPs), while achieving improvements in both pixel-wise accuracy and visual quality. The source code is available at: https://github.com/akramkhatami/IBMDN.
Authors: Luxi He, Xiangyu Qi, Michel Liao, Inyoung Cheong, Prateek Mittal, Danqi Chen, Peter Henderson
Abstract: The latest Audio Language Models (Audio LMs) process speech directly instead of relying on a separate transcription step. This shift preserves detailed information, such as intonation or the presence of multiple speakers, that would otherwise be lost in transcription. However, it also introduces new safety risks, including the potential misuse of speaker identity cues and other sensitive vocal attributes, which could have legal implications. In this paper, we urge a closer examination of how these models are built and deployed. Our experiments show that end-to-end modeling, compared with cascaded pipelines, creates socio-technical safety risks such as identity inference, biased decision-making, and emotion detection. This raises concerns about whether Audio LMs store voiceprints and function in ways that create uncertainty under existing legal regimes. We then argue that the Principle of Least Privilege should be considered to guide the development and deployment of these models. Specifically, evaluations should assess (1) the privacy and safety risks associated with end-to-end modeling; and (2) the appropriate scope of information access. Finally, we highlight related gaps in current audio LM benchmarks and identify key open research questions, both technical and policy-related, that must be addressed to enable the responsible deployment of end-to-end Audio LMs.
Authors: Yudong Yang, Jimin Zhuang, Guangzhi Sun, Changli Tang, Yixuan Li, Peihan Li, Yifan Jiang, Wei Li, Zejun Ma, Chao Zhang
Abstract: Audio often serves as an auxiliary modality in video understanding tasks of audio-visual large language models (LLMs), merely assisting in the comprehension of visual information. However, a thorough understanding of videos significantly depends on auditory information, as audio offers critical context, emotional cues, and semantic meaning that visual data alone often lacks. This paper proposes an audio-centric video understanding benchmark (AVUT) to evaluate the video comprehension capabilities of multimodal LLMs with a particular focus on auditory information. AVUT introduces a suite of carefully designed audio-centric tasks, holistically testing the understanding of both audio content and audio-visual interactions in videos. Moreover, this work points out the text shortcut problem that largely exists in other benchmarks where the correct answer can be found from question text alone without needing videos. AVUT addresses this problem by proposing a answer permutation-based filtering mechanism. A thorough evaluation across a diverse range of open-source and proprietary multimodal LLMs is performed, followed by the analyses of deficiencies in audio-visual LLMs. Demos and data are available at https://github.com/lark-png/AVUT.
Authors: Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
Abstract: Traffic crash detection in long-form surveillance videos is essential for improving emergency response and infrastructure planning, yet remains difficult due to the brief and infrequent nature of crash events. We present \textbf{HybridMamba}, a novel architecture integrating visual transformers with state-space temporal modeling to achieve high-precision crash time localization. Our approach introduces multi-level token compression and hierarchical temporal processing to maintain computational efficiency without sacrificing temporal resolution. Evaluated on a large-scale dataset from the Iowa Department of Transportation, HybridMamba achieves a mean absolute error of \textbf{1.50 seconds} for 2-minute videos ($p<0.01$ compared to baselines), with \textbf{65.2%} of predictions falling within one second of the ground truth. It outperforms recent video-language models (e.g., TimeChat, VideoLLaMA-2) by up to 3.95 seconds while using significantly fewer parameters (3B vs. 13--72B). Our results demonstrate effective temporal localization across various video durations (2--40 minutes) and diverse environmental conditions, highlighting HybridMamba's potential for fine-grained temporal localization in traffic surveillance while identifying challenges that remain for extended deployment.
Authors: Guohao Huo, Zibo Lin, Zitong Wang, Ruiting Dai, Hao Tang
Abstract: Ophthalmic diseases pose a significant global health burden. However, traditional diagnostic methods and existing monocular image-based deep learning approaches often overlook the pathological correlations between the two eyes. In practical medical robotic diagnostic scenarios, paired retinal images (binocular fundus images) are frequently required as diagnostic evidence. To address this, we propose DMS-Net-a dual-modal multi-scale siamese network for binocular retinal image classification. The framework employs a weight-sharing siamese ResNet-152 architecture to concurrently extract deep semantic features from bilateral fundus images. To tackle challenges like indistinct lesion boundaries and diffuse pathological distributions, we introduce the OmniPool Spatial Integrator Module (OSIM), which achieves multi-resolution feature aggregation through multi-scale adaptive pooling and spatial attention mechanisms. Furthermore, the Calibrated Analogous Semantic Fusion Module (CASFM) leverages spatial-semantic recalibration and bidirectional attention mechanisms to enhance cross-modal interaction, aggregating modality-agnostic representations of fundus structures. To fully exploit the differential semantic information of lesions present in bilateral fundus features, we introduce the Cross-Modal Contrastive Alignment Module (CCAM). Additionally, to enhance the aggregation of lesion-correlated semantic information, we introduce the Cross-Modal Integrative Alignment Module (CIAM). Evaluation on the ODIR-5K dataset demonstrates that DMS-Net achieves state-of-the-art performance with an accuracy of 82.9%, recall of 84.5%, and a Cohen's kappa coefficient of 83.2%, showcasing robust capacity in detecting symmetrical pathologies and improving clinical decision-making for ocular diseases. Code and the processed dataset will be released subsequently.
Authors: Michael Banf, Johannes Kuhn
Abstract: Large Language Models (LLMs) have shown remarkable capabilities across various domains, yet they struggle with knowledge-intensive tasks in areas that demand factual accuracy, e.g. industrial automation and healthcare. Key limitations include their tendency to hallucinate, lack of source traceability (provenance), and challenges in timely knowledge updates. Combining language models with knowledge graphs (GraphRAG) offers promising avenues for overcoming these deficits. However, a major challenge lies in creating such a knowledge graph in the first place. Here, we propose a novel approach that combines LLMs with a tripartite knowledge graph representation, which is constructed by connecting complex, domain-specific objects via a curated ontology of corresponding, domain-specific concepts to relevant sections within chunks of text through a concept-anchored pre-analysis of source documents starting from an initial lexical graph. Subsequently, we formulate LLM prompt creation as an unsupervised node classification problem allowing for the optimization of information density, coverage, and arrangement of LLM prompts at significantly reduced lengths. An initial experimental evaluation of our approach on a healthcare use case, involving multi-faceted analyses of patient anamneses given a set of medical concepts as well as a series of clinical guideline literature, indicates its potential to optimize information density, coverage, and arrangement of LLM prompts while significantly reducing their lengths, which, in turn, may lead to reduced costs as well as more consistent and reliable LLM outputs.
Authors: Akhiad Bercovich, Itay Levy, Izik Golan, Mohammad Dabbah, Ran El-Yaniv, Omri Puny, Ido Galil, Zach Moshe, Tomer Ronen, Najeeb Nabwani, Ido Shahaf, Oren Tropp, Ehud Karpas, Ran Zilberstein, Jiaqi Zeng, Soumye Singhal, Alexander Bukharin, Yian Zhang, Tugrul Konuk, Gerald Shen, Ameya Sunil Mahabaleshwarkar, Bilal Kartal, Yoshi Suhara, Olivier Delalleau, Zijia Chen, Zhilin Wang, David Mosallanezhad, Adi Renduchintala, Haifeng Qian, Dima Rekesh, Fei Jia, Somshubra Majumdar, Vahid Noroozi, Wasi Uddin Ahmad, Sean Narenthiran, Aleksander Ficek, Mehrzad Samadi, Jocelyn Huang, Siddhartha Jain, Igor Gitman, Ivan Moshkov, Wei Du, Shubham Toshniwal, George Armstrong, Branislav Kisacanin, Matvei Novikov, Daria Gitman, Evelina Bakhturina, Prasoon Varshney, Makesh Narsimhan, Jane Polak Scowcroft, John Kamalu, Dan Su, Kezhi Kong, Markus Kliegl, Rabeeh Karimi Mahabadi, Ying Lin, Sanjeev Satheesh, Jupinder Parmar, Pritam Gundecha, Brandon Norick, Joseph Jennings, Shrimai Prabhumoye, Syeda Nahida Akter, Mostofa Patwary, Abhinav Khattar, Deepak Narayanan, Roger Waleffe, Jimmy Zhang, Bor-Yiing Su, Guyue Huang, Terry Kong, Parth Chadha, Sahil Jain, Christine Harvey, Elad Segal, Jining Huang, Sergey Kashirsky, Robert McQueen, Izzy Putterman, George Lam, Arun Venkatesan, Sherry Wu, Vinh Nguyen, Manoj Kilaru, Andrew Wang, Anna Warno, Abhilash Somasamudramath, Sandip Bhaskar, Maka Dong, Nave Assaf, Shahar Mor, Omer Ullman Argov, Scot Junkin, Oleksandr Romanenko, Pedro Larroy, Monika Katariya, Marco Rovinelli, Viji Balas, Nicholas Edelman, Anahita Bhiwandiwalla, Muthu Subramaniam, Smita Ithape, Karthik Ramamoorthy, Yuting Wu, Suguna Varshini Velury, Omri Almog, Joyjit Daw, Denys Fridman, Erick Galinkin, Michael Evans, Shaona Ghosh, Katherine Luna, Leon Derczynski, Nikki Pope, Eileen Long, Seth Schneider, Guillermo Siman, Tomasz Grzegorzek, Pablo Ribalta, Monika Katariya, Chris Alexiuk, Joey Conway, Trisha Saar, Ann Guan, Krzysztof Pawelec, Shyamala Prayaga, Oleksii Kuchaiev, Boris Ginsburg, Oluwatobi Olabiyi, Kari Briski, Jonathan Cohen, Bryan Catanzaro, Jonah Alben, Yonatan Geifman, Eric Chung
Abstract: We introduce the Llama-Nemotron series of models, an open family of heterogeneous reasoning models that deliver exceptional reasoning capabilities, inference efficiency, and an open license for enterprise use. The family comes in three sizes -- Nano (8B), Super (49B), and Ultra (253B) -- and performs competitively with state-of-the-art reasoning models such as DeepSeek-R1 while offering superior inference throughput and memory efficiency. In this report, we discuss the training procedure for these models, which entails using neural architecture search from Llama 3 models for accelerated inference, knowledge distillation, and continued pretraining, followed by a reasoning-focused post-training stage consisting of two main parts: supervised fine-tuning and large scale reinforcement learning. Llama-Nemotron models are the first open-source models to support a dynamic reasoning toggle, allowing users to switch between standard chat and reasoning modes during inference. To further support open research and facilitate model development, we provide the following resources: 1. We release the Llama-Nemotron reasoning models -- LN-Nano, LN-Super, and LN-Ultra -- under the commercially permissive NVIDIA Open Model License Agreement. 2. We release the complete post-training dataset: Llama-Nemotron-Post-Training-Dataset. 3. We also release our training codebases: NeMo, NeMo-Aligner, and Megatron-LM.
Authors: Guangzhi Sun, Potsawee Manakul, Xiao Zhan, Mark Gales
Abstract: Unlearning has emerged as a critical capability for large language models (LLMs) to support data privacy, regulatory compliance, and ethical AI deployment. Recent techniques often rely on obfuscation by injecting incorrect or irrelevant information to suppress knowledge. Such methods effectively constitute knowledge addition rather than true removal, often leaving models vulnerable to probing. In this paper, we formally distinguish unlearning from obfuscation and introduce a probing-based evaluation framework to assess whether existing approaches genuinely remove targeted information. Moreover, we propose DF-MCQ, a novel unlearning method that flattens the model predictive distribution over automatically generated multiple-choice questions using KL-divergence, effectively removing knowledge about target individuals and triggering appropriate refusal behaviour. Experimental results demonstrate that DF-MCQ achieves unlearning with over 90% refusal rate and a random choice-level uncertainty that is much higher than obfuscation on probing questions.
Authors: Tasnim Shahriar
Abstract: This paper presents a comprehensive evaluation of lightweight deep learning models for image classification, emphasizing their suitability for deployment in resource-constrained environments such as low-memory devices. Five state-of-the-art architectures - MobileNetV3 Small, ResNet18, SqueezeNet, EfficientNetV2-S, and ShuffleNetV2 - are benchmarked across three diverse datasets: CIFAR-10, CIFAR-100, and Tiny ImageNet. The models are assessed using four key performance metrics: classification accuracy, inference time, floating-point operations (FLOPs), and model size. Additionally, we investigate the impact of hyperparameter tuning, data augmentation, and training paradigms by comparing pretrained models with scratch-trained counterparts, focusing on MobileNetV3 Small. Our findings reveal that transfer learning significantly enhances model accuracy and computational efficiency, particularly for complex datasets like Tiny ImageNet. EfficientNetV2 consistently achieves the highest accuracy, while MobileNetV3 offers the best balance between accuracy and efficiency, and SqueezeNet excels in inference speed and compactness. This study highlights critical trade-offs between accuracy and efficiency, offering actionable insights for deploying lightweight models in real-world applications where computational resources are limited. By addressing these challenges, this research contributes to optimizing deep learning systems for edge computing and mobile platforms.
Authors: Xiaoyu Xu, Minxin Du, Qingqing Ye, Haibo Hu
Abstract: Large language models (LLMs) trained over extensive corpora risk memorizing sensitive, copyrighted, or toxic content. To address this, we propose \textbf{OBLIVIATE}, a robust unlearning framework that removes targeted data while preserving model utility. The framework follows a structured process: extracting target tokens, building retain sets, and fine-tuning with a tailored loss function comprising three components -- masking, distillation, and world fact. Using low-rank adapters (LoRA) ensures efficiency without compromising unlearning quality. We conduct experiments on multiple datasets, including Harry Potter series, WMDP, and TOFU, using a comprehensive suite of metrics: \emph{forget quality} (via a new document-level memorization score), \emph{model utility}, and \emph{fluency}. Results demonstrate its effectiveness in resisting membership inference attacks, minimizing the impact on retained data, and maintaining robustness across diverse scenarios.
Authors: Jingjie Zheng, Aryo Pradipta Gema, Giwon Hong, Xuanli He, Pasquale Minervini, Youcheng Sun, Qiongkai Xu
Abstract: Retrieval Augmented Generation (RAG) frameworks improve the accuracy of large language models (LLMs) by integrating external knowledge from retrieved documents, thereby overcoming the limitations of models' static intrinsic knowledge. However, these systems are susceptible to adversarial attacks that manipulate the retrieval process by introducing documents that are adversarial yet semantically similar to the query. Notably, while these adversarial documents resemble the query, they exhibit weak similarity to benign documents in the retrieval set. Thus, we propose a simple yet effective Graph-based Reranking against Adversarial Document Attacks (GRADA) framework aiming at preserving retrieval quality while significantly reducing the success of adversaries. Our study evaluates the effectiveness of our approach through experiments conducted on five LLMs: GPT-3.5-Turbo, GPT-4o, Llama3.1-8b, Llama3.1-70b, and Qwen2.5-7b. We use three datasets to assess performance, with results from the Natural Questions dataset demonstrating up to an 80% reduction in attack success rates while maintaining minimal loss in accuracy.
Authors: Assaf Ben-Kish, Itamar Zimerman, M. Jehanzeb Mirza, Lior Wolf, James Glass, Leonid Karlinsky, Raja Giryes
Abstract: A recent trend in LLMs is developing recurrent sub-quadratic models that improve long-context processing efficiency. We investigate leading large long-context models, focusing on how their fixed-size recurrent memory affects their performance. Our experiments reveal that, even when these models are trained for extended contexts, their use of long contexts remains underutilized. Specifically, we demonstrate that a chunk-based inference procedure, which identifies and processes only the most relevant portion of the input can mitigate recurrent memory failures and be effective for many long-context tasks: On LongBench, our method improves the overall performance of Falcon3-Mamba-Inst-7B by 14%, Falcon-Mamba-Inst-7B by 28%, RecurrentGemma-IT-9B by 50%, and RWKV6-Finch-7B by 51%. Surprisingly, this simple approach also leads to state-of-the-art results in the challenging LongBench v2 benchmark, showing competitive performance with equivalent size Transformers. Furthermore, our findings raise questions about whether recurrent models genuinely exploit long-range dependencies, as our single-chunk strategy delivers stronger performance - even in tasks that presumably require cross-context relations.
Authors: Qi Feng
Abstract: Video-based spatial cognition is vital for robotics and embodied AI but challenges current Vision-Language Models (VLMs). This paper makes two key contributions. First, we introduce ViCA (Visuospatial Cognitive Assistant)-322K, a diverse dataset of 322,003 QA pairs from real-world indoor videos (ARKitScenes, ScanNet, ScanNet++), offering supervision for 3D metadata-grounded queries and video-based complex reasoning. Second, we develop ViCA-7B, fine-tuned on ViCA-322K, which achieves new state-of-the-art on all eight VSI-Bench tasks, outperforming existing models, including larger ones (e.g., +26.1 on Absolute Distance). For interpretability, we present ViCA-Thinking-2.68K, a dataset with explicit reasoning chains, and fine-tune ViCA-7B to create ViCA-7B-Thinking, a model that articulates its spatial reasoning. Our work highlights the importance of targeted data and suggests paths for improved temporal-spatial modeling. We release all resources to foster research in robust visuospatial intelligence.
Authors: Qi Feng
Abstract: While Multimodal Large Language Models (MLLMs) excel at general vision-language tasks, visuospatial cognition - reasoning about spatial layouts, relations, and dynamics - remains a significant challenge. Existing models often lack the necessary architectural components and specialized training data for fine-grained spatial understanding. We introduce ViCA2 (Visuospatial Cognitive Assistant 2), a novel MLLM designed to enhance spatial reasoning. ViCA2 features a dual vision encoder architecture integrating SigLIP for semantics and Hiera for spatial structure, coupled with a token ratio control mechanism for efficiency. We also developed ViCA-322K, a new large-scale dataset with over 322,000 spatially grounded question-answer pairs for targeted instruction tuning. On the challenging VSI-Bench benchmark, our ViCA2-7B model achieves a state-of-the-art average score of 56.8, significantly surpassing larger open-source models (e.g., LLaVA-NeXT-Video-72B, 40.9) and leading proprietary models (Gemini-1.5 Pro, 45.4). This demonstrates the effectiveness of our approach in achieving strong visuospatial intelligence with a compact model. We release ViCA2, its codebase, and the ViCA-322K dataset to facilitate further research.
Authors: Ander Artola Velasco, Stratis Tsirtsis, Nastaran Okati, Manuel Gomez-Rodriguez
Abstract: State-of-the-art large language models require specialized hardware and substantial energy to operate. As a consequence, cloud-based services that provide access to large language models have become very popular. In these services, the price users pay for an output provided by a model depends on the number of tokens the model uses to generate it -- they pay a fixed price per token. In this work, we show that this pricing mechanism creates a financial incentive for providers to strategize and misreport the (number of) tokens a model used to generate an output, and users cannot prove, or even know, whether a provider is overcharging them. However, we also show that, if an unfaithful provider is obliged to be transparent about the generative process used by the model, misreporting optimally without raising suspicion is hard. Nevertheless, as a proof-of-concept, we develop an efficient heuristic algorithm that allows providers to significantly overcharge users without raising suspicion. Crucially, we demonstrate that the cost of running the algorithm is lower than the additional revenue from overcharging users, highlighting the vulnerability of users under the current pay-per-token pricing mechanism. Further, we show that, to eliminate the financial incentive to strategize, a pricing mechanism must price tokens linearly on their character count. While this makes a provider's profit margin vary across tokens, we introduce a simple prescription under which the provider who adopts such an incentive-compatible pricing mechanism can maintain the average profit margin they had under the pay-per-token pricing mechanism. Along the way, to illustrate and complement our theoretical results, we conduct experiments with several large language models from the $\texttt{Llama}$, $\texttt{Gemma}$ and $\texttt{Ministral}$ families, and input prompts from the LMSYS Chatbot Arena platform.
Authors: Yu Zhang, Yuqi Xie, Huihan Liu, Rutav Shah, Michael Wan, Linxi Fan, Yuke Zhu
Abstract: Imitation learning advances robot capabilities by enabling the acquisition of diverse behaviors from human demonstrations. However, large-scale datasets used for policy training often introduce substantial variability in quality, which can negatively impact performance. As a result, automatically curating datasets by filtering low-quality samples to improve quality becomes essential. Existing robotic curation approaches rely on costly manual annotations and perform curation at a coarse granularity, such as the dataset or trajectory level, failing to account for the quality of individual state-action pairs. To address this, we introduce SCIZOR, a self-supervised data curation framework that filters out low-quality state-action pairs to improve the performance of imitation learning policies. SCIZOR targets two complementary sources of low-quality data: suboptimal data, which hinders learning with undesirable actions, and redundant data, which dilutes training with repetitive patterns. SCIZOR leverages a self-supervised task progress predictor for suboptimal data to remove samples lacking task progression, and a deduplication module operating on joint state-action representation for samples with redundant patterns. Empirically, we show that SCIZOR enables imitation learning policies to achieve higher performance with less data, yielding an average improvement of 15.4% across multiple benchmarks. More information is available at: https://ut-austin-rpl.github.io/SCIZOR/
Authors: Wonjun Yi, Wonho Jung, Hyeonuk Nam, Kangmin Jang, Yong-Hwa Park
Abstract: The increasing complexity of rotating machinery and the diversity of operating conditions, such as rotating speed and varying torques, have amplified the challenges in fault diagnosis in scenarios requiring domain adaptation, particularly involving compound faults. This study addresses these challenges by introducing a novel multi-output classification (MOC) framework tailored for domain adaptation in partially labeled target datasets. Unlike conventional multi-class classification (MCC) approaches, the MOC framework classifies the severity levels of compound faults simultaneously. Furthermore, we explore various single-task and multi-task architectures applicable to the MOC formulation-including shared trunk and cross-talk-based designs-for compound fault diagnosis under partially labeled conditions. Based on this investigation, we propose a novel cross-talk architecture, residual neural dimension reductor (RNDR), that enables selective information sharing across diagnostic tasks, effectively enhancing classification performance in compound fault scenarios. In addition, frequency-layer normalization was incorporated to improve domain adaptation performance on motor vibration data. Compound fault conditions were implemented using a motor-based test setup and evaluated across six domain adaptation scenarios. The experimental results demonstrate its superior macro F1 performance compared to baseline models. We further showed that the structural advantage of RNDR is more pronounced in compound fault settings through a single-fault comparison. We also found that frequency-layer normalization fits the fault diagnosis task better than conventional methods. Lastly, we analyzed the RNDR with various conditions, other models with increased number of parameters, and compared with the ablated RNDR structure.
Authors: Celia Cintas, Miriam Rateike, Erik Miehling, Elizabeth Daly, Skyler Speakman
Abstract: We present a study on how and where personas -- defined by distinct sets of human characteristics, values, and beliefs -- are encoded in the representation space of large language models (LLMs). Using a range of dimension reduction and pattern recognition methods, we first identify the model layers that show the greatest divergence in encoding these representations. We then analyze the activations within a selected layer to examine how specific personas are encoded relative to others, including their shared and distinct embedding spaces. We find that, across multiple pre-trained decoder-only LLMs, the analyzed personas show large differences in representation space only within the final third of the decoder layers. We observe overlapping activations for specific ethical perspectives -- such as moral nihilism and utilitarianism -- suggesting a degree of polysemy. In contrast, political ideologies like conservatism and liberalism appear to be represented in more distinct regions. These findings help to improve our understanding of how LLMs internally represent information and can inform future efforts in refining the modulation of specific human traits in LLM outputs. Warning: This paper includes potentially offensive sample statements.
Authors: Ziyan Luo, Tianwei Ni, Pierre-Luc Bacon, Doina Precup, Xujie Si
Abstract: A key approach to state abstraction is approximating behavioral metrics (notably, bisimulation metrics) in the observation space and embedding these learned distances in the representation space. While promising for robustness to task-irrelevant noise, as shown in prior work, accurately estimating these metrics remains challenging, requiring various design choices that create gaps between theory and practice. Prior evaluations focus mainly on final returns, leaving the quality of learned metrics and the source of performance gains unclear. To systematically assess how metric learning works in deep reinforcement learning (RL), we evaluate five recent approaches, unified conceptually as isometric embeddings with varying design choices. We benchmark them with baselines across 20 state-based and 14 pixel-based tasks, spanning 370 task configurations with diverse noise settings. Beyond final returns, we introduce the evaluation of a denoising factor to quantify the encoder's ability to filter distractions. To further isolate the effect of metric learning, we propose and evaluate an isolated metric estimation setting, in which the encoder is influenced solely by the metric loss. Finally, we release an open-source, modular codebase to improve reproducibility and support future research on metric learning in deep RL.
Authors: Rynaa Grover, Jayant Sravan Tamarapalli, Sahiti Yerramilli, Nilay Pande
Abstract: Multimodal Large Language Models (MLLMs) excel at high-level visual reasoning, but their performance on nuanced perceptual tasks remains surprisingly limited. We present HueManity, a benchmark designed to assess visual perception in MLLMs. The dataset comprises 83,850 images featuring two-character alphanumeric strings embedded in Ishihara test style dot patterns, challenging models on precise pattern recognition. Our evaluation of nine state-of-the-art MLLMs on HueManity demonstrates a significant performance deficit compared to human and traditional computer vision baselines. The best-performing MLLM achieved a 33.6% accuracy on the numeric `easy' task and a striking 3% on the alphanumeric `hard' task. In contrast, human participants achieved near-perfect scores (100% and 95.6%), and a fine-tuned ResNet50 model reached accuracies of 96.5% and 94.5%. These results highlight a critical gap in the visual capabilities of current MLLMs. Our analysis further explores potential architectural and training-paradigm factors contributing to this perceptual gap in MLLMs. We open-source HueManity dataset and code to foster further research in improving perceptual robustness of MLLMs.
Authors: Yutong Zhou, Masahiro Ryo
Abstract: Explaining why the species lives at a particular location is important for understanding ecological systems and conserving biodiversity. However, existing ecological workflows are fragmented and often inaccessible to non-specialists. We propose an end-to-end visual-to-causal framework that transforms a species image into interpretable causal insights about its habitat preference. The system integrates species recognition, global occurrence retrieval, pseudo-absence sampling, and climate data extraction. We then discover causal structures among environmental features and estimate their influence on species occurrence using modern causal inference methods. Finally, we generate statistically grounded, human-readable causal explanations from structured templates and large language models. We demonstrate the framework on a bee and a flower species and report early results as part of an ongoing project, showing the potential of the multimodal AI assistant backed up by a recommended ecological modeling practice for describing species habitat in human-understandable language. Our code is available at: https://github.com/Yutong-Zhou-cv/BioX.
Authors: Nathaniel Getachew, Abulhair Saparov
Abstract: We introduce $\texttt{StorySim}$, a programmable framework for synthetically generating stories to evaluate the theory of mind (ToM) and world modeling (WM) capabilities of large language models (LLMs). Unlike prior benchmarks that may suffer from contamination in pretraining data, $\texttt{StorySim}$ produces novel, compositional story prompts anchored by a highly controllable $\texttt{Storyboard}$, enabling precise manipulation of character perspectives and events. We use this framework to design first- and second-order ToM tasks alongside WM tasks that control for the ability to track and model mental states. Our experiments across a suite of state-of-the-art LLMs reveal that most models perform better on WM tasks than ToM tasks, and that models tend to perform better reasoning with humans compared to inanimate objects. Additionally, our framework enabled us to find evidence of heuristic behavior such as recency bias and an over-reliance on earlier events in the story. All code for generating data and evaluations is freely available.
Authors: Hamza Rasaee, Taha Koleilat, Hassan Rivaz
Abstract: Accurate and generalizable object segmentation in ultrasound imaging remains a significant challenge due to anatomical variability, diverse imaging protocols, and limited annotated data. In this study, we propose a prompt-driven vision-language model (VLM) that integrates Grounding DINO with SAM2 (Segment Anything Model2) to enable object segmentation across multiple ultrasound organs. A total of 18 public ultrasound datasets, encompassing the breast, thyroid, liver, prostate, kidney, and paraspinal muscle, were utilized. These datasets were divided into 15 for fine-tuning and validation of Grounding DINO using Low Rank Adaptation (LoRA) to the ultrasound domain, and 3 were held out entirely for testing to evaluate performance in unseen distributions. Comprehensive experiments demonstrate that our approach outperforms state-of-the-art segmentation methods, including UniverSeg, MedSAM, MedCLIP-SAM, BiomedParse, and SAMUS on most seen datasets while maintaining strong performance on unseen datasets without additional fine-tuning. These results underscore the promise of VLMs in scalable and robust ultrasound image analysis, reducing dependence on large, organ-specific annotated datasets. We will publish our code on code.sonography.ai after acceptance.
Authors: Sanjeda Akter, Ibne Farabi Shihab, Anuj Sharma
Abstract: Crash detection from video feeds is a critical problem in intelligent transportation systems. Recent developments in large language models (LLMs) and vision-language models (VLMs) have transformed how we process, reason about, and summarize multimodal information. This paper surveys recent methods leveraging LLMs for crash detection from video data. We present a structured taxonomy of fusion strategies, summarize key datasets, analyze model architectures, compare performance benchmarks, and discuss ongoing challenges and opportunities. Our review provides a foundation for future research in this fast-growing intersection of video understanding and foundation models.
Authors: Fangqiu Yi, Jingyu Xu, Jiawei Shao, Chi Zhang, Xuelong Li
Abstract: Perceptual studies demonstrate that conditional diffusion models excel at reconstructing video content aligned with human visual perception. Building on this insight, we propose a video compression framework that leverages conditional diffusion models for perceptually optimized reconstruction. Specifically, we reframe video compression as a conditional generation task, where a generative model synthesizes video from sparse, yet informative signals. Our approach introduces three key modules: (1) Multi-granular conditioning that captures both static scene structure and dynamic spatio-temporal cues; (2) Compact representations designed for efficient transmission without sacrificing semantic richness; (3) Multi-condition training with modality dropout and role-aware embeddings, which prevent over-reliance on any single modality and enhance robustness. Extensive experiments show that our method significantly outperforms both traditional and neural codecs on perceptual quality metrics such as Fr\'echet Video Distance (FVD) and LPIPS, especially under high compression ratios.
Authors: Arup Kumar Sahoo, Itzik Klein
Abstract: A fundamental requirement for full autonomy in mobile robots is accurate navigation even in situations where satellite navigation or cameras are unavailable. In such practical situations, relying only on inertial sensors will result in navigation solution drift due to the sensors' inherent noise and error terms. One of the emerging solutions to mitigate drift is to maneuver the robot in a snake-like slithering motion to increase the inertial signal-to-noise ratio, allowing the regression of the mobile robot position. In this work, we propose MoRPI-PINN as a physics-informed neural network framework for accurate inertial-based mobile robot navigation. By embedding physical laws and constraints into the training process, MoRPI-PINN is capable of providing an accurate and robust navigation solution. Using real-world experiments, we show accuracy improvements of over 85% compared to other approaches. MoRPI-PINN is a lightweight approach that can be implemented even on edge devices and used in any typical mobile robot application.
Authors: Supantho Rakshit, Adele Goldberg
Abstract: The usage-based constructionist (UCx) approach to language posits that language comprises a network of learned form-meaning pairings (constructions) whose use is largely determined by their meanings or functions, requiring them to be graded and probabilistic. This study investigates whether the internal representations in Large Language Models (LLMs) reflect the proposed function-infused gradience. We analyze representations of the English Double Object (DO) and Prepositional Object (PO) constructions in Pythia-$1.4$B, using a dataset of $5000$ sentence pairs systematically varied by human-rated preference strength for DO or PO. Geometric analyses show that the separability between the two constructions' representations, as measured by energy distance or Jensen-Shannon divergence, is systematically modulated by gradient preference strength, which depends on lexical and functional properties of sentences. That is, more prototypical exemplars of each construction occupy more distinct regions in activation space, compared to sentences that could have equally well have occured in either construction. These results provide evidence that LLMs learn rich, meaning-infused, graded representations of constructions and offer support for geometric measures for representations in LLMs.
Authors: Guilherme Guerino, Luiz Rodrigues, Luana Bianchini, Mariana Alves, Marcelo Marinho, Thomaz Veloso, Valmir Macario, Diego Dermeval, Thales Vieira, Ig Bittencourt, Seiji Isotani
Abstract: This study explores the integration of Augmented Intelligence (AuI) in Intelligent Tutoring Systems (ITS) to address challenges in Artificial Intelligence in Education (AIED), including teacher involvement, AI reliability, and resource accessibility. We present MathAIde, an ITS that uses computer vision and AI to correct mathematics exercises from student work photos and provide feedback. The system was designed through a collaborative process involving brainstorming with teachers, high-fidelity prototyping, A/B testing, and a real-world case study. Findings emphasize the importance of a teacher-centered, user-driven approach, where AI suggests remediation alternatives while teachers retain decision-making. Results highlight efficiency, usability, and adoption potential in classroom contexts, particularly in resource-limited environments. The study contributes practical insights into designing ITSs that balance user needs and technological feasibility, while advancing AIED research by demonstrating the effectiveness of a mixed-methods, user-centered approach to implementing AuI in educational technologies.
Authors: Thomas Thebaud, Yen-Ju Lu, Matthew Wiesner, Peter Viechnicki, Najim Dehak
Abstract: In dialogue transcription pipelines, Large Language Models (LLMs) are frequently employed in post-processing to improve grammar, punctuation, and readability. We explore a complementary post-processing step: enriching transcribed dialogues by adding metadata tags for speaker characteristics such as age, gender, and emotion. Some of the tags are global to the entire dialogue, while some are time-variant. Our approach couples frozen audio foundation models, such as Whisper or WavLM, with a frozen LLAMA language model to infer these speaker attributes, without requiring task-specific fine-tuning of either model. Using lightweight, efficient connectors to bridge audio and language representations, we achieve competitive performance on speaker profiling tasks while preserving modularity and speed. Additionally, we demonstrate that a frozen LLAMA model can compare x-vectors directly, achieving an Equal Error Rate of 8.8% in some scenarios.
Authors: Yan Zhang, Ahmad Mohammad Saber, Amr Youssef, Deepa Kundur
Abstract: Modern power grids face unprecedented complexity from Distributed Energy Resources (DERs), Electric Vehicles (EVs), and extreme weather, while also being increasingly exposed to cyberattacks that can trigger grid violations. This paper introduces Grid-Agent, an autonomous AI-driven framework that leverages Large Language Models (LLMs) within a multi-agent system to detect and remediate violations. Grid-Agent integrates semantic reasoning with numerical precision through modular agents: a planning agent generates coordinated action sequences using power flow solvers, while a validation agent ensures stability and safety through sandboxed execution with rollback mechanisms. To enhance scalability, the framework employs an adaptive multi-scale network representation that dynamically adjusts encoding schemes based on system size and complexity. Violation resolution is achieved through optimizing switch configurations, battery deployment, and load curtailment. Our experiments on IEEE and CIGRE benchmark networks, including the IEEE 69-bus, CIGRE MV, IEEE 30-bus test systems, demonstrate superior mitigation performance, highlighting Grid-Agent's suitability for modern smart grids requiring rapid, adaptive response.
Authors: Andrew Brown, Muhammad Roman, Barry Devereux
Abstract: This systematic review of the research literature on retrieval-augmented generation (RAG) provides a focused analysis of the most highly cited studies published between 2020 and May 2025. A total of 128 articles met our inclusion criteria. The records were retrieved from ACM Digital Library, IEEE Xplore, Scopus, ScienceDirect, and the Digital Bibliography and Library Project (DBLP). RAG couples a neural retriever with a generative language model, grounding output in up-to-date, non-parametric memory while retaining the semantic generalisation stored in model weights. Guided by the PRISMA 2020 framework, we (i) specify explicit inclusion and exclusion criteria based on citation count and research questions, (ii) catalogue datasets, architectures, and evaluation practices, and (iii) synthesise empirical evidence on the effectiveness and limitations of RAG. To mitigate citation-lag bias, we applied a lower citation-count threshold to papers published in 2025 so that emerging breakthroughs with naturally fewer citations were still captured. This review clarifies the current research landscape, highlights methodological gaps, and charts priority directions for future research.
Authors: Yaying Luo, Hui Fang, Zhu Sun
Abstract: Conversational Recommender Systems (CRS) provide personalized services through multi-turn interactions, yet most existing methods overlook users' heterogeneous decision-making styles and knowledge levels, which constrains both accuracy and efficiency. To address this gap, we propose CT-CRS (Consumer Type-Enhanced Conversational Recommender System), a framework that integrates consumer type modeling into dialogue recommendation. Based on consumer type theory, we define four user categories--dependent, efficient, cautious, and expert--derived from two dimensions: decision-making style (maximizers vs. satisficers) and knowledge level (high vs. low). CT-CRS employs interaction histories and fine-tunes the large language model to automatically infer user types in real time, avoiding reliance on static questionnaires. We incorporate user types into state representation and design a type-adaptive policy that dynamically adjusts recommendation granularity, diversity, and attribute query complexity. To further optimize the dialogue policy, we adopt Inverse Reinforcement Learning (IRL), enabling the agent to approximate expert-like strategies conditioned on consumer type. Experiments on LastFM, Amazon-Book, and Yelp show that CTCRS improves recommendation success rate and reduces interaction turns compared to strong baselines. Ablation studies confirm that both consumer type modeling and IRL contribute significantly to performance gains. These results demonstrate that CT-CRS offers a scalable and interpretable solution for enhancing CRS personalization through the integration of psychological modeling and advanced policy optimization.
Authors: V Venktesh, Mandeep Rathee, Avishek Anand
Abstract: Test-time scaling (TTS) has emerged as a new frontier for scaling the performance of Large Language Models. In test-time scaling, by using more computational resources during inference, LLMs can improve their reasoning process and task performance. Several approaches have emerged for TTS such as distilling reasoning traces from another model or exploring the vast decoding search space by employing a verifier. The verifiers serve as reward models that help score the candidate outputs from the decoding process to diligently explore the vast solution space and select the best outcome. This paradigm commonly termed has emerged as a superior approach owing to parameter free scaling at inference time and high performance gains. The verifiers could be prompt-based, fine-tuned as a discriminative or generative model to verify process paths, outcomes or both. Despite their widespread adoption, there is no detailed collection, clear categorization and discussion of diverse verification approaches and their training mechanisms. In this survey, we cover the diverse approaches in the literature and present a unified view of verifier training, types and their utility in test-time scaling. Our repository can be found at https://github.com/elixir-research-group/Verifierstesttimescaling.github.io.
URLs: https://github.com/elixir-research-group/Verifierstesttimescaling.github.io.
Authors: Kamel Kamel, Keshav Sood, Hridoy Sankar Dutta, Sunil Aryal
Abstract: Voice authentication has undergone significant changes from traditional systems that relied on handcrafted acoustic features to deep learning models that can extract robust speaker embeddings. This advancement has expanded its applications across finance, smart devices, law enforcement, and beyond. However, as adoption has grown, so have the threats. This survey presents a comprehensive review of the modern threat landscape targeting Voice Authentication Systems (VAS) and Anti-Spoofing Countermeasures (CMs), including data poisoning, adversarial, deepfake, and adversarial spoofing attacks. We chronologically trace the development of voice authentication and examine how vulnerabilities have evolved in tandem with technological advancements. For each category of attack, we summarize methodologies, highlight commonly used datasets, compare performance and limitations, and organize existing literature using widely accepted taxonomies. By highlighting emerging risks and open challenges, this survey aims to support the development of more secure and resilient voice authentication systems.
Authors: Qiming Guo, Jinwen Tang, Xingran Huang
Abstract: We introduce Advertisement Embedding Attacks (AEA), a new class of LLM security threats that stealthily inject promotional or malicious content into model outputs and AI agents. AEA operate through two low-cost vectors: (1) hijacking third-party service-distribution platforms to prepend adversarial prompts, and (2) publishing back-doored open-source checkpoints fine-tuned with attacker data. Unlike conventional attacks that degrade accuracy, AEA subvert information integrity, causing models to return covert ads, propaganda, or hate speech while appearing normal. We detail the attack pipeline, map five stakeholder victim groups, and present an initial prompt-based self-inspection defense that mitigates these injections without additional model retraining. Our findings reveal an urgent, under-addressed gap in LLM security and call for coordinated detection, auditing, and policy responses from the AI-safety community.
Authors: Chengjun Zhang, Yuhao Zhang, Jie Yang, Mohamad Sawan
Abstract: Spiking Neural Networks (SNNs), inspired by the brain, are characterized by minimal power consumption and swift inference capabilities on neuromorphic hardware, and have been widely applied to various visual perception tasks. Current ANN-SNN conversion methods have achieved excellent results in classification tasks with ultra-low time-steps, but their performance in visual detection tasks remains suboptimal. In this paper, we propose a delay-spike approach to mitigate the issue of residual membrane potential caused by heterogeneous spiking patterns. Furthermore, we propose a novel temporal-dependent Integrate-and-Fire (tdIF) neuron architecture for SNNs. This enables Integrate-and-fire (IF) neurons to dynamically adjust their accumulation and firing behaviors based on the temporal order of time-steps. Our method enables spikes to exhibit distinct temporal properties, rather than relying solely on frequency-based representations. Moreover, the tdIF neuron maintains energy consumption on par with traditional IF neuron. We demonstrate that our method achieves more precise feature representation with lower time-steps, enabling high performance and ultra-low latency in visual detection tasks. In this study, we conduct extensive evaluation of the tdIF method across two critical vision tasks: object detection and lane line detection. The results demonstrate that the proposed method surpasses current ANN-SNN conversion approaches, achieving state-of-the-art performance with ultra-low latency (within 5 time-steps).
Authors: Delin Qu, Haoming Song, Qizhi Chen, Zhaoqing Chen, Xianqiang Gao, Xinyi Ye, Qi Lv, Modi Shi, Guanghui Ren, Cheng Ruan, Maoqing Yao, Haoran Yang, Jiacheng Bao, Bin Zhao, Dong Wang
Abstract: The human ability to seamlessly perform multimodal reasoning and physical interaction in the open world is a core goal for general-purpose embodied intelligent systems. Recent vision-language-action (VLA) models, which are co-trained on large-scale robot and visual-text data, have demonstrated notable progress in general robot control. However, they still fail to achieve human-level flexibility in interleaved reasoning and interaction. In this work, introduce EO-Robotics, consists of EO-1 model and EO-Data1.5M dataset. EO-1 is a unified embodied foundation model that achieves superior performance in multimodal embodied reasoning and robot control through interleaved vision-text-action pre-training. The development of EO-1 is based on two key pillars: (i) a unified architecture that processes multimodal inputs indiscriminately (image, text, video, and action), and (ii) a massive, high-quality multimodal embodied reasoning dataset, EO-Data1.5M, which contains over 1.5 million samples with emphasis on interleaved vision-text-action comprehension. EO-1 is trained through synergies between auto-regressive decoding and flow matching denoising on EO-Data1.5M, enabling seamless robot action generation and multimodal embodied reasoning. Extensive experiments demonstrate the effectiveness of interleaved vision-text-action learning for open-world understanding and generalization, validated through a variety of long-horizon, dexterous manipulation tasks across multiple embodiments. This paper details the architecture of EO-1, the data construction strategy of EO-Data1.5M, and the training methodology, offering valuable insights for developing advanced embodied foundation models.
Authors: Guofu Liao, Taotao Wang, Shengli Zhang, Jiqun Zhang, Shi Long, Dacheng Tao
Abstract: Fine-tuning large language models (LLMs) is crucial for adapting them to specific tasks, yet it remains computationally demanding and raises concerns about correctness and privacy, particularly in untrusted environments. Although parameter-efficient methods like Low-Rank Adaptation (LoRA) significantly reduce resource requirements, ensuring the security and verifiability of fine-tuning under zero-knowledge constraints remains an unresolved challenge. To address this, we introduce zkLoRA, the first framework to integrate LoRA fine-tuning with zero-knowledge proofs (ZKPs), achieving provable security and correctness. zkLoRA employs advanced cryptographic techniques -- such as lookup arguments, sumcheck protocols, and polynomial commitments -- to verify both arithmetic and non-arithmetic operations in Transformer-based architectures. The framework provides end-to-end verifiability for forward propagation, backward propagation, and parameter updates during LoRA fine-tuning, while safeguarding the privacy of model parameters and training data. Leveraging GPU-based implementations, zkLoRA demonstrates practicality and efficiency through experimental validation on open-source LLMs like LLaMA, scaling up to 13 billion parameters. By combining parameter-efficient fine-tuning with ZKPs, zkLoRA bridges a critical gap, enabling secure and trustworthy deployment of LLMs in sensitive or untrusted environments.
Authors: W. F. Lamberti, S. R. Lawrence, D. White, S. Kim, S. Abdullah
Abstract: Generative AI (GAI) tools have seen rapid adoption in educational settings, yet their role in fostering critical thinking remains underexplored. While previous studies have examined GAI as a tutor for specific lessons or as a tool for completing assignments, few have addressed how students critically evaluate the accuracy and appropriateness of GAI-generated responses. This pilot study investigates students' ability to apply structured critical thinking when assessing Generative AI outputs in introductory Computational and Data Science courses. Given that GAI tools often produce contextually flawed or factually incorrect answers, we designed learning activities that require students to analyze, critique, and revise AI-generated solutions. Our findings offer initial insights into students' ability to engage critically with GAI content and lay the groundwork for more comprehensive studies in future semesters.
Authors: Kairong Han, Wenshuo Zhao, Ziyu Zhao, JunJian Ye, Lujia Pan, Kun Kuang
Abstract: Large Language Models (LLMs) have achieved remarkable success across various domains. However, a fundamental question remains: Can LLMs effectively utilize causal knowledge for prediction and generation? Through empirical studies, we find that LLMs trained directly on large-scale data often capture spurious correlations rather than true causal relationships, leading to suboptimal performance, especially in out-of-distribution (OOD) scenarios. To address this challenge, we propose Causal Attention Tuning (CAT), a novel approach that injects fine-grained causal knowledge into the attention mechanism. We propose an automated pipeline that leverages human priors to automatically generate token-level causal signals and introduce the Re-Attention mechanism to guide training, helping the model focus on causal structures while mitigating noise and biases in attention scores. Experimental results on our proposed Spurious Token Game (STG) benchmark and multiple downstream tasks demonstrate that our approach effectively leverages causal knowledge for prediction and remains robust in OOD scenarios. The CAT achieves an average improvement of 5.76% on the STG dataset and 1.56% on downstream tasks. Notably, the OOD performance of the Llama-3.1-8B model on STG_M increased from 64.5% to 90.5%, and Qwen's OOD performance on the STG_H dataset improved from 25.4% to 55.9%. Implementation details can be found at https://github.com/Kairong-Han/CAT.
Authors: Akis Nousias, Stavros Nousias
Abstract: Currently, prominent Transformer architectures applied on graphs and meshes for shape analysis tasks employ traditional attention layers that heavily utilize spectral features requiring costly eigenvalue decomposition-based methods. To encode the mesh structure, these methods derive positional embeddings, that heavily rely on eigenvalue decomposition based operations, e.g. on the Laplacian matrix, or on heat-kernel signatures, which are then concatenated to the input features. This paper proposes a novel approach inspired by the explicit construction of the Hodge Laplacian operator in Discrete Exterior Calculus as a product of discrete Hodge operators and exterior derivatives, i.e. $(L := \star_0^{-1} d_0^T \star_1 d_0)$. We adjust the Transformer architecture in a novel deep learning layer that utilizes the multi-head attention mechanism to approximate Hodge matrices $\star_0$, $\star_1$ and $\star_2$ and learn families of discrete operators $L$ that act on mesh vertices, edges and faces. Our approach results in a computationally-efficient architecture that achieves comparable performance in mesh segmentation and classification tasks, through a direct learning framework, while eliminating the need for costly eigenvalue decomposition operations or complex preprocessing operations.
Authors: Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
Abstract: Sparse-reward reinforcement learning (RL) remains fundamentally hard: without structure, any agent needs $\Omega(|\mathcal{S}||\mathcal{A}|/p)$ samples to recover rewards. We introduce Policy-Aware Matrix Completion (PAMC) as a first concrete step toward a structural reward learning framework. Our key idea is to exploit approximate low-rank + sparse structure in the reward matrix, under policy-biased (MNAR) sampling. We prove recovery guarantees with inverse-propensity weighting, and establish a visitation-weighted error-to-regret bound linking completion error to control performance. Importantly, when assumptions weaken, PAMC degrades gracefully: confidence intervals widen and the algorithm abstains, ensuring safe fallback to exploration. Empirically, PAMC improves sample efficiency across Atari-26 (10M steps), DM Control, MetaWorld MT50, D4RL offline RL, and preference-based RL benchmarks, outperforming DrQ-v2, DreamerV3, Agent57, T-REX/D-REX, and PrefPPO under compute-normalized comparisons. Our results highlight PAMC as a practical and principled tool when structural rewards exist, and as a concrete first instantiation of a broader structural reward learning perspective.
Authors: Kyra Wilson, Mattea Sim, Anna-Maria Gueorguieva, Aylin Caliskan
Abstract: In this study, we conduct a resume-screening experiment (N=528) where people collaborate with simulated AI models exhibiting race-based preferences (bias) to evaluate candidates for 16 high and low status occupations. Simulated AI bias approximates factual and counterfactual estimates of racial bias in real-world AI systems. We investigate people's preferences for White, Black, Hispanic, and Asian candidates (represented through names and affinity groups on quality-controlled resumes) across 1,526 scenarios and measure their unconscious associations between race and status using implicit association tests (IATs), which predict discriminatory hiring decisions but have not been investigated in human-AI collaboration. When making decisions without AI or with AI that exhibits no race-based preferences, people select all candidates at equal rates. However, when interacting with AI favoring a particular group, people also favor those candidates up to 90% of the time, indicating a significant behavioral shift. The likelihood of selecting candidates whose identities do not align with common race-status stereotypes can increase by 13% if people complete an IAT before conducting resume screening. Finally, even if people think AI recommendations are low quality or not important, their decisions are still vulnerable to AI bias under certain circumstances. This work has implications for people's autonomy in AI-HITL scenarios, AI and work, design and evaluation of AI hiring systems, and strategies for mitigating bias in collaborative decision-making tasks. In particular, organizational and regulatory policy should acknowledge the complex nature of AI-HITL decision making when implementing these systems, educating people who use them, and determining which are subject to oversight.
Authors: Yuxuan Hu, Jihao Liu, Ke Wang, Jinliang Zhen, Weikang Shi, Manyuan Zhang, Qi Dou, Rui Liu, Aojun Zhou, Hongsheng Li
Abstract: Recent progress in Large Language Models (LLMs) has opened new avenues for solving complex optimization problems, including Neural Architecture Search (NAS). However, existing LLM-driven NAS approaches rely heavily on prompt engineering and domain-specific tuning, limiting their practicality and scalability across diverse tasks. In this work, we propose LM-Searcher, a novel framework that leverages LLMs for cross-domain neural architecture optimization without the need for extensive domain-specific adaptation. Central to our approach is NCode, a universal numerical string representation for neural architectures, which enables cross-domain architecture encoding and search. We also reformulate the NAS problem as a ranking task, training LLMs to select high-performing architectures from candidate pools using instruction-tuning samples derived from a novel pruning-based subspace sampling strategy. Our curated dataset, encompassing a wide range of architecture-performance pairs, encourages robust and transferable learning. Comprehensive experiments demonstrate that LM-Searcher achieves competitive performance in both in-domain (e.g., CNNs for image classification) and out-of-domain (e.g., LoRA configurations for segmentation and generation) tasks, establishing a new paradigm for flexible and generalizable LLM-based architecture search. The datasets and models will be released at https://github.com/Ashone3/LM-Searcher.
Authors: Yuming Li, Yikai Wang, Yuying Zhu, Zhongyu Zhao, Ming Lu, Qi She, Shanghang Zhang
Abstract: Recent advancements in aligning image and video generative models via GRPO have achieved remarkable gains in enhancing human preference alignment. However, these methods still face high computational costs from on-policy rollouts and excessive SDE sampling steps, as well as training instability due to sparse rewards. In this paper, we propose BranchGRPO, a novel method that introduces a branch sampling policy updating the SDE sampling process. By sharing computation across common prefixes and pruning low-reward paths and redundant depths, BranchGRPO substantially lowers the per-update compute cost while maintaining or improving exploration diversity. This work makes three main contributions: (1) a branch sampling scheme that reduces rollout and training cost; (2) a tree-based advantage estimator incorporating dense process-level rewards; and (3) pruning strategies exploiting path and depth redundancy to accelerate convergence and boost performance. Experiments on image and video preference alignment show that BranchGRPO improves alignment scores by 16% over strong baselines, while cutting training time by 50%.
Authors: Christo Mathew, Wentian Wang, Jacob Feldman, Lazaros K. Gallos, Paul B. Kantor, Vladimir Menkov, Hao Wang
Abstract: We investigate reinforcement learning in the Game Of Hidden Rules (GOHR) environment, a complex puzzle in which an agent must infer and execute hidden rules to clear a 6$\times$6 board by placing game pieces into buckets. We explore two state representation strategies, namely Feature-Centric (FC) and Object-Centric (OC), and employ a Transformer-based Advantage Actor-Critic (A2C) algorithm for training. The agent has access only to partial observations and must simultaneously infer the governing rule and learn the optimal policy through experience. We evaluate our models across multiple rule-based and trial-list-based experimental setups, analyzing transfer effects and the impact of representation on learning efficiency.
Authors: Jack Wilkie, Hanan Hindy, Ivan Andonovic, Christos Tachtatzis, Robert Atkinson
Abstract: Malware classification is a contemporary and ongoing challenge in cyber-security: modern obfuscation techniques are able to evade traditional static analysis, while dynamic analysis is too resource intensive to be deployed at a large scale. One prominent line of research addresses these limitations by converting malware binaries into 2D images by heuristically reshaping them into a 2D grid before resizing using Lanczos resampling. These images can then be classified based on their textural information using computer vision approaches. While this approach can detect obfuscated malware more effectively than static analysis, the process of converting files into 2D images results in significant information loss due to both quantisation noise, caused by rounding to integer pixel values, and the introduction of 2D dependencies which do not exist in the original data. This loss of signal limits the classification performance of the downstream model. This work addresses these weaknesses by instead resizing the files into 1D signals which avoids the need for heuristic reshaping, and additionally these signals do not suffer from quantisation noise due to being stored in a floating-point format. It is shown that existing 2D CNN architectures can be readily adapted to classify these 1D signals for improved performance. Furthermore, a bespoke 1D convolutional neural network, based on the ResNet architecture and squeeze-and-excitation layers, was developed to classify these signals and evaluated on the MalNet dataset. It was found to achieve state-of-the-art performance on binary, type, and family level classification with F1 scores of 0.874, 0.503, and 0.507, respectively, paving the way for future models to operate on the proposed signal modality.
Authors: Victor Toscano-Duran, Rocio Gonzalez-Diaz, Miguel A. Guti\'errez-Naranjo
Abstract: While it is well-established that artificial neural networks are universal approximators for continuous functions on compact domains, many modern approaches rely on deep or overparameterized architectures that incur high computational costs. In this paper, a new type of small shallow neural network, called the Barycentric Neural Network (BNN), is proposed, which leverages a fixed set of base points and their barycentric coordinates to define both its structure and its parameters. We demonstrate that our BNN enables the exact representation of continuous piecewise linear functions (CPLFs), ensuring strict continuity across segments. Since any continuous function over a compact domain can be approximated arbitrarily well by CPLFs, the BNN naturally emerges as a flexible and interpretable tool for function approximation. Beyond the use of this representation, the main contribution of the paper is the introduction of a new variant of persistent entropy, a topological feature that is stable and scale invariant, called the length-weighted persistent entropy (LWPE), which is weighted by the lifetime of topological features. Our framework, which combines the BNN with a loss function based on our LWPE, aims to provide flexible and geometrically interpretable approximations of nonlinear continuous functions in resource-constrained settings, such as those with limited base points for BNN design and few training epochs. Instead of optimizing internal weights, our approach directly optimizes the base points that define the BNN. Experimental results show that our approach achieves superior and faster approximation performance compared to classical loss functions such as MSE, RMSE, MAE, and log-cosh.
Authors: Wenxuan Huang, Shuang Chen, Zheyong Xie, Shaosheng Cao, Shixiang Tang, Yufan Shen, Qingyu Yin, Wenbo Hu, Xiaoman Wang, Yuntian Tang, Junbo Qiao, Yue Guo, Yao Hu, Zhenfei Yin, Philip Torr, Yu Cheng, Wanli Ouyang, Shaohui Lin
Abstract: Unified multimodal understanding and generation models recently have achieve significant improvement in image generation capability, yet a large gap remains in instruction following and detail preservation compared to systems that tightly couple comprehension with generation such as GPT-4o. Motivated by recent advances in interleaving reasoning, we explore whether such reasoning can further improve Text-to-Image (T2I) generation. We introduce Interleaving Reasoning Generation (IRG), a framework that alternates between text-based thinking and image synthesis: the model first produces a text-based thinking to guide an initial image, then reflects on the result to refine fine-grained details, visual quality, and aesthetics while preserving semantics. To train IRG effectively, we propose Interleaving Reasoning Generation Learning (IRGL), which targets two sub-goals: (1) strengthening the initial think-and-generate stage to establish core content and base quality, and (2) enabling high-quality textual reflection and faithful implementation of those refinements in a subsequent image. We curate IRGL-300K, a dataset organized into six decomposed learning modes that jointly cover learning text-based thinking, and full thinking-image trajectories. Starting from a unified foundation model that natively emits interleaved text-image outputs, our two-stage training first builds robust thinking and reflection, then efficiently tunes the IRG pipeline in the full thinking-image trajectory data. Extensive experiments show SoTA performance, yielding absolute gains of 5-10 points on GenEval, WISE, TIIF, GenAI-Bench, and OneIG-EN, alongside substantial improvements in visual quality and fine-grained fidelity. The code, model weights and datasets will be released in: https://github.com/Osilly/Interleaving-Reasoning-Generation .
URLs: https://github.com/Osilly/Interleaving-Reasoning-Generation