Authors: Avishek Choudhury, Yeganeh Shahsavar, Hamid Shamszare
Abstract: Large language models (LLMs) increasingly serve as interactive healthcare resources, yet user acceptance remains underexplored. This study examines how ease of use, perceived usefulness, trust, and risk perception interact to shape intentions to adopt DeepSeek, an emerging LLM-based platform, for healthcare purposes. A cross-sectional survey of 556 participants from India, the United Kingdom, and the United States was conducted to measure perceptions and usage patterns. Structural equation modeling assessed both direct and indirect effects, including potential quadratic relationships. Results revealed that trust plays a pivotal mediating role: ease of use exerts a significant indirect effect on usage intentions through trust, while perceived usefulness contributes to both trust development and direct adoption. By contrast, risk perception negatively affects usage intent, emphasizing the importance of robust data governance and transparency. Notably, significant non-linear paths were observed for ease of use and risk, indicating threshold or plateau effects. The measurement model demonstrated strong reliability and validity, supported by high composite reliabilities, average variance extracted, and discriminant validity measures. These findings extend technology acceptance and health informatics research by illuminating the multifaceted nature of user adoption in sensitive domains. Stakeholders should invest in trust-building strategies, user-centric design, and risk mitigation measures to encourage sustained and safe uptake of LLMs in healthcare. Future work can employ longitudinal designs or examine culture-specific variables to further clarify how user perceptions evolve over time and across different regulatory environments. Such insights are critical for harnessing AI to enhance outcomes.
Authors: Michal Bravansky, Vaclav Kubon, Suhas Hariharan, Robert Kirk
Abstract: Interpreting data is central to modern research. Large language models (LLMs) show promise in providing such natural language interpretations of data, yet simple feature extraction methods such as prompting often fail to produce accurate and versatile descriptions for diverse datasets and lack control over granularity and scale. To address these limitations, we propose a domain-agnostic method for dataset featurization that provides precise control over the number of features extracted while maintaining compact and descriptive representations comparable to human expert labeling. Our method optimizes the selection of informative binary features by evaluating the ability of an LLM to reconstruct the original data using those features. We demonstrate its effectiveness in dataset modeling tasks and through two case studies: (1) Constructing a feature representation of jailbreak tactics that compactly captures both the effectiveness and diversity of a larger set of human-crafted attacks; and (2) automating the discovery of features that align with human preferences, achieving accuracy and robustness comparable to expert-crafted features. Moreover, we show that the pipeline scales effectively, improving as additional features are sampled, making it suitable for large and diverse datasets.
Authors: Rylan Schaeffer, Joshua Kazdan, John Hughes, Jordan Juravsky, Sara Price, Aengus Lynch, Erik Jones, Robert Kirk, Azalia Mirhoseini, Sanmi Koyejo
Abstract: Recent research across mathematical problem solving, proof assistant programming and multimodal jailbreaking documents a striking finding: when (multimodal) language model tackle a suite of tasks with multiple attempts per task -- succeeding if any attempt is correct -- then the negative log of the average success rate scales a power law in the number of attempts. In this work, we identify an apparent puzzle: a simple mathematical calculation predicts that on each problem, the failure rate should fall exponentially with the number of attempts. We confirm this prediction empirically, raising a question: from where does aggregate polynomial scaling emerge? We then answer this question by demonstrating per-problem exponential scaling can be made consistent with aggregate polynomial scaling if the distribution of single-attempt success probabilities is heavy tailed such that a small fraction of tasks with extremely low success probabilities collectively warp the aggregate success trend into a power law - even as each problem scales exponentially on its own. We further demonstrate that this distributional perspective explains previously observed deviations from power law scaling, and provides a simple method for forecasting the power law exponent with an order of magnitude lower relative error, or equivalently, ${\sim}2-4$ orders of magnitude less inference compute. Overall, our work contributes to a better understanding of how neural language model performance improves with scaling inference compute and the development of scaling-predictable evaluations of (multimodal) language models.
Authors: Peijie Zhao, Zunayed Arefin, Felipe Meneguzzi, Ramon Fraga Pereira
Abstract: In this demonstration, we develop IntentRec4Maps, a system to recognise users' intentions in interactive maps for real-world navigation. IntentRec4Maps uses the Google Maps Platform as the real-world interactive map, and a very effective approach for recognising users' intentions in real-time. We showcase the recognition process of IntentRec4Maps using two different Path-Planners and a Large Language Model (LLM). GitHub: https://github.com/PeijieZ/IntentRec4Maps
Authors: Lukasz Bartoszcze, Sarthak Munshi, Bryan Sukidi, Jennifer Yen, Zejia Yang, David Williams-King, Linh Le, Kosi Asuzu, Carsten Maple
Abstract: Large-language models are capable of completing a variety of tasks, but remain unpredictable and intractable. Representation engineering seeks to resolve this problem through a new approach utilizing samples of contrasting inputs to detect and edit high-level representations of concepts such as honesty, harmfulness or power-seeking. We formalize the goals and methods of representation engineering to present a cohesive picture of work in this emerging discipline. We compare it with alternative approaches, such as mechanistic interpretability, prompt-engineering and fine-tuning. We outline risks such as performance decrease, compute time increases and steerability issues. We present a clear agenda for future research to build predictable, dynamic, safe and personalizable LLMs.
Authors: Sangwon Seo, Bing Han, Rayan E. Harari, Roger D. Dias, Marco A. Zenati, Eduardo Salas, Vaibhav Unhelkar
Abstract: Coaches are vital for effective collaboration, but cost and resource constraints often limit their availability during real-world tasks. This limitation poses serious challenges in life-critical domains that rely on effective teamwork, such as healthcare and disaster response. To address this gap, we propose and realize an innovative application of AI: task-time team coaching. Specifically, we introduce Socratic, a novel AI system that complements human coaches by providing real-time guidance during task execution. Socratic monitors team behavior, detects misalignments in team members' shared understanding, and delivers automated interventions to improve team performance. We validated Socratic through two human subject experiments involving dyadic collaboration. The results demonstrate that the system significantly enhances team performance with minimal interventions. Participants also perceived Socratic as helpful and trustworthy, supporting its potential for adoption. Our findings also suggest promising directions both for AI research and its practical applications to enhance human teamwork.
Authors: Ruxiao Chen, Chenguang Wang, Yuran Sun, Xilei Zhao, Susu Xu
Abstract: Evacuation decision prediction is critical for efficient and effective wildfire response by helping emergency management anticipate traffic congestion and bottlenecks, allocate resources, and minimize negative impacts. Traditional statistical methods for evacuation decision prediction fail to capture the complex and diverse behavioral logic of different individuals. In this work, for the first time, we introduce FLARE, short for facilitating LLM for advanced reasoning on wildfire evacuation decision prediction, a Large Language Model (LLM)-based framework that integrates behavioral theories and models to streamline the Chain-of-Thought (CoT) reasoning and subsequently integrate with memory-based Reinforcement Learning (RL) module to provide accurate evacuation decision prediction and understanding. Our proposed method addresses the limitations of using existing LLMs for evacuation behavioral predictions, such as limited survey data, mismatching with behavioral theory, conflicting individual preferences, implicit and complex mental states, and intractable mental state-behavior mapping. Experiments on three post-wildfire survey datasets show an average of 20.47% performance improvement over traditional theory-informed behavioral models, with strong cross-event generalizability. Our complete code is publicly available at https://github.com/SusuXu-s-Lab/FLARE
Authors: Akhila Yerukola, Saadia Gabriel, Nanyun Peng, Maarten Sap
Abstract: Gestures are an integral part of non-verbal communication, with meanings that vary across cultures, and misinterpretations that can have serious social and diplomatic consequences. As AI systems become more integrated into global applications, ensuring they do not inadvertently perpetuate cultural offenses is critical. To this end, we introduce Multi-Cultural Set of Inappropriate Gestures and Nonverbal Signs (MC-SIGNS), a dataset of 288 gesture-country pairs annotated for offensiveness, cultural significance, and contextual factors across 25 gestures and 85 countries. Through systematic evaluation using MC-SIGNS, we uncover critical limitations: text-to-image (T2I) systems exhibit strong US-centric biases, performing better at detecting offensive gestures in US contexts than in non-US ones; large language models (LLMs) tend to over-flag gestures as offensive; and vision-language models (VLMs) default to US-based interpretations when responding to universal concepts like wishing someone luck, frequently suggesting culturally inappropriate gestures. These findings highlight the urgent need for culturally-aware AI safety mechanisms to ensure equitable global deployment of AI technologies.
Authors: Shinwoo Park, Hyundong Jin, Jeong-won Cha, Yo-Sub Han
Abstract: Recent progress in large language models (LLMs) for code generation has raised serious concerns about intellectual property protection. Malicious users can exploit LLMs to produce paraphrased versions of proprietary code that closely resemble the original. While the potential for LLM-assisted code paraphrasing continues to grow, research on detecting it remains limited, underscoring an urgent need for detection system. We respond to this need by proposing two tasks. The first task is to detect whether code generated by an LLM is a paraphrased version of original human-written code. The second task is to identify which LLM is used to paraphrase the original code. For these tasks, we construct a dataset LPcode consisting of pairs of human-written code and LLM-paraphrased code using various LLMs. We statistically confirm significant differences in the coding styles of human-written and LLM-paraphrased code, particularly in terms of naming consistency, code structure, and readability. Based on these findings, we develop LPcodedec, a detection method that identifies paraphrase relationships between human-written and LLM-generated code, and discover which LLM is used for the paraphrasing. LPcodedec outperforms the best baselines in two tasks, improving F1 scores by 2.64% and 15.17% while achieving speedups of 1,343x and 213x, respectively.
Authors: Tianyi Zhuang, Chuqiao Kuang, Xiaoguang Li, Yihua Teng, Jihao Wu, Yasheng Wang, Lifeng Shang
Abstract: We present DocPuzzle, a rigorously constructed benchmark for evaluating long-context reasoning capabilities in large language models (LLMs). This benchmark comprises 100 expert-level QA problems requiring multi-step reasoning over long real-world documents. To ensure the task quality and complexity, we implement a human-AI collaborative annotation-validation pipeline. DocPuzzle introduces an innovative evaluation framework that mitigates guessing bias through checklist-guided process analysis, establishing new standards for assessing reasoning capacities in LLMs. Our evaluation results show that: 1)Advanced slow-thinking reasoning models like o1-preview(69.7%) and DeepSeek-R1(66.3%) significantly outperform best general instruct models like Claude 3.5 Sonnet(57.7%); 2)Distilled reasoning models like DeepSeek-R1-Distill-Qwen-32B(41.3%) falls far behind the teacher model, suggesting challenges to maintain the generalization of reasoning capabilities relying solely on distillation.
Authors: Beibei Xiong, Hangyu Lv, Haojia Shan, Jianlin Wang, Zhengfeng Yang, Lihong Zhi
Abstract: Large language models (LLMs) have significantly advanced formal theorem proving, yet the scarcity of high-quality training data constrains their capabilities in complex mathematical domains. Combinatorics, a cornerstone of mathematics, provides essential tools for analyzing discrete structures and solving optimization problems. However, its inherent complexity makes it particularly challenging for automated theorem proving (ATP) for combinatorial identities. To address this, we manually construct LeanComb, combinatorial identities benchmark in Lean, which is, to our knowledge, the first formalized theorem proving benchmark built for combinatorial identities. We develop an Automated Theorem Generator for Combinatorial Identities, ATG4CI, which combines candidate tactics suggested by a self-improving large language model with a Reinforcement Learning Tree Search approach for tactic prediction. By utilizing ATG4CI, we generate a LeanComb-Enhanced dataset comprising 260K combinatorial identities theorems, each with a complete formal proof in Lean, and experimental evaluations demonstrate that models trained on this dataset can generate more effective tactics, thereby improving success rates in automated theorem proving for combinatorial identities.
Authors: Hannah Calzi Kleidermacher, James Zou
Abstract: Scientific research is inherently global. However, the vast majority of academic journals are published exclusively in English, creating barriers for non-native-English-speaking researchers. In this study, we leverage large language models (LLMs) to translate published scientific articles while preserving their native JATS XML formatting, thereby developing a practical, automated approach for implementation by academic journals. Using our approach, we translate articles across multiple scientific disciplines into 28 languages. To evaluate translation accuracy, we introduce a novel question-and-answer (QA) benchmarking method, in which an LLM generates comprehension-based questions from the original text and then answers them based on the translated text. Our benchmark results show an average performance of 95.9%, showing that the key scientific details are accurately conveyed. In a user study, we translate the scientific papers of 15 researchers into their native languages, finding that the authors consistently found the translations to accurately capture the original information in their articles. Interestingly, a third of the authors found many technical terms "overtranslated," expressing a preference to keep terminology more familiar in English untranslated. Finally, we demonstrate how in-context learning techniques can be used to align translations with domain-specific preferences such as mitigating overtranslation, highlighting the adaptability and utility of LLM-driven scientific translation. The code and translated articles are available at https://hankleid.github.io/ProjectMundo.
Authors: Lars Krupp, Daniel Gei{\ss}ler, Paul Lukowicz, Jakob Karolus
Abstract: Improvements in the area of large language models have shifted towards the construction of models capable of using external tools and interpreting their outputs. These so-called web agents have the ability to interact autonomously with the internet. This allows them to become powerful daily assistants handling time-consuming, repetitive tasks while supporting users in their daily activities. While web agent research is thriving, the sustainability aspect of this research direction remains largely unexplored. We provide an initial exploration of the energy and CO2 cost associated with web agents. Our results show how different philosophies in web agent creation can severely impact the associated expended energy. We highlight lacking transparency regarding the disclosure of model parameters and processes used for some web agents as a limiting factor when estimating energy consumption. As such, our work advocates a change in thinking when evaluating web agents, warranting dedicated metrics for energy consumption and sustainability.
Authors: Tahsin Alamgir Kheya, Mohamed Reda Bouadjenek, Sunil Aryal
Abstract: Recommendation systems are now an integral part of our daily lives. We rely on them for tasks such as discovering new movies, finding friends on social media, and connecting job seekers with relevant opportunities. Given their vital role, we must ensure these recommendations are free from societal stereotypes. Therefore, evaluating and addressing such biases in recommendation systems is crucial. Previous work evaluating the fairness of recommended items fails to capture certain nuances as they mainly focus on comparing performance metrics for different sensitive groups. In this paper, we introduce a set of comprehensive metrics for quantifying gender bias in recommendations. Specifically, we show the importance of evaluating fairness on a more granular level, which can be achieved using our metrics to capture gender bias using categories of recommended items like genres for movies. Furthermore, we show that employing a category-aware fairness metric as a regularization term along with the main recommendation loss during training can help effectively minimize bias in the models' output. We experiment on three real-world datasets, using five baseline models alongside two popular fairness-aware models, to show the effectiveness of our metrics in evaluating gender bias. Our metrics help provide an enhanced insight into bias in recommended items compared to previous metrics. Additionally, our results demonstrate how incorporating our regularization term significantly improves the fairness in recommendations for different categories without substantial degradation in overall recommendation performance.
Authors: Suozhi Huang, Peiyang Song, Robert Joseph George, Anima Anandkumar
Abstract: Mathematical reasoning remains a significant challenge for Large Language Models (LLMs) due to hallucinations. When combined with formal proof assistants like Lean, these hallucinations can be eliminated through rigorous verification, making theorem proving reliable. However, even with formal verification, LLMs still struggle with long proofs and complex mathematical formalizations. While Lean with LLMs offers valuable assistance with retrieving lemmas, generating tactics, or even complete proofs, it lacks a crucial capability: providing a sense of proof progress. This limitation particularly impacts the overall development efficiency in large formalization projects. We introduce LeanProgress, a method that predicts the progress in the proof. Training and evaluating our models made on a large corpus of Lean proofs from Lean Workbook Plus and Mathlib4 and how many steps remain to complete it, we employ data preprocessing and balancing techniques to handle the skewed distribution of proof lengths. Our experiments show that LeanProgress achieves an overall prediction accuracy of 75.1\% in predicting the amount of progress and, hence, the remaining number of steps. When integrated into a best-first search framework using Reprover, our method shows a 3.8\% improvement on Mathlib4 compared to baseline performances of 41.2\%, particularly for longer proofs. These results demonstrate how proof progress prediction can enhance both automated and interactive theorem proving, enabling users to make more informed decisions about proof strategies.
Authors: Michele Fiori, Davide Mor, Gabriele Civitarese, Claudio Bettini
Abstract: Sensor-based Human Activity Recognition (HAR) in smart home environments is crucial for several applications, especially in the healthcare domain. The majority of the existing approaches leverage deep learning models. While these approaches are effective, the rationale behind their outputs is opaque. Recently, eXplainable Artificial Intelligence (XAI) approaches emerged to provide intuitive explanations to the output of HAR models. To the best of our knowledge, these approaches leverage classic deep models like CNNs or RNNs. Recently, Graph Neural Networks (GNNs) proved to be effective for sensor-based HAR. However, existing approaches are not designed with explainability in mind. In this work, we propose the first explainable Graph Neural Network explicitly designed for smart home HAR. Our results on two public datasets show that this approach provides better explanations than state-of-the-art methods while also slightly improving the recognition rate.
Authors: Chiara Lindloff, Ingo Siegert
Abstract: The debate around bias in AI systems is central to discussions on algorithmic fairness. However, the term bias often lacks a clear definition, despite frequently being contrasted with fairness, implying that an unbiased model is inherently fair. In this paper, we challenge this assumption and argue that a precise conceptualization of bias is necessary to effectively address fairness concerns. Rather than viewing bias as inherently negative or unfair, we highlight the importance of distinguishing between bias and discrimination. We further explore how this shift in focus can foster a more constructive discourse within academic debates on fairness in AI systems.
Authors: Li Lei, Jia Sen, Wang Jianhao, An Zhaochong, Li Jiaang, Hwang Jenq-Neng, Belongie Serge
Abstract: Advancements in Multimodal Large Language Models (MLLMs) have improved human motion understanding. However, these models remain constrained by their "instruct-only" nature, lacking interactivity and adaptability for diverse analytical perspectives. To address these challenges, we introduce ChatMotion, a multimodal multi-agent framework for human motion analysis. ChatMotion dynamically interprets user intent, decomposes complex tasks into meta-tasks, and activates specialized function modules for motion comprehension. It integrates multiple specialized modules, such as the MotionCore, to analyze human motion from various perspectives. Extensive experiments demonstrate ChatMotion's precision, adaptability, and user engagement for human motion understanding.
Authors: Guoxin Wang, Minyu Gao, Shuai Yang, Ya Zhang, Lizhi He, Liang Huang, Hanlin Xiao, Yexuan Zhang, Wanyue Li, Lu Chen, Jintao Fei, Xin Li
Abstract: Large language models (LLMs), particularly those with reasoning capabilities, have rapidly advanced in recent years, demonstrating significant potential across a wide range of applications. However, their deployment in healthcare, especially in disease reasoning tasks, is hindered by the challenge of acquiring expert-level cognitive data. In this paper, we introduce Citrus, a medical language model that bridges the gap between clinical expertise and AI reasoning by emulating the cognitive processes of medical experts. The model is trained on a large corpus of simulated expert disease reasoning data, synthesized using a novel approach that accurately captures the decision-making pathways of clinicians. This approach enables Citrus to better simulate the complex reasoning processes involved in diagnosing and treating medical conditions.To further address the lack of publicly available datasets for medical reasoning tasks, we release the last-stage training data, including a custom-built medical diagnostic dialogue dataset. This open-source contribution aims to support further research and development in the field. Evaluations using authoritative benchmarks such as MedQA, covering tasks in medical reasoning and language understanding, show that Citrus achieves superior performance compared to other models of similar size. These results highlight Citrus potential to significantly enhance medical decision support systems, providing a more accurate and efficient tool for clinical decision-making.
Authors: Sirisha Velampalli, Chandrashekar Muniyappa
Abstract: The extraction of information from semi-structured text, such as resumes, has long been a challenge due to the diverse formatting styles and subjective content organization. Conventional solutions rely on specialized logic tailored for specific use cases. However, we propose a revolutionary approach leveraging structured Graphs, Natural Language Processing (NLP), and Deep Learning. By abstracting intricate logic into Graph structures, we transform raw data into a comprehensive Knowledge Graph. This innovative framework enables precise information extraction and sophisticated querying. We systematically construct dictionaries assigning skill weights, paving the way for nuanced talent analysis. Our system not only benefits job recruiters and curriculum designers but also empowers job seekers with targeted query-based filtering and ranking capabilities.
Authors: Sepehr Asgarian, Qayam Jetha, Jouhyun Jeon
Abstract: In the competitive landscape of advertising, success hinges on effectively navigating and leveraging complex interactions among consumers, advertisers, and advertisement platforms. These multifaceted interactions compel advertisers to optimize strategies for modeling consumer behavior, enhancing brand recall, and tailoring advertisement content. To address these challenges, we present MindMem, a multimodal predictive model for advertisement memorability. By integrating textual, visual, and auditory data, MindMem achieves state-of-the-art performance, with a Spearman's correlation coefficient of 0.631 on the LAMBDA and 0.731 on the Memento10K dataset, consistently surpassing existing methods. Furthermore, our analysis identified key factors influencing advertisement memorability, such as video pacing, scene complexity, and emotional resonance. Expanding on this, we introduced MindMem-ReAd (MindMem-Driven Re-generated Advertisement), which employs Large Language Model-based simulations to optimize advertisement content and placement, resulting in up to a 74.12% improvement in advertisement memorability. Our results highlight the transformative potential of Artificial Intelligence in advertising, offering advertisers a robust tool to drive engagement, enhance competitiveness, and maximize impact in a rapidly evolving market.
Authors: Minhua Lin, Hui Liu, Xianfeng Tang, Jingying Zeng, Zhenwei Dai, Chen Luo, Zheng Li, Xiang Zhang, Qi He, Suhang Wang
Abstract: Search plays a fundamental role in problem-solving across various domains, with most real-world decision-making problems being solvable through systematic search. Drawing inspiration from recent discussions on search and learning, we systematically explore the complementary relationship between search and Large Language Models (LLMs) from three perspectives. First, we analyze how learning can enhance search efficiency and propose Search via Learning (SeaL), a framework that leverages LLMs for effective and efficient search. Second, we further extend SeaL to SeaL-C to ensure rigorous completeness during search. Our evaluation across three real-world planning tasks demonstrates that SeaL achieves near-perfect accuracy while reducing search spaces by up to 99.1% compared to traditional approaches. Finally, we explore how far LLMs are from real search by investigating whether they can develop search capabilities independently. Our analysis reveals that while current LLMs struggle with efficient search in complex problems, incorporating systematic search strategies significantly enhances their problem-solving capabilities. These findings not only validate the effectiveness of our approach but also highlight the need for improving LLMs' search abilities for real-world applications.
Authors: Jaron Maene, Luc De Raedt
Abstract: Algebraic model counting unifies many inference tasks on logic formulas by exploiting semirings. Rather than focusing on inference, we consider learning, especially in statistical-relational and neurosymbolic AI, which combine logical, probabilistic and neural representations. Concretely, we show that the very same semiring perspective of algebraic model counting also applies to learning. This allows us to unify various learning algorithms by generalizing gradients and backpropagation to different semirings. Furthermore, we show how cancellation and ordering properties of a semiring can be exploited for more memory-efficient backpropagation. This allows us to obtain some interesting variations of state-of-the-art gradient-based optimisation methods for probabilistic logical models. We also discuss why algebraic model counting on tractable circuits does not lead to more efficient second-order optimization. Empirically, our algebraic backpropagation exhibits considerable speed-ups as compared to existing approaches.
Authors: Nils Wandel, David Stotko, Alexander Schier, Reinhard Klein
Abstract: Grading student assignments in STEM courses is a laborious and repetitive task for tutors, often requiring a week to assess an entire class. For students, this delay of feedback prevents iterating on incorrect solutions, hampers learning, and increases stress when exercise scores determine admission to the final exam. Recent advances in AI-assisted education, such as automated grading and tutoring systems, aim to address these challenges by providing immediate feedback and reducing grading workload. However, existing solutions often fall short due to privacy concerns, reliance on proprietary closed-source models, lack of support for combining Markdown, LaTeX and Python code, or excluding course tutors from the grading process. To overcome these limitations, we introduce PyEvalAI, an AI-assisted evaluation system, which automatically scores Jupyter notebooks using a combination of unit tests and a locally hosted language model to preserve privacy. Our approach is free, open-source, and ensures tutors maintain full control over the grading process. A case study demonstrates its effectiveness in improving feedback speed and grading efficiency for exercises in a university-level course on numerics.
Authors: Chanwoo Park, Seungju Han, Xingzhi Guo, Asuman Ozdaglar, Kaiqing Zhang, Joo-Kyung Kim
Abstract: Leveraging multiple large language models (LLMs) to build collaborative multi-agentic workflows has demonstrated significant potential. However, most previous studies focus on prompting the out-of-the-box LLMs, relying on their innate capability for collaboration, which may not improve LLMs' performance as shown recently. In this paper, we introduce a new post-training paradigm MAPoRL (Multi-Agent Post-co-training for collaborative LLMs with Reinforcement Learning), to explicitly elicit the collaborative behaviors and further unleash the power of multi-agentic LLM frameworks. In MAPoRL, multiple LLMs first generate their own responses independently and engage in a multi-turn discussion to collaboratively improve the final answer. In the end, a MAPoRL verifier evaluates both the answer and the discussion, by assigning a score that verifies the correctness of the answer, while adding incentives to encourage corrective and persuasive discussions. The score serves as the co-training reward, and is then maximized through multi-agent RL. Unlike existing LLM post-training paradigms, MAPoRL advocates the co-training of multiple LLMs together using RL for better generalization. Accompanied by analytical insights, our experiments demonstrate that training individual LLMs alone is insufficient to induce effective collaboration. In contrast, multi-agent co-training can boost the collaboration performance across benchmarks, with generalization to unseen domains.
Authors: Donghyun Kim, Sriram Ravula, Taemin Ha, Alexandros G. Dimakis, Daehyeok Kim, Aditya Akella
Abstract: Computer system workload traces, which record hardware or software events during application execution, are essential for understanding the behavior of complex systems and managing their processing and memory resources. However, obtaining real-world traces can be challenging due to the significant collection overheads in performance and privacy concerns that arise in proprietary systems. As a result, synthetic trace generation is considered a promising alternative to using traces collected in real-world production deployments. This paper proposes to train a large language model (LLM) to generate synthetic workload traces, specifically microservice call graphs. To capture complex and arbitrary hierarchical structures and implicit constraints in such traces, we fine-tune LLMs to generate each layer recursively, making call graph generation a sequence of easier steps. To further enforce learning constraints in traces and generate uncommon situations, we apply additional instruction tuning steps to align our model with the desired trace features. Our evaluation results show that our model can generate diverse realistic traces under various conditions and outperform existing methods in accuracy and validity. We show that our synthetically generated traces can effectively substitute real-world data in optimizing or tuning systems management tasks. We also show that our model can be adapted to perform key downstream trace-related tasks, specifically, predicting key trace features and infilling missing data given partial traces. Codes are available in https://github.com/ldos-project/TraceLLM.
Authors: Xiaoqing Zhang, Yuhan Liu, Flood Sung, Xiuying Chen, Rui Yan
Abstract: Code generation is crucial in software engineering for automating the coding process efficiently. While test-time computation methods show promise, they suffer from high latency due to multiple computation rounds. To overcome this, we introduce ThinkCoder, a framework that combines thorough exploration with optimal refinement. The exploration phase diversifies the solution space by searching for potential solutions, followed by a refinement phase that enhances precision. This approach allows us to select the best solution through careful consideration before taking action, avoiding excessive trial and error. To further minimize test-time computation overhead, we introduce preference-driven optimization with Reinforced Self-Training (ReST), which uses exploration trajectories from ThinkCoder to guide LLM's evolution. By learning preferences, this approach improves LLM's exploration efficiency, reducing computational costs while maintaining accuracy. ThinkCoder boosts the performance of multiple base LLMs, excelling on benchmarks like HumanEval and MBPP. Compared to SOTA models, it improves Pass@1 by 1.5\% over MapCoder with just 21.7\% of the computation cost. Against AgentCoder, ThinkCoder achieves a 0.6\% higher Pass@1 after 2 rounds, outperforming AgentCoder's 5 rounds. Additionally, ReST with success trajectories enhances efficiency, allowing models like LLaMA2-7B to achieve competitive results using only 20\% of the computational resources. These results highlight the framework's effectiveness and scalability.
Authors: Vaibhav Tupe, Shrinath Thube
Abstract: The rapid advancement of Generative AI has catalyzed the emergence of autonomous AI agents, presenting unprecedented challenges for enterprise computing infrastructures. Current enterprise API architectures are predominantly designed for human-driven, predefined interaction patterns, rendering them ill-equipped to support intelligent agents' dynamic, goal-oriented behaviors. This research systematically examines the architectural adaptations for enterprise APIs to support AI agentic workflows effectively. Through a comprehensive analysis of existing API design paradigms, agent interaction models, and emerging technological constraints, the paper develops a strategic framework for API transformation. The study employs a mixed-method approach, combining theoretical modeling, comparative analysis, and exploratory design principles to address critical challenges in standardization, performance, and intelligent interaction. The proposed research contributes a conceptual model for next-generation enterprise APIs that can seamlessly integrate with autonomous AI agent ecosystems, offering significant implications for future enterprise computing architectures.
Authors: Xiaowei Jiang, Yanan Chen, Nikhil Ranjan Pal, Yu-Cheng Chang, Yunkai Yang, Thomas Do, Chin-Teng Lin
Abstract: Fuzzy logic provides a robust framework for enhancing explainability, particularly in domains requiring the interpretation of complex and ambiguous signals, such as brain-computer interface (BCI) systems. Despite significant advances in deep learning, interpreting human emotions remains a formidable challenge. In this work, we present iFuzzyAffectDuo, a novel computational model that integrates a dual-filter fuzzy neural network architecture for improved detection and interpretation of emotional states from neuroimaging data. The model introduces a new membership function (MF) based on the Laplace distribution, achieving superior accuracy and interpretability compared to traditional approaches. By refining the extraction of neural signals associated with specific emotions, iFuzzyAffectDuo offers a human-understandable framework that unravels the underlying decision-making processes. We validate our approach across three neuroimaging datasets using functional Near-Infrared Spectroscopy (fNIRS) and Electroencephalography (EEG), demonstrating its potential to advance affective computing. These findings open new pathways for understanding the neural basis of emotions and their application in enhancing human-computer interaction.
Authors: Xiaolin Li, Binhua Huang, Barry Cardiff, Deepu John
Abstract: DCentNet is a novel decentralized multistage signal classification approach designed for biomedical data from IoT wearable sensors, integrating early exit points (EEP) to enhance energy efficiency and processing speed. Unlike traditional centralized processing methods, which result in high energy consumption and latency, DCentNet partitions a single CNN model into multiple sub-networks using EEPs. By introducing encoder-decoder pairs at EEPs, the system compresses large feature maps before transmission, significantly reducing wireless data transfer and power usage. If an input is confidently classified at an EEP, processing stops early, optimizing efficiency. Initial sub-networks can be deployed on fog or edge devices to further minimize energy consumption. A genetic algorithm is used to optimize EEP placement, balancing performance and complexity. Experimental results on ECG classification show that with one EEP, DCentNet reduces wireless data transmission by 94.54% and complexity by 21%, while maintaining original accuracy and sensitivity. With two EEPs, sensitivity reaches 98.36%, accuracy 97.74%, wireless data transmission decreases by 91.86%, and complexity is reduced by 22%. Implemented on an ARM Cortex-M4 MCU, DCentNet achieves an average power saving of 73.6% compared to continuous wireless ECG transmission.
Authors: David Noever, Forrest McKee
Abstract: This study addresses challenges in reverse logistics, a frequently overlooked but essential component of last-mile delivery, particularly in disaster relief scenarios where infrastructure disruptions demand adaptive solutions. While hub-and-spoke logistics networks excel at long-distance scalability, they often fail to optimize closely spaced spokes reliant on distant hubs, introducing inefficiencies in transit times and resource allocation. Using 20 Apple AirTags embedded in packages, this research provides empirical insights into logistical flows, capturing granular spatial and temporal data through Bluetooth LE (BLE) 5 trackers integrated with the Apple Find My network. These trackers demonstrated their value in monitoring dynamic cargo movements, enabling real-time adjustments in mobile hub placement and route optimization, particularly in disaster relief contexts like Hurricane Helene. A novel application of discrete event simulation (DES) further explored the saddle point in hub-spoke configurations, where excessive hub reliance clashes with diminishing spoke interaction demand. By coupling simulation results with empirical AirTag tracking, the study highlights the potential of BLE technology to refine reverse logistics, reduce delays, and improve operational flexibility in both routine and crisis-driven delivery networks.
Authors: Marcos Soto (Universidad Loyola Andaluc\'ia)
Abstract: In industrial environments, data acquisition accuracy is crucial for process control and optimization. Wireless telemetry has proven to be a valuable tool for improving efficiency in well-testing operations, enabling bidirectional communication and real-time control of downhole tools. However, high sampling frequencies present challenges in telemetry, including data storage, transmission, computational resource consumption, and battery life of wireless devices. This study explores how optimizing data acquisition strategies can reduce aliasing effects and systematic errors while improving sampling rates without compromising measurement accuracy. A reduction of 80% in sampling frequency was achieved without degrading measurement quality, demonstrating the potential for resource optimization in industrial environments.
Authors: Ding Hu, Pengxiang Hua, Zhen Huang
Abstract: The development of artificial intelligence (AI) techniques has brought revolutionary changes across various realms. In particular, the use of AI-assisted methods to accelerate chemical research has become a popular and rapidly growing trend, leading to numerous groundbreaking works. In this paper, we provide a comprehensive review of current AI techniques in chemistry from a computational perspective, considering various aspects in the design of methods. We begin by discussing the characteristics of data from diverse sources, followed by an overview of various representation methods. Next, we review existing models for several topical tasks in the field, and conclude by highlighting some key challenges that warrant further attention.
Authors: Mehran Shabanpour, Kasra Rad, Sadaf Khademi, Arash Mohammadi
Abstract: High-Density surface Electromyography (HDsEMG) has emerged as a pivotal resource for Human-Computer Interaction (HCI), offering direct insights into muscle activities and motion intentions. However, a significant challenge in practical implementations of HD-sEMG-based models is the low accuracy of inter-session and inter-subject classification. Variability between sessions can reach up to 40% due to the inherent temporal variability of HD-sEMG signals. Targeting this challenge, the paper introduces the MoEMba framework, a novel approach leveraging Selective StateSpace Models (SSMs) to enhance HD-sEMG-based gesture recognition. The MoEMba framework captures temporal dependencies and cross-channel interactions through channel attention techniques. Furthermore, wavelet feature modulation is integrated to capture multi-scale temporal and spatial relations, improving signal representation. Experimental results on the CapgMyo HD-sEMG dataset demonstrate that MoEMba achieves a balanced accuracy of 56.9%, outperforming its state-of-the-art counterparts. The proposed framework's robustness to session-to-session variability and its efficient handling of high-dimensional multivariate time series data highlight its potential for advancing HD-sEMG-powered HCI systems.
Authors: B\'alint T\'oth, Dominik Senti, Thorir Mar Ingolfsson, Jeffrey Zweidler, Alexandre Elsig, Luca Benini, Yawei Li
Abstract: Blood pressure (BP) is a key indicator of cardiovascular health. As hypertension remains a global cause of morbidity and mortality, accurate, continuous, and non-invasive BP monitoring is therefore of paramount importance. Photoplethysmography (PPG) and electrocardiography (ECG) can potentially enable continuous BP monitoring, yet training accurate and robust machine learning (ML) models remains challenging due to variability in data quality and patient-specific factors. Recently, multiple research groups explored Electroencephalographic (EEG)--based foundation models and demonstrated their exceptional ability to learn rich temporal resolution. Considering the morphological similarities between different biosignals, the question arises of whether a model pre-trained on one modality can effectively be exploited to improve the accuracy of a different signal type. In this work, we take an initial step towards generalized biosignal foundation models by investigating whether model representations learned from abundant EEG data can effectively be transferred to ECG/PPG data solely with fine-tuning, without the need for large-scale additional pre-training, for the BP estimation task. Evaluations on the MIMIC-III and VitalDB datasets demonstrate that our approach achieves near state-of-the-art accuracy for diastolic BP (mean absolute error of 1.57 mmHg) and surpasses by 1.5x the accuracy of prior works for systolic BP (mean absolute error 2.72 mmHg). Additionally, we perform dynamic INT8 quantization, reducing the smallest model size by over 3.5x (from 13.73 MB down to 3.83 MB) while preserving performance, thereby enabling unobtrusive, real-time BP monitoring on resource-constrained wearable devices.
Authors: Francesco Stefano Carzaniga, Gary Tom Hoppeler, Michael Hersche, Kaspar Anton Schindler, Abbas Rahimi
Abstract: All data modalities are not created equal, even when the signal they measure comes from the same source. In the case of the brain, two of the most important data modalities are the scalp electroencephalogram (EEG), and the intracranial electroencephalogram (iEEG). They are used by human experts, supported by deep learning (DL) models, to accomplish a variety of tasks, such as seizure detection and motor imagery classification. Although the differences between EEG and iEEG are well understood by human experts, the performance of DL models across these two modalities remains under-explored. To help characterize the importance of clean data on the performance of DL models, we propose BrainCodec, a high-fidelity EEG and iEEG neural compressor. We find that training BrainCodec on iEEG and then transferring to EEG yields higher reconstruction quality than training on EEG directly. In addition, we also find that training BrainCodec on both EEG and iEEG improves fidelity when reconstructing EEG. Our work indicates that data sources with higher SNR, such as iEEG, provide better performance across the board also in the medical time-series domain. BrainCodec also achieves up to a 64x compression on iEEG and EEG without a notable decrease in quality. BrainCodec markedly surpasses current state-of-the-art compression models both in final compression ratio and in reconstruction fidelity. We also evaluate the fidelity of the compressed signals objectively on a seizure detection and a motor imagery task performed by standard DL models. Here, we find that BrainCodec achieves a reconstruction fidelity high enough to ensure no performance degradation on the downstream tasks. Finally, we collect the subjective assessment of an expert neurologist, that confirms the high reconstruction quality of BrainCodec in a realistic scenario. The code is available at https://github.com/IBM/eeg-ieeg-brain-compressor.
Authors: Younghoon Na
Abstract: The analysis of lifelogs can yield valuable insights into an individual's daily life, particularly with regard to their health and well-being. The accurate assessment of quality of life is necessitated by the use of diverse sensors and precise synchronization. To rectify this issue, this study proposes the image-based sleep quality and stress level estimation flow (PixleepFlow). PixleepFlow employs a conversion methodology into composite image data to examine sleep patterns and their impact on overall health. Experiments were conducted using lifelog datasets to ascertain the optimal combination of data formats. In addition, we identified which sensor information has the greatest influence on the quality of life through Explainable Artificial Intelligence(XAI). As a result, PixleepFlow produced more significant results than various data formats. This study was part of a written-based competition, and the additional findings from the lifelog dataset are detailed in Section Section IV. More information about PixleepFlow can be found at https://github.com/seongjiko/Pixleep.
Authors: Younghoon Na
Abstract: Sleep profoundly affects our health, and sleep deficiency or disorders can cause physical and mental problems. % Despite significant findings from previous studies, challenges persist in optimizing deep learning models, especially in multi-modal learning for high-accuracy sleep stage classification. Our research introduces MC2SleepNet (Multi-modal Cross-masking with Contrastive learning for Sleep stage classification Network). It aims to facilitate the effective collaboration between Convolutional Neural Networks (CNNs) and Transformer architectures for multi-modal training with the help of contrastive learning and cross-masking. % Raw single channel EEG signals and corresponding spectrogram data provide differently characterized modalities for multi-modal learning. Our MC2SleepNet has achieved state-of-the-art performance with an accuracy of both 84.6% on the SleepEDF-78 and 88.6% accuracy on the Sleep Heart Health Study (SHHS). These results demonstrate the effective generalization of our proposed network across both small and large datasets.
Authors: Xu Wang, Jiaju Kang, Puyu Han
Abstract: We present ECG-Expert-QA, a comprehensive multimodal dataset designed for evaluating diagnostic capabilities in ECG interpretation, integrating real clinical data with systematically generated synthetic cases. The dataset encompasses six fundamental diagnostic tasks, comprising 47,211 meticulously curated question-answer pairs that span a spectrum of clinical scenarios, from basic rhythm analysis to complex case interpretation. By simulating challenging clinical cases through a rigorous medical knowledge-guided process, ECG-Expert-QA not only enhances the availability of annotated diagnostic data but also significantly increases the complexity and diversity of clinical presentations, including rare cardiac conditions and temporal progression patterns. This design enables comprehensive evaluation of medical language models across multiple dimensions, including diagnostic accuracy, clinical reasoning, and knowledge integration. To facilitate global research collaboration, ECG-Expert-QA is available in both Chinese and English versions, with rigorous quality control ensuring linguistic and clinical consistency. The dataset's challenging diagnostic tasks, which include interpretation of complex arrhythmias, identification of subtle ischemic changes, and integration of clinical context, establish it as an effective benchmark for advancing AI-assisted ECG interpretation and pushing the boundaries of current diagnostic models. Our dataset is open-source and available at https://github.com/Zaozzz/ECG-Expert-QA.
Authors: Jarod L\'evy, Mingfang Zhang, Svetlana Pinet, J\'er\'emy Rapin, Hubert Banville, St\'ephane d'Ascoli, Jean-R\'emi King
Abstract: Modern neuroprostheses can now restore communication in patients who have lost the ability to speak or move. However, these invasive devices entail risks inherent to neurosurgery. Here, we introduce a non-invasive method to decode the production of sentences from brain activity and demonstrate its efficacy in a cohort of 35 healthy volunteers. For this, we present Brain2Qwerty, a new deep learning architecture trained to decode sentences from either electro- (EEG) or magneto-encephalography (MEG), while participants typed briefly memorized sentences on a QWERTY keyboard. With MEG, Brain2Qwerty reaches, on average, a character-error-rate (CER) of 32% and substantially outperforms EEG (CER: 67%). For the best participants, the model achieves a CER of 19%, and can perfectly decode a variety of sentences outside of the training set. While error analyses suggest that decoding depends on motor processes, the analysis of typographical errors suggests that it also involves higher-level cognitive factors. Overall, these results narrow the gap between invasive and non-invasive methods and thus open the path for developing safe brain-computer interfaces for non-communicating patients.
Authors: Cheol-Hui Lee, Hakseung Kim, Byung C. Yoon, Dong-Joo Kim
Abstract: Sleep is essential for maintaining human health and quality of life. Analyzing physiological signals during sleep is critical in assessing sleep quality and diagnosing sleep disorders. However, manual diagnoses by clinicians are time-intensive and subjective. Despite advances in deep learning that have enhanced automation, these approaches remain heavily dependent on large-scale labeled datasets. This study introduces SynthSleepNet, a multimodal hybrid self-supervised learning framework designed for analyzing polysomnography (PSG) data. SynthSleepNet effectively integrates masked prediction and contrastive learning to leverage complementary features across multiple modalities, including electroencephalogram (EEG), electrooculography (EOG), electromyography (EMG), and electrocardiogram (ECG). This approach enables the model to learn highly expressive representations of PSG data. Furthermore, a temporal context module based on Mamba was developed to efficiently capture contextual information across signals. SynthSleepNet achieved superior performance compared to state-of-the-art methods across three downstream tasks: sleep-stage classification, apnea detection, and hypopnea detection, with accuracies of 89.89%, 99.75%, and 89.60%, respectively. The model demonstrated robust performance in a semi-supervised learning environment with limited labels, achieving accuracies of 87.98%, 99.37%, and 77.52% in the same tasks. These results underscore the potential of the model as a foundational tool for the comprehensive analysis of PSG data. SynthSleepNet demonstrates comprehensively superior performance across multiple downstream tasks compared to other methodologies, making it expected to set a new standard for sleep disorder monitoring and diagnostic systems.
Authors: Thomas Screven, Andras Necz, Jason Smucny, Ian Davidson
Abstract: Although great advances in the analysis of neuroimaging data have been made, a major challenge is a lack of training data. This is less problematic in tasks such as diagnosis, where much data exists, but particularly prevalent in harder problems such as predicting treatment responses (prognosis), where data is focused and hence limited. Here, we address the learning from small data problems for medical imaging using graph neural networks. This is particularly challenging as the information about the patients is themselves graphs (regions of interest connectivity graphs). We show how a spectral representation of the connectivity data allows for efficient propagation that can yield approximately 12\% improvement over traditional deep learning methods using the exact same data. We show that our method's superior performance is due to a data smoothing result that can be measured by closing the number of triangle inequalities and thereby satisfying transitivity.
Authors: Hagen Holthusen, Kevin Linka, Ellen Kuhl, Tim Brepols
Abstract: We present a methodology for designing a generalized dual potential, or pseudo potential, for inelastic Constitutive Artificial Neural Networks (iCANNs). This potential, expressed in terms of stress invariants, inherently satisfies thermodynamic consistency for large deformations. In comparison to our previous work, the new potential captures a broader spectrum of material behaviors, including pressure-sensitive inelasticity. To this end, we revisit the underlying thermodynamic framework of iCANNs for finite strain inelasticity and derive conditions for constructing a convex, zero-valued, and non-negative dual potential. To embed these principles in a neural network, we detail the architecture's design, ensuring a priori compliance with thermodynamics. To evaluate the proposed architecture, we study its performance and limitations discovering visco-elastic material behavior, though the method is not limited to visco-elasticity. In this context, we investigate different aspects in the strategy of discovering inelastic materials. Our results indicate that the novel architecture robustly discovers interpretable models and parameters, while autonomously revealing the degree of inelasticity. The iCANN framework, implemented in JAX, is publicly accessible at https://doi.org/10.5281/zenodo.14894687.
Authors: Ruoyu Guo, Haochen Qiu
Abstract: Making consistently profitable financial decisions in a continuously evolving and volatile stock market has always been a difficult task. Professionals from different disciplines have developed foundational theories to anticipate price movement and evaluate securities such as the famed Capital Asset Pricing Model (CAPM). In recent years, the role of artificial intelligence (AI) in asset pricing has been growing. Although the black-box nature of deep learning models lacks interpretability, they have continued to solidify their position in the financial industry. We aim to further enhance AI's potential and utility by introducing a return-weighted loss function that will drive top growth while providing the ML models a limited amount of information. Using only publicly accessible stock data (open/close/high/low, trading volume, sector information) and several technical indicators constructed from them, we propose an efficient daily trading system that detects top growth opportunities. Our best models achieve 61.73% annual return on daily rebalancing with an annualized Sharpe Ratio of 1.18 over 1340 testing days from 2019 to 2024, and 37.61% annual return with an annualized Sharpe Ratio of 0.97 over 1360 testing days from 2005 to 2010. The main drivers for success, especially independent of any domain knowledge, are the novel return-weighted loss function, the integration of categorical and continuous data, and the ML model architecture. We also demonstrate the superiority of our novel loss function over traditional loss functions via several performance metrics and statistical evidence.
Authors: Mingfu Liang, Xi Liu, Rong Jin, Boyang Liu, Qiuling Suo, Qinghai Zhou, Song Zhou, Laming Chen, Hua Zheng, Zhiyuan Li, Shali Jiang, Jiyan Yang, Xiaozhen Xia, Fan Yang, Yasmine Badr, Ellie Wen, Shuyu Xu, Hansey Chen, Zhengyu Zhang, Jade Nie, Chunzhi Yang, Zhichen Zeng, Weilin Zhang, Xingliang Huang, Qianru Li, Shiquan Wang, Evelyn Lyu, Wenjing Lu, Rui Zhang, Wenjun Wang, Jason Rudy, Mengyue Hang, Kai Wang, Yinbin Ma, Shuaiwen Wang, Sihan Zeng, Tongyi Tang, Xiaohan Wei, Longhao Jin, Jamey Zhang, Marcus Chen, Jiayi Zhang, Angie Huang, Chi Zhang, Zhengli Zhao, Jared Yang, Qiang Jin, Xian Chen, Amit Anand Amlesahwaram, Lexi Song, Liang Luo, Yuchen Hao, Nan Xiao, Yavuz Yetim, Luoshang Pan, Gaoxiang Liu, Yuxi Hu, Yuzhen Huang, Jackie Xu, Rich Zhu, Xin Zhang, Yiqun Liu, Hang Yin, Yuxin Chen, Buyun Zhang, Xiaoyi Liu, Sylvia Wang, Wenguang Mao, Zhijing Li, Qin Huang, Chonglin Sun, Shupin Mao, Jingzheng Qin, Peggy Yao, Jae-Woo Choi, Bin Gao, Ernest Wang, Lei Zhang, Wen-Yen Chen, Ted Lee, Jay Zha, Yi Meng, Alex Gong, Edison Gao, Alireza Vahdatpour, Yiping Han, Yantao Yao, Toshinari Kureha, Shuo Chang, Musharaf Sultan, John Bocharov, Sagar Chordia, Xiaorui Gan, Peng Sun, Rocky Liu, Bo Long, Wenlin Chen, Santanu Kolay, Huayu Li
Abstract: Ads recommendation is a prominent service of online advertising systems and has been actively studied. Recent studies indicate that scaling-up and advanced design of the recommendation model can bring significant performance improvement. However, with a larger model scale, such prior studies have a significantly increasing gap from industry as they often neglect two fundamental challenges in industrial-scale applications. First, training and inference budgets are restricted for the model to be served, exceeding which may incur latency and impair user experience. Second, large-volume data arrive in a streaming mode with data distributions dynamically shifting, as new users/ads join and existing users/ads leave the system. We propose the External Large Foundation Model (ExFM) framework to address the overlooked challenges. Specifically, we develop external distillation and a data augmentation system (DAS) to control the computational cost of training/inference while maintaining high performance. We design the teacher in a way like a foundation model (FM) that can serve multiple students as vertical models (VMs) to amortize its building cost. We propose Auxiliary Head and Student Adapter to mitigate the data distribution gap between FM and VMs caused by the streaming data issue. Comprehensive experiments on internal industrial-scale applications and public datasets demonstrate significant performance gain by ExFM.
Authors: Tokey Tahmid, Mark Gates, Piotr Luszczek, Catherine D. Schuman
Abstract: In this era of AI revolution, massive investments in large-scale data-driven AI systems demand high-performance computing, consuming tremendous energy and resources. This trend raises new challenges in optimizing sustainability without sacrificing scalability or performance. Among the energy-efficient alternatives of the traditional Von Neumann architecture, neuromorphic computing and its Spiking Neural Networks (SNNs) are a promising choice due to their inherent energy efficiency. However, in some real-world application scenarios such as complex continuous control tasks, SNNs often lack the performance optimizations that traditional artificial neural networks have. Researchers have addressed this by combining SNNs with Deep Reinforcement Learning (DeepRL), yet scalability remains unexplored. In this paper, we extend our previous work on SpikeRL, which is a scalable and energy efficient framework for DeepRL-based SNNs for continuous control. In our initial implementation of SpikeRL framework, we depended on the population encoding from the Population-coded Spiking Actor Network (PopSAN) method for our SNN model and implemented distributed training with Message Passing Interface (MPI) through mpi4py. Also, further optimizing our model training by using mixed-precision for parameter updates. In our new SpikeRL framework, we have implemented our own DeepRL-SNN component with population encoding, and distributed training with PyTorch Distributed package with NCCL backend while still optimizing with mixed precision training. Our new SpikeRL implementation is 4.26X faster and 2.25X more energy efficient than state-of-the-art DeepRL-SNN methods. Our proposed SpikeRL framework demonstrates a truly scalable and sustainable solution for complex continuous control tasks in real-world applications.
Authors: Zetian Sun, Dongfang Li, Baotian Hu, Jun Yu, Min Zhang
Abstract: In the Large Language Model(LLM) reasoning scenario, people often estimate state value via Monte Carlo sampling. Though Monte Carlo estimation is an elegant method with less inductive bias, noise and errors are inevitably introduced due to the limited sampling. To handle the problem, we inject the structural prior into the value representation and transfer the scalar value into the expectation of a pre-defined categorical distribution, representing the noise and errors from a distribution perspective. Specifically, by treating the result of Monte Carlo sampling as a single sample from the prior ground-truth Binomial distribution, we quantify the sampling error as the mismatch between posterior estimated distribution and ground-truth distribution, which is thus optimized via distribution selection optimization. We test the performance of value-based process verifiers on Best-of-N task and Beam search task. Compared with the scalar value representation, we show that reasonable structural prior injection induced by different objective functions or optimization methods can improve the performance of value-based process verifiers for about 1$\sim$2 points at little-to-no cost. We also show that under different structural prior, the verifiers' performances vary greatly despite having the same optimal solution, indicating the importance of reasonable structural prior injection.
Authors: Sumei Fan, Deyun Zhang, Yue Wang, Shijia Geng, Kun Lu, Meng Sang, Weilun Xu, Haixue Wang, Qinghao Zhao, Chuandong Cheng, Peng Wang, Shenda Hong
Abstract: In recent years, wearable devices have revolutionized cardiac monitoring by enabling continuous, non-invasive ECG recording in real-world settings. Despite these advances, the accuracy of ECG parameter calculations (PR interval, QRS interval, QT interval, etc.) from wearables remains to be rigorously validated against conventional ECG machines and expert clinician assessments. In this large-scale, multicenter study, we evaluated FeatureDB, a novel algorithm for automated computation of ECG parameters from wearable single-lead signals Three diverse datasets were employed: the AHMU-FH dataset (n=88,874), the CSE dataset (n=106), and the HeartVoice-ECG-lite dataset (n=369) with annotations provided by two experienced cardiologists. FeatureDB demonstrates a statistically significant correlation with key parameters (PR interval, QRS duration, QT interval, and QTc) calculated by standard ECG machines and annotated by clinical doctors. Bland-Altman analysis confirms a high level of agreement.Moreover,FeatureDB exhibited robust diagnostic performance in detecting Long QT syndrome (LQT) and atrioventricular block interval abnormalities (AVBI),with excellent area under the ROC curve (LQT: 0.836, AVBI: 0.861),accuracy (LQT: 0.856, AVBI: 0.845),sensitivity (LQT: 0.815, AVBI: 0.877),and specificity (LQT: 0.856, AVBI: 0.845).This further validates its clinical reliability. These results validate the clinical applicability of FeatureDB for wearable ECG analysis and highlight its potential to bridge the gap between traditional diagnostic methods and emerging wearable technologies.Ultimately,this study supports integrating wearable ECG devices into large-scale cardiovascular disease management and early intervention strategies,and it highlights the potential of wearable ECG technologies to deliver accurate,clinically relevant cardiac monitoring while advancing broader applications in cardiovascular care.
Authors: Andrzej Cichocki
Abstract: In this paper we propose and investigate a new class of Generalized Exponentiated Gradient (GEG) algorithms using Mirror Descent (MD) approaches, and applying as a regularization function the Bregman divergence with two-parameter deformation of logarithm as a link function. This link function (referred to as the Euler logarithm) is associated with a wide class of generalized entropies. In order to derive novel GEG/MD updates, we estimate generalized exponential function, which closely approximates the inverse of the Euler two-parameter logarithm. The characteristic/shape and properties of the Euler logarithm and its inverse -- deformed exponential functions are tuned by two or even more hyperparameters. By learning these hyperparameters, we can adapt to distribution of training data, and we can adjust them to achieve desired properties of gradient descent algorithms. The concept of generalized entropies and associated deformed logarithms provide deeper insight into novel gradient descent updates. In literature, there exist nowadays over fifty mathematically well-defined entropic functionals and associated deformed logarithms, so impossible to investigate all of them in one research paper. Therefore, we focus here on a wide-class of trace-form entropies and associated generalized logarithm. We applied the developed algorithms for Online Portfolio Selection (OPLS) in order to improve its performance and robustness.
Authors: Qiheng Sun, Hongwei Zhang, Haocheng Xia, Jiayao Zhang, Jinfei Liu, Kui Ren
Abstract: Large language models (LLMs) have achieved remarkable success on various aspects of human life. However, one of the major challenges in deploying these models is the substantial memory consumption required to store key-value pairs (KV), which imposes significant resource demands. Recent research has focused on KV cache budget allocation, with several approaches proposing head-level budget distribution by evaluating the importance of individual attention heads. These methods, however, assess the importance of heads independently, overlooking their cooperative contributions within the model, which may result in a deviation from their true impact on model performance. In light of this limitation, we propose CoKV, a novel method that models the cooperation between heads in model inference as a cooperative game. By evaluating the contribution of each head within the cooperative game, CoKV can allocate the cache budget more effectively. Extensive experiments show that CoKV achieves state-of-the-art performance on the LongBench benchmark using LLama-3-8B-Instruct and Mistral-7B models.
Authors: Alice Natalina Caragliano, Filippo Ruffini, Carlo Greco, Edy Ippolito, Michele Fiore, Claudia Tacconi, Lorenzo Nibid, Giuseppe Perrone, Sara Ramella, Paolo Soda, Valerio Guarrasi
Abstract: Non-small cell lung cancer (NSCLC) remains a major global health challenge, with high post-surgical recurrence rates underscoring the need for accurate pathological response predictions to guide personalized treatments. Although artificial intelligence models show promise in this domain, their clinical adoption is limited by the lack of medically grounded guidance during training, often resulting in non-explainable intrinsic predictions. To address this, we propose Doctor-in-the-Loop, a novel framework that integrates expert-driven domain knowledge with explainable artificial intelligence techniques, directing the model toward clinically relevant anatomical regions and improving both interpretability and trustworthiness. Our approach employs a gradual multi-view strategy, progressively refining the model's focus from broad contextual features to finer, lesion-specific details. By incorporating domain insights at every stage, we enhance predictive accuracy while ensuring that the model's decision-making process aligns more closely with clinical reasoning. Evaluated on a dataset of NSCLC patients, Doctor-in-the-Loop delivers promising predictive performance and provides transparent, justifiable outputs, representing a significant step toward clinically explainable artificial intelligence in oncology.
Authors: Yijia Xiao, Wanjia Zhao, Junkai Zhang, Yiqiao Jin, Han Zhang, Zhicheng Ren, Renliang Sun, Haixin Wang, Guancheng Wan, Pan Lu, Xiao Luo, Yu Zhang, James Zou, Yizhou Sun, Wei Wang
Abstract: Protein-specific large language models (Protein LLMs) are revolutionizing protein science by enabling more efficient protein structure prediction, function annotation, and design. While existing surveys focus on specific aspects or applications, this work provides the first comprehensive overview of Protein LLMs, covering their architectures, training datasets, evaluation metrics, and diverse applications. Through a systematic analysis of over 100 articles, we propose a structured taxonomy of state-of-the-art Protein LLMs, analyze how they leverage large-scale protein sequence data for improved accuracy, and explore their potential in advancing protein engineering and biomedical research. Additionally, we discuss key challenges and future directions, positioning Protein LLMs as essential tools for scientific discovery in protein science. Resources are maintained at https://github.com/Yijia-Xiao/Protein-LLM-Survey.
Authors: Enrico Ahlers, Peter Feuer-Forson, Gregor Hartmann, Rolf Mitzner, Peter Baumg\"artel, Jens Viefhaus
Abstract: In this study, we present a method to create a robust inverse surrogate model for a soft X-ray spectrometer. During a beamtime at an electron storage ring, such as BESSY II, instrumentation and beamlines are required to be correctly aligned and calibrated for optimal experimental conditions. In order to automate these processes, machine learning methods can be developed and implemented, but in many cases these methods require the use of an inverse model which maps the output of the experiment, such as a detector image, to the parameters of the device. Due to limited experimental data, such models are often trained with simulated data, which creates the challenge of compensating for the inherent differences between simulation and experiment. In order to close this gap, we demonstrate the application of data augmentation and adversarial domain adaptation techniques, with which we can predict absolute coordinates for the automated alignment of our spectrometer. Bridging the simulation-experiment gap with minimal real-world data opens new avenues for automated experimentation using machine learning in scientific instrumentation.
Authors: Namkyeong Lee, Edward De Brouwer, Ehsan Hajiramezanali, Chanyoung Park, Gabriele Scalia
Abstract: Recent advances in large language models (LLMs) have shown great potential to accelerate drug discovery. However, the specialized nature of biochemical data often necessitates costly domain-specific fine-tuning, posing critical challenges. First, it hinders the application of more flexible general-purpose LLMs in cutting-edge drug discovery tasks. More importantly, it impedes the rapid integration of the vast amounts of scientific data continuously generated through experiments and research. To investigate these challenges, we propose CLADD, a retrieval-augmented generation (RAG)-empowered agentic system tailored to drug discovery tasks. Through the collaboration of multiple LLM agents, CLADD dynamically retrieves information from biomedical knowledge bases, contextualizes query molecules, and integrates relevant evidence to generate responses -- all without the need for domain-specific fine-tuning. Crucially, we tackle key obstacles in applying RAG workflows to biochemical data, including data heterogeneity, ambiguity, and multi-source integration. We demonstrate the flexibility and effectiveness of this framework across a variety of drug discovery tasks, showing that it outperforms general-purpose and domain-specific LLMs as well as traditional deep learning approaches.
Authors: Kavosh Asadi, Julien Han, Xingzi Xu, Dominique Perrault-Joncas, Shoham Sabach, Karim Bouyarmane, Mohammad Ghavamzadeh
Abstract: Direct preference optimization (DPO)-style algorithms have emerged as a promising approach for solving the alignment problem in AI. We present a novel perspective that formulates these algorithms as implicit classification algorithms. This classification framework enables us to recover many variants of DPO-style algorithms by choosing appropriate classification labels and loss functions. We then leverage this classification framework to demonstrate that the underlying problem solved in these algorithms is under-specified, making them susceptible to probability collapse of the winner-loser responses. We address this by proposing a set of constraints designed to control the movement of probability mass between the winner and loser in the reference and target policies. Our resulting algorithm, which we call Constrained Controlled Classification DPO (\texttt{C-3DPO}), has a meaningful RLHF interpretation. By hedging against probability collapse, \texttt{C-3DPO} provides practical improvements over vanilla \texttt{DPO} when aligning several large language models using standard preference datasets.
Authors: Yujie Feng, Xujia Wang, Zexin Lu, Shenghong Fu, Guangyuan Shi, Yongxin Xu, Yasha Wang, Philip S. Yu, Xu Chu, Xiao-Ming Wu
Abstract: Continual learning (CL) is crucial for deploying large language models (LLMs) in dynamic real-world environments without costly retraining. While recent model ensemble and model merging methods guided by parameter importance have gained popularity, they often struggle to balance knowledge transfer and forgetting, mainly due to the reliance on static importance estimates during sequential training. In this paper, we present Recurrent-KIF, a novel CL framework for Recurrent Knowledge Identification and Fusion, which enables dynamic estimation of parameter importance distributions to enhance knowledge transfer. Inspired by human continual learning, Recurrent-KIF employs an inner loop that rapidly adapts to new tasks while identifying important parameters, coupled with an outer loop that globally manages the fusion of new and historical knowledge through redundant knowledge pruning and key knowledge merging. These inner-outer loops iteratively perform multiple rounds of fusion, allowing Recurrent-KIF to leverage intermediate training information and adaptively adjust fusion strategies based on evolving importance distributions. Extensive experiments on two CL benchmarks with various model sizes (from 770M to 13B) demonstrate that Recurrent-KIF effectively mitigates catastrophic forgetting and enhances knowledge transfer.
Authors: Fran\c{c}ois Charton
Abstract: This paper documents Int2Int, an open source code base for using transformers on problems of mathematical research, with a focus on number theory and other problems involving integers. Int2Int is a complete PyTorch implementation of a transformer architecture, together with training and evaluation loops, and classes and functions to represent, generate and decode common mathematical objects. Ancillary code for data preparation, and Jupyter Notebooks for visualizing experimental results are also provided. This document presents the main features of Int2Int, serves as its user manual, and provides guidelines on how to extend it. Int2Int is released under the MIT licence, at https://github.com/FacebookResearch/Int2Int.
Authors: Hantao Lou, Changye Li, Jiaming Ji, Yaodong Yang
Abstract: With the integration of image modality, the semantic space of multimodal large language models (MLLMs) is more complex than text-only models, making their interpretability more challenging and their alignment less stable, particularly susceptible to low-quality data, which can lead to inconsistencies between modalities, hallucinations, and biased outputs. As a result, developing interpretability methods for MLLMs is crucial for improving alignment quality and efficiency. In text-only LLMs, Sparse Autoencoders (SAEs) have gained attention for their ability to interpret latent representations. However, extending SAEs to multimodal settings presents new challenges due to modality fusion and the difficulty of isolating cross-modal representations. To address these challenges, we introduce SAE-V, a mechanistic interpretability framework that extends the SAE paradigm to MLLMs. By identifying and analyzing interpretable features along with their corresponding data, SAE-V enables fine-grained interpretation of both model behavior and data quality, facilitating a deeper understanding of cross-modal interactions and alignment dynamics. Moreover, by utilizing cross-modal feature weighting, SAE-V provides an intrinsic data filtering mechanism to enhance model alignment without requiring additional models. Specifically, when applied to the alignment process of MLLMs, SAE-V-based data filtering methods could achieve more than 110% performance with less than 50% data. Our results highlight SAE-V's ability to enhance interpretability and alignment in MLLMs, providing insights into their internal mechanisms.
Authors: Jiaming Zhang, Mingxi Lei, Meng Ding, Mengdi Li, Zihang Xiang, Difei Xu, Jinhui Xu, Di Wang
Abstract: Reinforcement Learning with Human Feedback (RLHF) has emerged as an influential technique, enabling the alignment of large language models (LLMs) with human preferences. Despite the promising potential of RLHF, how to protect user preference privacy has become a crucial issue. Most previous work has focused on using differential privacy (DP) to protect the privacy of individual data. However, they have concentrated primarily on item-level privacy protection and have unsatisfactory performance for user-level privacy, which is more common in RLHF. This study proposes a novel framework, AUP-RLHF, which integrates user-level label DP into RLHF. We first show that the classical random response algorithm, which achieves an acceptable performance in item-level privacy, leads to suboptimal utility when in the user-level settings. We then establish a lower bound for the user-level label DP-RLHF and develop the AUP-RLHF algorithm, which guarantees $(\varepsilon, \delta)$ user-level privacy and achieves an improved estimation error. Experimental results show that AUP-RLHF outperforms existing baseline methods in sentiment generation and summarization tasks, achieving a better privacy-utility trade-off.
Authors: Zihao Lin, Samyadeep Basu, Mohammad Beigi, Varun Manjunatha, Ryan A. Rossi, Zichao Wang, Yufan Zhou, Sriram Balasubramanian, Arman Zarei, Keivan Rezaei, Ying Shen, Barry Menglong Yao, Zhiyang Xu, Qin Liu, Yuxiang Zhang, Yan Sun, Shilong Liu, Li Shen, Hongxuan Li, Soheil Feizi, Lifu Huang
Abstract: The rise of foundation models has transformed machine learning research, prompting efforts to uncover their inner workings and develop more efficient and reliable applications for better control. While significant progress has been made in interpreting Large Language Models (LLMs), multimodal foundation models (MMFMs) - such as contrastive vision-language models, generative vision-language models, and text-to-image models - pose unique interpretability challenges beyond unimodal frameworks. Despite initial studies, a substantial gap remains between the interpretability of LLMs and MMFMs. This survey explores two key aspects: (1) the adaptation of LLM interpretability methods to multimodal models and (2) understanding the mechanistic differences between unimodal language models and crossmodal systems. By systematically reviewing current MMFM analysis techniques, we propose a structured taxonomy of interpretability methods, compare insights across unimodal and multimodal architectures, and highlight critical research gaps.
Authors: Li Dong, Feibo Jiang, Yubo Peng
Abstract: Unmanned Aerial Vehicles (UAVs) in Wireless Power Transfer (WPT)-assisted Internet of Things (IoT) systems face the following challenges: limited resources and suboptimal trajectory planning. Reinforcement learning-based trajectory planning schemes face issues of low search efficiency and learning instability when optimizing large-scale systems. To address these issues, we present an Attention-based UAV Trajectory Optimization (AUTO) framework based on the graph transformer, which consists of an Attention Trajectory Optimization Model (ATOM) and a Trajectory lEarNing Method based on Actor-critic (TENMA). In ATOM, a graph encoder is used to calculate the self-attention characteristics of all IoTDs, and a trajectory decoder is developed to optimize the number and trajectories of UAVs. TENMA then trains the ATOM using an improved Actor-Critic method, in which the real reward of the system is applied as the baseline to reduce variances in the critic network. This method is suitable for high-quality and large-scale multi-UAV trajectory planning. Finally, we develop numerous experiments, including a hardware experiment in the field case, to verify the feasibility and efficiency of the AUTO framework.
Authors: Zheli Xiong
Abstract: This paper presents a comprehensive study on the use of ensemble Reinforcement Learning (RL) models in financial trading strategies, leveraging classifier models to enhance performance. By combining RL algorithms such as A2C, PPO, and SAC with traditional classifiers like Support Vector Machines (SVM), Decision Trees, and Logistic Regression, we investigate how different classifier groups can be integrated to improve risk-return trade-offs. The study evaluates the effectiveness of various ensemble methods, comparing them with individual RL models across key financial metrics, including Cumulative Returns, Sharpe Ratios (SR), Calmar Ratios, and Maximum Drawdown (MDD). Our results demonstrate that ensemble methods consistently outperform base models in terms of risk-adjusted returns, providing better management of drawdowns and overall stability. However, we identify the sensitivity of ensemble performance to the choice of variance threshold {\tau}, highlighting the importance of dynamic {\tau} adjustment to achieve optimal performance. This study emphasizes the value of combining RL with classifiers for adaptive decision-making, with implications for financial trading, robotics, and other dynamic environments.
Authors: Simin Chen, Yiming Chen, Zexin Li, Yifan Jiang, Zhongwei Wan, Yixin He, Dezhi Ran, Tianle Gu, Haizhou Li, Tao Xie, Baishakhi Ray
Abstract: Data contamination has received increasing attention in the era of large language models (LLMs) due to their reliance on vast Internet-derived training corpora. To mitigate the risk of potential data contamination, LLM benchmarking has undergone a transformation from static to dynamic benchmarking. In this work, we conduct an in-depth analysis of existing static to dynamic benchmarking methods aimed at reducing data contamination risks. We first examine methods that enhance static benchmarks and identify their inherent limitations. We then highlight a critical gap-the lack of standardized criteria for evaluating dynamic benchmarks. Based on this observation, we propose a series of optimal design principles for dynamic benchmarking and analyze the limitations of existing dynamic benchmarks. This survey provides a concise yet comprehensive overview of recent advancements in data contamination research, offering valuable insights and a clear guide for future research efforts. We maintain a GitHub repository to continuously collect both static and dynamic benchmarking methods for LLMs. The repository can be found at this link.
Authors: Nicolas Bessone
Abstract: Graph Neural Networks (GNNs) have emerged as a powerful tool for learning on graph-structured data, finding applications in numerous domains including social network analysis and molecular biology. Within this broad category, Asynchronous Recurrent Graph Neural Networks (ARGNNs) stand out for their ability to capture complex dependencies in dynamic graphs, resembling living organisms' intricate and adaptive nature. However, their complexity often leads to large and computationally expensive models. Therefore, pruning unnecessary edges becomes crucial for enhancing efficiency without significantly compromising performance. This paper presents a dynamic pruning method based on graph spectral theory, leveraging the imaginary component of the eigenvalues of the network graph's Laplacian.
Authors: Tasfiq E. Alam, Md Manjurul Ahsan, Shivakumar Raman
Abstract: Bearings play an integral role in ensuring the reliability and efficiency of rotating machinery - reducing friction and handling critical loads. Bearing failures that constitute up to 90% of mechanical faults highlight the imperative need for reliable condition monitoring and fault detection. This study proposes a multimodal bearing fault classification approach that relies on vibration and motor phase current signals within a one-dimensional convolutional neural network (1D CNN) framework. The method fuses features from multiple signals to enhance the accuracy of fault detection. Under the baseline condition (1,500 rpm, 0.7 Nm load torque, and 1,000 N radial force), the model reaches an accuracy of 96% with addition of L2 regularization. This represents a notable improvement of 2% compared to the non-regularized model. In addition, the model demonstrates robust performance across three distinct operating conditions by employing transfer learning (TL) strategies. Among the tested TL variants, the approach that preserves parameters up to the first max-pool layer and then adjusts subsequent layers achieves the highest performance. While this approach attains excellent accuracy across varied conditions, it requires more computational time due to its greater number of trainable parameters. To address resource constraints, less computationally intensive models offer feasible trade-offs, albeit at a slight accuracy cost. Overall, this multimodal 1D CNN framework with late fusion and TL strategies lays a foundation for more accurate, adaptable, and efficient bearing fault classification in industrial environments with variable operating conditions.
Authors: Cl\'ementine Berger (IP Paris, IDS, S2A), Roland Badeau (IP Paris, IDS, S2A), Slim Essid (IP Paris, IDS, S2A)
Abstract: People often listen to music in noisy environments, seeking to isolate themselves from ambient sounds. Indeed, a music signal can mask some of the noise's frequency components due to the effect of simultaneous masking. In this article, we propose a neural network based on a psychoacoustic masking model, designed to enhance the music's ability to mask ambient noise by reshaping its spectral envelope with predicted filter frequency responses. The model is trained with a perceptual loss function that balances two constraints: effectively masking the noise while preserving the original music mix and the user's chosen listening level. We evaluate our approach on simulated data replicating a user's experience of listening to music with headphones in a noisy environment. The results, based on defined objective metrics, demonstrate that our system improves the state of the art.
Authors: Hongyu Zhou, Zorah L\"ahner
Abstract: With the rising popularity of 3D Gaussian splatting and the expanse of applications from rendering to 3D reconstruction, there comes also a need for geometry processing applications directly on this new representation. While considering the centers of Gaussians as a point cloud or meshing them is an option that allows to apply existing algorithms, this might ignore information present in the data or be unnecessarily expensive. Additionally, Gaussian splatting tends to contain a large number of outliers which do not affect the rendering quality but need to be handled correctly in order not to produce noisy results in geometry processing applications. In this work, we propose a formulation to compute the Laplace-Beltrami operator, a widely used tool in geometry processing, directly on Gaussian splatting using the Mahalanobis distance. While conceptually similar to a point cloud Laplacian, our experiments show superior accuracy on the point clouds encoded in the Gaussian splatting centers and, additionally, the operator can be used to evaluate the quality of the output during optimization.
Authors: Tomer Raz, Michael Shalyt, Elyasheev Leibtag, Rotem Kalisch, Yaron Hadad, Ido Kaminer
Abstract: The constant $\pi$ has fascinated scholars for centuries, inspiring the derivation of countless formulas rooted in profound mathematical insight. This abundance of formulas raises a question: Are they interconnected, and can a unifying structure explain their relationships? We propose a systematic methodology for discovering and proving formula equivalences, leveraging modern large language models, large-scale data processing, and novel mathematical algorithms. Analyzing 457,145 arXiv papers, over a third of the validated formulas for $\pi$ were proven to be derivable from a single mathematical object - including formulas by Euler, Gauss, Lord Brouncker, and newer ones from algorithmic discoveries by the Ramanujan Machine. Our approach extends to other constants, such as $e$, $\zeta(3)$, and Catalan's constant, proving its broad applicability. This work represents a step toward the automatic unification of mathematical knowledge, laying a foundation for AI-driven discoveries of connections across scientific domains.
Authors: Zhenheng Tang, Xiang Liu, Qian Wang, Peijie Dong, Bingsheng He, Xiaowen Chu, Bo Li
Abstract: Motivated by reducing the computational and storage costs of LLMs, model compression and KV cache compression have attracted much attention from researchers. However, current methods predominantly emphasize maintaining the performance of compressed LLMs, as measured by perplexity or simple accuracy on tasks of common sense knowledge QA and basic arithmetic reasoning. In this blog, we present a brief review of recent advancements in LLMs related to retrieval-augmented generation, multi-step reasoning, external tools, and computational expressivity, all of which substantially enhance LLM performance. Then, we propose a lottery LLM hypothesis suggesting that for a given LLM and task, there exists a smaller lottery LLM capable of producing the same performance as the original LLM with the assistance of multi-step reasoning and external tools. Based on the review of current progress in LLMs, we discuss and summarize the essential capabilities that the lottery LLM and KV cache compression must possess, which are currently overlooked in existing methods.
Authors: Lucas Beerens, Alex D. Richardson, Kaicheng Zhang, Dongdong Chen
Abstract: The proliferation of text-to-image diffusion models has raised significant privacy and security concerns, particularly regarding the generation of copyrighted or harmful images. To address these issues, research on machine unlearning has developed various concept erasure methods, which aim to remove the effect of unwanted data through post-hoc training. However, we show these erasure techniques are vulnerable, where images of supposedly erased concepts can still be generated using adversarially crafted prompts. We introduce RECORD, a coordinate-descent-based algorithm that discovers prompts capable of eliciting the generation of erased content. We demonstrate that RECORD significantly beats the attack success rate of current state-of-the-art attack methods. Furthermore, our findings reveal that models subjected to concept erasure are more susceptible to adversarial attacks than previously anticipated, highlighting the urgency for more robust unlearning approaches. We open source all our code at https://github.com/LucasBeerens/RECORD
Authors: Rohit Saxena, Pasquale Minervini, Frank Keller
Abstract: Generating accurate and concise textual summaries from multimodal documents is challenging, especially when dealing with visually complex content like scientific posters. We introduce PosterSum, a novel benchmark to advance the development of vision-language models that can understand and summarize scientific posters into research paper abstracts. Our dataset contains 16,305 conference posters paired with their corresponding abstracts as summaries. Each poster is provided in image format and presents diverse visual understanding challenges, such as complex layouts, dense text regions, tables, and figures. We benchmark state-of-the-art Multimodal Large Language Models (MLLMs) on PosterSum and demonstrate that they struggle to accurately interpret and summarize scientific posters. We propose Segment & Summarize, a hierarchical method that outperforms current MLLMs on automated metrics, achieving a 3.14% gain in ROUGE-L. This will serve as a starting point for future research on poster summarization.
Authors: Fahim Tajwar, Yiding Jiang, Abitha Thankaraj, Sumaita Sadia Rahman, J Zico Kolter, Jeff Schneider, Ruslan Salakhutdinov
Abstract: Efficient exploration is essential for intelligent systems interacting with their environment, but existing language models often fall short in scenarios that require strategic information gathering. In this paper, we present PAPRIKA, a fine-tuning approach that enables language models to develop general decision-making capabilities that are not confined to particular environments. By training on synthetic interaction data from different tasks that require diverse strategies, PAPRIKA teaches models to explore and adapt their behavior on a new task based on environment feedback in-context without more gradient updates. Experimental results show that models fine-tuned with PAPRIKA can effectively transfer their learned decision-making capabilities to entirely unseen tasks without additional training. Unlike traditional training, our approach's primary bottleneck lies in sampling useful interaction data instead of model updates. To improve sample efficiency, we propose a curriculum learning strategy that prioritizes sampling trajectories from tasks with high learning potential. These results suggest a promising path towards AI systems that can autonomously solve novel sequential decision-making problems that require interactions with the external world.
Authors: Christopher Zerafa, Pauline Galea, Cristiana Sebu
Abstract: This review explores the integration of deep learning (DL) with full-waveform inversion (FWI) for enhanced seismic imaging and subsurface characterization. It covers FWI and DL fundamentals, geophysical applications (velocity estimation, deconvolution, tomography), and challenges (model complexity, data quality). The review also outlines future research directions, including hybrid, generative, and physics-informed models for improved accuracy, efficiency, and reliability in subsurface property estimation. The synergy between DL and FWI has the potential to transform geophysics, providing new insights into Earth's subsurface.
Authors: Jakub Binkowski, Denis Janiak, Albert Sawczyn, Bogdan Gabrys, Tomasz Kajdanowicz
Abstract: Large Language Models (LLMs) have demonstrated remarkable performance across various tasks but remain prone to hallucinations. Detecting hallucinations is essential for safety-critical applications, and recent methods leverage attention map properties to this end, though their effectiveness remains limited. In this work, we investigate the spectral features of attention maps by interpreting them as adjacency matrices of graph structures. We propose the $\text{LapEigvals}$ method, which utilises the top-$k$ eigenvalues of the Laplacian matrix derived from the attention maps as an input to hallucination detection probes. Empirical evaluations demonstrate that our approach achieves state-of-the-art hallucination detection performance among attention-based methods. Extensive ablation studies further highlight the robustness and generalisation of $\text{LapEigvals}$, paving the way for future advancements in the hallucination detection domain.
Authors: Tian Yu Liu, Alessandro Achille, Matthew Trager, Aditya Golatkar, Luca Zancato, Stefano Soatto
Abstract: Providing Large Language Models with relevant contextual knowledge at inference time has been shown to greatly improve the quality of their generations. This is often achieved by prepending informative passages of text, or 'contexts', retrieved from external knowledge bases to their input. However, processing additional contexts online incurs significant computation costs that scale with their length. State Space Models (SSMs) offer a promising solution by allowing a database of contexts to be mapped onto fixed-dimensional states from which to start the generation. A key challenge arises when attempting to leverage information present across multiple contexts, since there is no straightforward way to condition generation on multiple independent states in existing SSMs. To address this, we leverage a simple mathematical relation derived from SSM dynamics to compose multiple states into one that efficiently approximates the effect of concatenating textual contexts. Since the temporal ordering of contexts can often be uninformative, we enforce permutation-invariance by efficiently averaging states obtained via our composition algorithm across all possible context orderings. We evaluate our resulting method on WikiText and MSMARCO in both zero-shot and fine-tuned settings, and show that we can match the strongest performing baseline while enjoying on average 5.4x speedup.
Authors: Christopher Zerafa, Pauline Galea, Cristiana Sebu
Abstract: FWI seeks to achieve a high-resolution model of the subsurface through the application of multi-variate optimization to the seismic inverse problem. Although now a mature technology, FWI has limitations related to the choice of the appropriate solver for the forward problem in challenging environments requiring complex assumptions, and very wide angle and multi-azimuth data necessary for full reconstruction are often not available. Deep Learning techniques have emerged as excellent optimization frameworks. These exist between data and theory-guided methods. Data-driven methods do not impose a wave propagation model and are not exposed to modelling errors. On the contrary, deterministic models are governed by the laws of physics. Application of seismic FWI has recently started to be investigated within Deep Learning. This has focussed on the time-domain approach, while the pseudo-spectral domain has not been yet explored. However, classical FWI experienced major breakthroughs when pseudo-spectral approaches were employed. This work addresses the lacuna that exists in incorporating the pseudo-spectral approach within Deep Learning. This has been done by re-formulating the pseudo-spectral FWI problem as a Deep Learning algorithm for a data-driven pseudo-spectral approach. A novel DNN framework is proposed. This is formulated theoretically, qualitatively assessed on synthetic data, applied to a two-dimensional Marmousi dataset and evaluated against deterministic and time-based approaches. Inversion of data-driven pseudo-spectral DNN was found to outperform classical FWI for deeper and over-thrust areas. This is due to the global approximator nature of the technique and hence not bound by forward-modelling physical constraints from ray-tracing.
Authors: Adrian Thummerer, Erik van der Bijl, Arthur Jr Galapon, Florian Kamp, Mark Savenije, Christina Muijs, Shafak Aluwini, Roel J. H. M. Steenbakkers, Stephanie Beuel, Martijn P. W. Intven, Johannes A. Langendijk, Stefan Both, Stefanie Corradini, Viktor Rogowski, Maarten Terpstra, Niklas Wahl, Christopher Kurz, Guillaume Landry, Matteo Maspero
Abstract: Medical imaging is essential in modern radiotherapy, supporting diagnosis, treatment planning, and monitoring. Synthetic imaging, particularly synthetic computed tomography (sCT), is gaining traction in radiotherapy. The SynthRAD2025 dataset and Grand Challenge promote advancements in sCT generation by providing a benchmarking platform for algorithms using cone-beam CT (CBCT) and magnetic resonance imaging (MRI). The dataset includes 2362 cases: 890 MRI-CT and 1472 CBCT-CT pairs from head-and-neck, thoracic, and abdominal cancer patients treated at five European university medical centers (UMC Groningen, UMC Utrecht, Radboud UMC, LMU University Hospital Munich, and University Hospital of Cologne). Data were acquired with diverse scanners and protocols. Pre-processing, including rigid and deformable image registration, ensures high-quality, modality-aligned images. Extensive quality assurance validates image consistency and usability. All imaging data is provided in MetaImage (.mha) format, ensuring compatibility with medical image processing tools. Metadata, including acquisition parameters and registration details, is available in structured CSV files. To maintain dataset integrity, SynthRAD2025 is divided into training (65%), validation (10%), and test (25%) sets. The dataset is accessible at https://doi.org/10.5281/zenodo.14918089 under the SynthRAD2025 collection. This dataset supports benchmarking and the development of synthetic imaging techniques for radiotherapy applications. Use cases include sCT generation for MRI-only and MR-guided photon/proton therapy, CBCT-based dose calculations, and adaptive radiotherapy workflows. By integrating diverse acquisition settings, SynthRAD2025 fosters robust, generalizable image synthesis algorithms, advancing personalized cancer care and adaptive radiotherapy.
Authors: Stig Hellemans, Andres Algaba, Sam Verboven, Vincent Ginis
Abstract: Counterfactual explanations provide actionable insights to achieve desired outcomes by suggesting minimal changes to input features. However, existing methods rely on fixed sets of mutable features, which makes counterfactual explanations inflexible for users with heterogeneous real-world constraints. Here, we introduce Flexible Counterfactual Explanations, a framework incorporating counterfactual templates, which allows users to dynamically specify mutable features at inference time. In our implementation, we use Generative Adversarial Networks (FCEGAN), which align explanations with user-defined constraints without requiring model retraining or additional optimization. Furthermore, FCEGAN is designed for black-box scenarios, leveraging historical prediction datasets to generate explanations without direct access to model internals. Experiments across economic and healthcare datasets demonstrate that FCEGAN significantly improves counterfactual explanations' validity compared to traditional benchmark methods. By integrating user-driven flexibility and black-box compatibility, counterfactual templates support personalized explanations tailored to user constraints.
Authors: Sangwon Seo, Vaibhav Unhelkar
Abstract: Successful collaboration requires team members to stay aligned, especially in complex sequential tasks. Team members must dynamically coordinate which subtasks to perform and in what order. However, real-world constraints like partial observability and limited communication bandwidth often lead to suboptimal collaboration. Even among expert teams, the same task can be executed in multiple ways. To develop multi-agent systems and human-AI teams for such tasks, we are interested in data-driven learning of multimodal team behaviors. Multi-Agent Imitation Learning (MAIL) provides a promising framework for data-driven learning of team behavior from demonstrations, but existing methods struggle with heterogeneous demonstrations, as they assume that all demonstrations originate from a single team policy. Hence, in this work, we introduce DTIL: a hierarchical MAIL algorithm designed to learn multimodal team behaviors in complex sequential tasks. DTIL represents each team member with a hierarchical policy and learns these policies from heterogeneous team demonstrations in a factored manner. By employing a distribution-matching approach, DTIL mitigates compounding errors and scales effectively to long horizons and continuous state representations. Experimental results show that DTIL outperforms MAIL baselines and accurately models team behavior across a variety of collaborative scenarios.
Authors: Christopher Zerafa, Pauline Galea, Cristiana Sebu
Abstract: Full-Waveform Inversion seeks to achieve a high-resolution model of the subsurface through the application of multi-variate optimization to the seismic inverse problem. Although now a mature technology, FWI has limitations related to the choice of the appropriate solver for the forward problem in challenging environments requiring complex assumptions, and very wide angle and multi-azimuth data necessary for full reconstruction are often not available. Deep Learning techniques have emerged as excellent optimization frameworks. Data-driven methods do not impose a wave propagation model and are not exposed to modelling errors. On the contrary, deterministic models are governed by the laws of physics. Seismic FWI has recently started to be investigated as a Deep Learning framework. Focus has been on the time-domain, while the pseudo-spectral domain has not been yet explored. However, classical FWI experienced major breakthroughs when pseudo-spectral approaches were employed. This work addresses the lacuna that exists in incorporating the pseudo-spectral approach within Deep Learning. This has been done by re-formulating the pseudo-spectral FWI problem as a Deep Learning algorithm for a theory-driven pseudo-spectral approach. A novel Recurrent Neural Network framework is proposed. This is qualitatively assessed on synthetic data, applied to a two-dimensional Marmousi dataset and evaluated against deterministic and time-based approaches. Pseudo-spectral theory-guided FWI using RNN was shown to be more accurate than classical FWI with only 0.05 error tolerance and 1.45\% relative percent-age error. Indeed, this provides more stable convergence, able to identify faults better and has more low frequency content than classical FWI. Moreover, RNN was more suited than classical FWI at edge detection in the shallow and deep sections due to cleaner receiver residuals.
Authors: Manuj Kant, Sareh Nabi, Manav Kant, Roland Scharrer, Megan Ma, Marzieh Nabi
Abstract: Legal services rely heavily on text processing. While large language models (LLMs) show promise, their application in legal contexts demands higher accuracy, repeatability, and transparency. Logic programs, by encoding legal concepts as structured rules and facts, offer reliable automation, but require sophisticated text extraction. We propose a neuro-symbolic approach that integrates LLMs' natural language understanding with logic-based reasoning to address these limitations. As a legal document case study, we applied neuro-symbolic AI to coverage-related queries in insurance contracts using both closed and open-source LLMs. While LLMs have improved in legal reasoning, they still lack the accuracy and consistency required for complex contract analysis. In our analysis, we tested three methodologies to evaluate whether a specific claim is covered under a contract: a vanilla LLM, an unguided approach that leverages LLMs to encode both the contract and the claim, and a guided approach that uses a framework for the LLM to encode the contract. We demonstrated the promising capabilities of LLM + Logic in the guided approach.
Authors: Horst K. Hahn, Matthias S. May, Volker Dicken, Michael Walz, Rainer E{\ss}eling, Bianca Lassen-Schmidt, Robert Rischen, Jens Vogel-Claussen, Konstantin Nikolaou, J\"org Barkhausen
Abstract: Lung cancer is the second most common cancer and the leading cause of cancer-related deaths worldwide. Survival largely depends on tumor stage at diagnosis, and early detection with low-dose CT can significantly reduce mortality in high-risk patients. AI can improve the detection, measurement, and characterization of pulmonary nodules while reducing assessment time. However, the training data, functionality, and performance of available AI systems vary considerably, complicating software selection and regulatory evaluation. Manufacturers must specify intended use and provide test statistics, but they can choose their training and test data, limiting standardization and comparability. Under the EU AI Act, consistent quality assurance is required for AI-based nodule detection, measurement, and characterization. This position paper proposes systematic quality assurance grounded in a validated reference dataset, including real screening cases plus phantom data to verify volume and growth rate measurements. Regular updates shall reflect demographic shifts and technological advances, ensuring ongoing relevance. Consequently, ongoing AI quality assurance is vital. Regulatory challenges are also adressed. While the MDR and the EU AI Act set baseline requirements, they do not adequately address self-learning algorithms or their updates. A standardized, transparent quality assessment - based on sensitivity, specificity, and volumetric accuracy - enables an objective evaluation of each AI solution's strengths and weaknesses. Establishing clear testing criteria and systematically using updated reference data lay the groundwork for comparable performance metrics, informing tenders, guidelines, and recommendations.
Authors: Sameer Neupane (University of Memphis), Poorvesh Dongre (Virginia Tech), Denis Gracanin (Virginia Tech), Santosh Kumar (University of Memphis)
Abstract: We use a duoethnographic approach to study how wearable-integrated LLM chatbots can assist with personalized stress management, addressing the growing need for immediacy and tailored interventions. Two researchers interacted with custom chatbots over 22 days, responding to wearable-detected physiological prompts, recording stressor phrases, and using them to seek tailored interventions from their LLM-powered chatbots. They recorded their experiences in autoethnographic diaries and analyzed them during weekly discussions, focusing on the relevance, clarity, and impact of chatbot-generated interventions. Results showed that even though most events triggered by the wearable were meaningful, only one in five warranted an intervention. It also showed that interventions tailored with brief event descriptions were more effective than generic ones. By examining the intersection of wearables and LLM, this research contributes to developing more effective, user-centric mental health tools for real-time stress relief and behavior change.
Authors: Bingxuan Li, Yiwei Wang, Jiuxiang Gu, Kai-Wei Chang, Nanyun Peng
Abstract: Chart generation aims to generate code to produce charts satisfying the desired visual properties, e.g., texts, layout, color, and type. It has great potential to empower the automatic professional report generation in financial analysis, research presentation, education, and healthcare. In this work, we build a vision-language model (VLM) based multi-agent framework for effective automatic chart generation. Generating high-quality charts requires both strong visual design skills and precise coding capabilities that embed the desired visual properties into code. Such a complex multi-modal reasoning process is difficult for direct prompting of VLMs. To resolve these challenges, we propose METAL, a multi-agent framework that decomposes the task of chart generation into the iterative collaboration among specialized agents. METAL achieves 5.2% improvement in accuracy over the current best result in the chart generation task. The METAL framework exhibits the phenomenon of test-time scaling: its performance increases monotonically as the logarithmic computational budget grows from 512 to 8192 tokens. In addition, we find that separating different modalities during the critique process of METAL boosts the self-correction capability of VLMs in the multimodal context.
Authors: Xinyi Song, Lina Lee, Kexin Xie, Xueying Liu, Xinwei Deng, Yili Hong
Abstract: The coding capabilities of large language models (LLMs) have opened up new opportunities for automatic statistical analysis in machine learning and data science. However, before their widespread adoption, it is crucial to assess the accuracy of code generated by LLMs. A major challenge in this evaluation lies in the absence of a benchmark dataset for statistical code (e.g., SAS and R). To fill in this gap, this paper introduces StatLLM, an open-source dataset for evaluating the performance of LLMs in statistical analysis. The StatLLM dataset comprises three key components: statistical analysis tasks, LLM-generated SAS code, and human evaluation scores. The first component includes statistical analysis tasks spanning a variety of analyses and datasets, providing problem descriptions, dataset details, and human-verified SAS code. The second component features SAS code generated by ChatGPT 3.5, ChatGPT 4.0, and Llama 3.1 for those tasks. The third component contains evaluation scores from human experts in assessing the correctness, effectiveness, readability, executability, and output accuracy of the LLM-generated code. We also illustrate the unique potential of the established benchmark dataset for (1) evaluating and enhancing natural language processing metrics, (2) assessing and improving LLM performance in statistical coding, and (3) developing and testing of next-generation statistical software - advancements that are crucial for data science and machine learning research.
Authors: Xi Liu, Yujun Zhao, Chun Yu Wan, Yang Zhang, Junwei Liu
Abstract: In recent years, with the rapid development of machine learning, physicists have been exploring its new applications in solving or alleviating the curse of dimensionality in many-body problems. In order to accurately reflect the underlying physics of the problem, domain knowledge must be encoded into the machine learning algorithms. In this work, inspired by field theory, we propose a new set of machine learning models called effective field neural networks (EFNNs) that can automatically and efficiently capture important many-body interactions through multiple self-refining processes. Taking the classical $3$-spin infinite-range model and the quantum double exchange model as case studies, we explicitly demonstrate that EFNNs significantly outperform fully-connected deep neural networks (DNNs) and the effective model. Furthermore, with the help of convolution operations, the EFNNs learned in a small system can be seamlessly used in a larger system without additional training and the relative errors even decrease, which further demonstrates the efficacy of EFNNs in representing core physical behaviors.
Authors: Denis Tarasov, Alexander Nikulin, Ilya Zisman, Albina Klepach, Andrei Polubarov, Nikita Lyubaykin, Alexander Derevyagin, Igor Kiselev, Vladislav Kurenkov
Abstract: In this work, we explore the integration of Reinforcement Learning (RL) approaches within a scalable offline In-Context RL (ICRL) framework. Through experiments across more than 150 datasets derived from GridWorld and MuJoCo environments, we demonstrate that optimizing RL objectives improves performance by approximately 40% on average compared to the widely established Algorithm Distillation (AD) baseline across various dataset coverages, structures, expertise levels, and environmental complexities. Our results also reveal that offline RL-based methods outperform online approaches, which are not specifically designed for offline scenarios. These findings underscore the importance of aligning the learning objectives with RL's reward-maximization goal and demonstrate that offline RL is a promising direction for application in ICRL settings.
Authors: Jason R. C. Nurse
Abstract: As technology has become more embedded into our society, the security of modern-day systems is paramount. One topic which is constantly under discussion is that of patching, or more specifically, the installation of updates that remediate security vulnerabilities in software or hardware systems. This continued deliberation is motivated by complexities involved with patching; in particular, the various incentives and disincentives for organizations and their cybersecurity teams when deciding whether to patch. In this paper, we take a fresh look at the question of patching and critically explore why organizations and IT/security teams choose to patch or decide against it (either explicitly or due to inaction). We tackle this question by aggregating and synthesizing prominent research and industry literature on the incentives and disincentives for patching, specifically considering the human aspects in the context of these motives. Through this research, this study identifies key motivators such as organizational needs, the IT/security team's relationship with vendors, and legal and regulatory requirements placed on the business and its staff. There are also numerous significant reasons discovered for why the decision is taken not to patch, including limited resources (e.g., person-power), challenges with manual patch management tasks, human error, bad patches, unreliable patch management tools, and the perception that related vulnerabilities would not be exploited. These disincentives, in combination with the motivators above, highlight the difficult balance that organizations and their security teams need to maintain on a daily basis. Finally, we conclude by discussing implications of these findings and important future considerations.
Authors: Yu Zhou, Bingxuan Li, Mohan Tang, Xiaomeng Jin, Te-Lin Wu, Kuan-Hao Huang, Heng Ji, Kai-Wei Chang, Nanyun Peng
Abstract: Large multimodal models (LMMs) often struggle to recognize novel concepts, as they rely on pre-trained knowledge and have limited ability to capture subtle visual details. Domain-specific knowledge gaps in training also make them prone to confusing visually similar, commonly misrepresented, or low-resource concepts. To help LMMs better align nuanced visual features with language, improving their ability to recognize and reason about novel or rare concepts, we propose a Contrastive visual Data Augmentation (CoDA) strategy. CoDA extracts key contrastive textual and visual features of target concepts against the known concepts they are misrecognized as, and then uses multimodal generative models to produce targeted synthetic data. Automatic filtering of extracted features and augmented images is implemented to guarantee their quality, as verified by human annotators. We show the effectiveness and efficiency of CoDA on low-resource concept and diverse scene recognition datasets including INaturalist and SUN. We additionally collect NovelSpecies, a benchmark dataset consisting of newly discovered animal species that are guaranteed to be unseen by LMMs. LLaVA-1.6 1-shot updating results on these three datasets show CoDA significantly improves SOTA visual data augmentation strategies by 12.3% (NovelSpecies), 5.1% (SUN), and 6.0% (iNat) absolute gains in accuracy.
Authors: Anna Ravera, Cristina Gena
Abstract: Generative AI systems are transforming content creation, but their usability remains a key challenge. This paper examines usability factors such as user experience, transparency, control, and cognitive load. Common challenges include unpredictability and difficulties in fine-tuning outputs. We review evaluation metrics like efficiency, learnability, and satisfaction, highlighting best practices from various domains. Improving interpretability, intuitive interfaces, and user feedback can enhance usability, making generative AI more accessible and effective.
Authors: Zhe Liu, Taekyu Kang, Haoyu Wang, Seyed Hossein Alavi, Vered Shwartz
Abstract: Effective conversational systems are expected to dynamically generate contextual follow-up questions to elicit new information while maintaining the conversation flow. While humans excel at asking diverse and informative questions by intuitively assessing both obtained and missing information, existing models often fall short of human performance on this task. To mitigate this, we propose a method that generates diverse and informative questions based on targeting unanswered information using a hypothetical LLM-generated "comprehensive answer". Our method is applied to augment an existing follow-up questions dataset. The experimental results demonstrate that language models fine-tuned on the augmented datasets produce follow-up questions of significantly higher quality and diversity. This promising approach could be effectively adopted to future work to augment information-seeking dialogues for reducing ambiguities and improving the accuracy of LLM answers.
Authors: Yuxuan Li, Hirokazu Shirado
Abstract: Large language models, when trained with reinforcement learning, demonstrate advanced problem-solving capabilities through reasoning techniques like chain of thoughts and reflection. However, it is unclear how these reasoning capabilities extend to social intelligence. In this study, we investigate how reasoning influences model outcomes in social dilemmas. First, we examine the effects of chain-of-thought and reflection techniques in a public goods game. We then extend our analysis to six economic games on cooperation and punishment, comparing off-the-shelf non-reasoning and reasoning models. We find that reasoning models reduce cooperation and norm enforcement, prioritizing individual rationality. Consequently, groups with more reasoning models exhibit less cooperation and lower gains through repeated interactions. These behaviors parallel human tendencies of "spontaneous giving and calculated greed." Our results suggest the need for AI architectures that incorporate social intelligence alongside reasoning capabilities to ensure that AI supports, rather than disrupts, human cooperative intuition.
Authors: Xiangwen Wang, Yibo Jacky Zhang, Zhoujie Ding, Katherine Tsai, Sanmi Koyejo
Abstract: Compound AI systems, comprising multiple interacting components such as LLM agents and external tools, demonstrate state-of-the-art results across diverse tasks. It is hence crucial to align components within the system to produce consistent results that match human expectations. However, conventional alignment methods, such as Direct Preference Optimization (DPO), are not directly applicable to compound AI systems. These challenges include the non-differentiable interactions between components, making end-to-end gradient optimization infeasible. Additionally, system-level preferences cannot be directly translated into component-level preferences, further complicating alignment. We address the issues by formulating compound AI systems as Directed Acyclic Graphs (DAGs), capturing the connections between agents and the data generation processes. We propose a system-level DPO (SysDPO) to jointly align compound systems by adapting the DPO to operate on these DAGs. We study the joint alignment of an LLM and a diffusion model to demonstrate the effectiveness of our approach. Our exploration provides insights into the alignment of compound AI systems and lays a foundation for future advancements.
Authors: Venkat Padmasola, Zhaotong Li, Rupak Chatterjee, Wesley Dyk
Abstract: We study the application of emerging photonic and quantum computing architectures to solving the Traveling Salesman Problem (TSP), a well-known NP-hard optimization problem. We investigate several approaches: Simulated Annealing (SA), Quadratic Unconstrained Binary Optimization (QUBO-Ising) methods implemented on quantum annealers and Optical Coherent Ising Machines, as well as the Quantum Approximate Optimization Algorithm (QAOA) and the Quantum Phase Estimation (QPE) algorithm on gate-based quantum computers. QAOA and QPE were tested on the IBM Quantum platform. The QUBO-Ising method was explored using the D-Wave quantum annealer, which operates on superconducting Josephson junctions, and the QCI Dirac machine, a nonlinear optoelectronic Ising machine. Gate-based quantum computers demonstrated accurate results for small TSP instances in simulation. However, real quantum devices are hindered by noise and limited scalability. Circuit complexity grows with problem size, restricting performance to TSP instances with a maximum of 6 nodes. In contrast, Ising-based architectures show improved scalability for larger problem sizes. SQUID-based Ising machines can handle TSP instances with up to 12 nodes, while nonlinear optoelectronic Ising machines extend this capability to 18 nodes. Nevertheless, the solutions tend to be suboptimal due to hardware limitations and challenges in achieving ground state convergence as the problem size increases. Despite these limitations, Ising machines demonstrate significant time advantages over classical methods, making them a promising candidate for solving larger-scale TSPs efficiently.
Authors: Keon Ju Maverick Lee, Jeff Ens, Sara Adkins, Pedro Sarmento, Mathieu Barthet, Philippe Pasquier
Abstract: The Musical Instrument Digital Interface (MIDI), introduced in 1983, revolutionized music production by allowing computers and instruments to communicate efficiently. MIDI files encode musical instructions compactly, facilitating convenient music sharing. They benefit Music Information Retrieval (MIR), aiding in research on music understanding, computational musicology, and generative music. The GigaMIDI dataset contains over 1.4 million unique MIDI files, encompassing 1.8 billion MIDI note events and over 5.3 million MIDI tracks. GigaMIDI is currently the largest collection of symbolic music in MIDI format available for research purposes under fair dealing. Distinguishing between non-expressive and expressive MIDI tracks is challenging, as MIDI files do not inherently make this distinction. To address this issue, we introduce a set of innovative heuristics for detecting expressive music performance. These include the Distinctive Note Velocity Ratio (DNVR) heuristic, which analyzes MIDI note velocity; the Distinctive Note Onset Deviation Ratio (DNODR) heuristic, which examines deviations in note onset times; and the Note Onset Median Metric Level (NOMML) heuristic, which evaluates onset positions relative to metric levels. Our evaluation demonstrates these heuristics effectively differentiate between non-expressive and expressive MIDI tracks. Furthermore, after evaluation, we create the most substantial expressive MIDI dataset, employing our heuristic, NOMML. This curated iteration of GigaMIDI encompasses expressively-performed instrument tracks detected by NOMML, containing all General MIDI instruments, constituting 31% of the GigaMIDI dataset, totalling 1,655,649 tracks.
Authors: Mahsa Salmani, Ilya Soloveychik
Abstract: The rapid development of the Transformer-based Large Language Models (LLMs) in recent years has been closely linked to their ever-growing and already enormous sizes. Many LLMs contain hundreds of billions of parameters and require dedicated hardware resources for training and inference. One of the key challenges inherent to the Transformer architecture is the requirement to support numerous non-linear transformations that involves normalization. For instance, each decoder block typically contains at least one Softmax operation and two Layernorms. The computation of the corresponding normalization scaling factors becomes a major bottleneck as it requires spatial collective operations. In other words, when it comes to the computation of denominators for Softmax and Layernorm, all vector elements must be aggregated into a single location, requiring significant communication. These collective operations slow down inference on Transformers by approximately 20%, defeating the whole purpose of distributed in-memory compute. In this work, we propose an extremely efficient technique that can completely hide the overhead caused by such collective operations. Note that each Softmax and Layernorm operation is typically followed by a linear layer. Since non-linear and linear operations are performed on different hardware engines, they can be easily parallelized once the algebra allows such commutation. By leveraging the inherent properties of linear operations, we can defer the normalization of the preceding Softmax and Layernorm until after the linear layer is computed. Now we can compute the collective scaling factors concurrently with the matrix multiplication and completely hide the latency of the former behind the latter. Such parallelization preserves the numerical accuracy while significantly improving the hardware utilization and reducing the overall latency.
Authors: Tony Shaska
Abstract: This paper presents a novel framework for graded neural networks (GNNs) built over graded vector spaces $\V_\w^n$, extending classical neural architectures by incorporating algebraic grading. Leveraging a coordinate-wise grading structure with scalar action $\lambda \star \x = (\lambda^{q_i} x_i)$, defined by a tuple $\w = (q_0, \ldots, q_{n-1})$, we introduce graded neurons, layers, activation functions, and loss functions that adapt to feature significance. Theoretical properties of graded spaces are established, followed by a comprehensive GNN design, addressing computational challenges like numerical stability and gradient scaling. Potential applications span machine learning and photonic systems, exemplified by high-speed laser-based implementations. This work offers a foundational step toward graded computation, unifying mathematical rigor with practical potential, with avenues for future empirical and hardware exploration.
Authors: Yuqing Wang, Xiao Yang
Abstract: Traditional security protection methods struggle to address sophisticated attack vectors in large-scale distributed systems, particularly when balancing detection accuracy with data privacy concerns. This paper presents a novel distributed security threat detection system that integrates federated learning with multimodal large language models (LLMs). Our system leverages federated learning to ensure data privacy while employing multimodal LLMs to process heterogeneous data sources including network traffic, system logs, images, and sensor data. Experimental evaluation on a 10TB distributed dataset demonstrates that our approach achieves 96.4% detection accuracy, outperforming traditional baseline models by 4.1 percentage points. The system reduces both false positive and false negative rates by 1.8 and 2.4 percentage points respectively. Performance analysis shows that our system maintains efficient processing capabilities in distributed environments, requiring 180 seconds for model training and 3.8 seconds for threat detection across the distributed network. These results demonstrate significant improvements in detection accuracy and computational efficiency while preserving data privacy, suggesting strong potential for real-world deployment in large-scale security systems.
Authors: Qile Jiang, Zhiwei Gao, George Em Karniadakis
Abstract: Large Language Models (LLMs) have emerged as powerful tools for tackling a wide range of problems, including those in scientific computing, particularly in solving partial differential equations (PDEs). However, different models exhibit distinct strengths and preferences, resulting in varying levels of performance. In this paper, we compare the capabilities of the most advanced LLMs--ChatGPT and DeepSeek--along with their reasoning-optimized versions in addressing computational challenges. Specifically, we evaluate their proficiency in solving traditional numerical problems in scientific computing as well as leveraging scientific machine learning techniques for PDE-based problems. We designed all our experiments so that a non-trivial decision is required, e.g. defining the proper space of input functions for neural operator learning. Our findings reveal that the latest model, ChatGPT o3-mini-high, usually delivers the most accurate results while also responding significantly faster than its reasoning counterpart, DeepSeek R1. This enhanced speed and accuracy make ChatGPT o3-mini-high a more practical and efficient choice for diverse computational tasks at this juncture.
Authors: Chris Dongjoo Kim, Sangwoo Moon, Jihwan Moon, Dongyeon Woo, Gunhee Kim
Abstract: As with many other problems, real-world regression is plagued by the presence of noisy labels, an inevitable issue that demands our attention. Fortunately, much real-world data often exhibits an intrinsic property of continuously ordered correlations between labels and features, where data points with similar labels are also represented with closely related features. In response, we propose a novel approach named ConFrag, where we collectively model the regression data by transforming them into disjoint yet contrasting fragmentation pairs. This enables the training of more distinctive representations, enhancing the ability to select clean samples. Our ConFrag framework leverages a mixture of neighboring fragments to discern noisy labels through neighborhood agreement among expert feature extractors. We extensively perform experiments on six newly curated benchmark datasets of diverse domains, including age prediction, price prediction, and music production year estimation. We also introduce a metric called Error Residual Ratio (ERR) to better account for varying degrees of label noise. Our approach consistently outperforms fourteen state-of-the-art baselines, being robust against symmetric and random Gaussian label noise.
Authors: Chengpiao Huang, Yuhang Wu, Kaizheng Wang
Abstract: We investigate the reliable use of simulated survey responses from large language models (LLMs) through the lens of uncertainty quantification. Our approach converts synthetic data into confidence sets for population parameters of human responses, addressing the distribution shift between the simulated and real populations. A key innovation lies in determining the optimal number of simulated responses: too many produce overly narrow confidence sets with poor coverage, while too few yield excessively loose estimates. To resolve this, our method adaptively selects the simulation sample size, ensuring valid average-case coverage guarantees. It is broadly applicable to any LLM, irrespective of its fidelity, and any procedure for constructing confidence sets. Additionally, the selected sample size quantifies the degree of misalignment between the LLM and the target human population. We illustrate our method on real datasets and LLMs.
Authors: Wei Liu, Yancheng He, Hui Huang, Chengwei Hu, Jiaheng Liu, Shilong Li, Wenbo Su, Bo Zheng
Abstract: With the development of large language models, their ability to follow simple instructions has significantly improved. However, adhering to complex instructions remains a major challenge. Current approaches to generating complex instructions are often irrelevant to the current instruction requirements or suffer from limited scalability and diversity. Moreover, methods such as back-translation, while effective for simple instruction generation, fail to leverage the rich contents and structures in large web corpora. In this paper, we propose a novel automatic iterative refinement framework to generate complex instructions with constraints, which not only better reflects the requirements of real scenarios but also significantly enhances LLMs' ability to follow complex instructions. The AIR framework consists of two stages: (1)Generate an initial instruction from a document; (2)Iteratively refine instructions with LLM-as-judge guidance by comparing the model's output with the document to incorporate valuable constraints. Finally, we construct the AIR-10K dataset with 10K complex instructions and demonstrate that instructions generated with our approach significantly improve the model's ability to follow complex instructions, outperforming existing methods for instruction generation.
Authors: Xiaomeng Jin, Hyeonjeong Ha, Jeonghwan Kim, Jiateng Liu, Zhenhailong Wang, Khanh Duy Nguyen, Ansel Blume, Nanyun Peng, Kai-wei Chang, Heng Ji
Abstract: Text-to-image (T2I) models enable rapid concept design, making them widely used in AI-driven design. While recent studies focus on generating semantic and stylistic variations of given design concepts, functional coherence--the integration of multiple affordances into a single coherent concept--remains largely overlooked. In this paper, we introduce SYNTHIA, a framework for generating novel, functionally coherent designs based on desired affordances. Our approach leverages a hierarchical concept ontology that decomposes concepts into parts and affordances, serving as a crucial building block for functionally coherent design. We also develop a curriculum learning scheme based on our ontology that contrastively fine-tunes T2I models to progressively learn affordance composition while maintaining visual novelty. To elaborate, we (i) gradually increase affordance distance, guiding models from basic concept-affordance association to complex affordance compositions that integrate parts of distinct affordances into a single, coherent form, and (ii) enforce visual novelty by employing contrastive objectives to push learned representations away from existing concepts. Experimental results show that SYNTHIA outperforms state-of-the-art T2I models, demonstrating absolute gains of 25.1% and 14.7% for novelty and functional coherence in human evaluation, respectively.
Authors: Yuqing Wang, Xiao Yang
Abstract: Cloud computing environments are increasingly vulnerable to security threats such as distributed denial-of-service (DDoS) attacks and SQL injection. Traditional security mechanisms, based on rule matching and feature recognition, struggle to adapt to evolving attack strategies. This paper proposes an adaptive security protection framework leveraging deep learning to construct a multi-layered defense architecture. The proposed system is evaluated in a real-world business environment, achieving a detection accuracy of 97.3%, an average response time of 18 ms, and an availability rate of 99.999%. Experimental results demonstrate that the proposed method significantly enhances detection accuracy, response efficiency, and resource utilization, offering a novel and effective approach to cloud computing security.
Authors: Wenlong Ji, Weizhe Yuan, Emily Getzen, Kyunghyun Cho, Michael I. Jordan, Song Mei, Jason E Weston, Weijie J. Su, Jing Xu, Linjun Zhang
Abstract: Large Language Models (LLMs) have emerged as transformative tools in artificial intelligence (AI), exhibiting remarkable capabilities across diverse tasks such as text generation, reasoning, and decision-making. While their success has primarily been driven by advances in computational power and deep learning architectures, emerging problems -- in areas such as uncertainty quantification, decision-making, causal inference, and distribution shift -- require a deeper engagement with the field of statistics. This paper explores potential areas where statisticians can make important contributions to the development of LLMs, particularly those that aim to engender trustworthiness and transparency for human users. Thus, we focus on issues such as uncertainty quantification, interpretability, fairness, privacy, watermarking and model adaptation. We also consider possible roles for LLMs in statistical analysis. By bridging AI and statistics, we aim to foster a deeper collaboration that advances both the theoretical foundations and practical applications of LLMs, ultimately shaping their role in addressing complex societal challenges.
Authors: Rui Liu, Yu Shen, Peng Gao, Pratap Tokekar, Ming Lin
Abstract: Multi-modality learning has become a crucial technique for improving the performance of machine learning applications across domains such as autonomous driving, robotics, and perception systems. While existing frameworks such as Auxiliary Modality Learning (AML) effectively utilize multiple data sources during training and enable inference with reduced modalities, they primarily operate in a single-agent context. This limitation is particularly critical in dynamic environments, such as connected autonomous vehicles (CAV), where incomplete data coverage can lead to decision-making blind spots. To address these challenges, we propose Collaborative Auxiliary Modality Learning ($\textbf{CAML}$), a novel multi-agent multi-modality framework that enables agents to collaborate and share multimodal data during training while allowing inference with reduced modalities per agent during testing. We systematically analyze the effectiveness of $\textbf{CAML}$ from the perspective of uncertainty reduction and data coverage, providing theoretical insights into its advantages over AML. Experimental results in collaborative decision-making for CAV in accident-prone scenarios demonstrate that \ours~achieves up to a ${\bf 58.13}\%$ improvement in accident detection. Additionally, we validate $\textbf{CAML}$ on real-world aerial-ground robot data for collaborative semantic segmentation, achieving up to a ${\bf 10.61}\%$ improvement in mIoU.
Authors: Hyeonjeong Ha, Qiusi Zhan, Jeonghwan Kim, Dimitrios Bralios, Saikrishna Sanniboina, Nanyun Peng, Kai-wei Chang, Daniel Kang, Heng Ji
Abstract: Multimodal large language models (MLLMs) equipped with Retrieval Augmented Generation (RAG) leverage both their rich parametric knowledge and the dynamic, external knowledge to excel in tasks such as Question Answering. While RAG enhances MLLMs by grounding responses in query-relevant external knowledge, this reliance poses a critical yet underexplored safety risk: knowledge poisoning attacks, where misinformation or irrelevant knowledge is intentionally injected into external knowledge bases to manipulate model outputs to be incorrect and even harmful. To expose such vulnerabilities in multimodal RAG, we propose MM-PoisonRAG, a novel knowledge poisoning attack framework with two attack strategies: Localized Poisoning Attack (LPA), which injects query-specific misinformation in both text and images for targeted manipulation, and Globalized Poisoning Attack (GPA) to provide false guidance during MLLM generation to elicit nonsensical responses across all queries. We evaluate our attacks across multiple tasks, models, and access settings, demonstrating that LPA successfully manipulates the MLLM to generate attacker-controlled answers, with a success rate of up to 56% on MultiModalQA. Moreover, GPA completely disrupts model generation to 0% accuracy with just a single irrelevant knowledge injection. Our results highlight the urgent need for robust defenses against knowledge poisoning to safeguard multimodal RAG frameworks.
Authors: Haris Riaz, Ellen Riloff, Mihai Surdeanu
Abstract: We propose a simple, unsupervised method that injects pragmatic principles in retrieval-augmented generation (RAG) frameworks such as Dense Passage Retrieval~\cite{karpukhin2020densepassageretrievalopendomain} to enhance the utility of retrieved contexts. Our approach first identifies which sentences in a pool of documents retrieved by RAG are most relevant to the question at hand, cover all the topics addressed in the input question and no more, and then highlights these sentences within their context, before they are provided to the LLM, without truncating or altering the context in any other way. We show that this simple idea brings consistent improvements in experiments on three question answering tasks (ARC-Challenge, PubHealth and PopQA) using five different LLMs. It notably enhances relative accuracy by up to 19.7\% on PubHealth and 10\% on ARC-Challenge compared to a conventional RAG system.
Authors: Zhiyu Yin, Kehai Chen, Xuefeng Bai, Ruili Jiang, Juntao Li, Hongdong Li, Jin Liu, Yang Xiang, Jun Yu, Min Zhang
Abstract: Video generation, by leveraging a dynamic visual generation method, pushes the boundaries of Artificial Intelligence Generated Content (AIGC). Video generation presents unique challenges beyond static image generation, requiring both high-quality individual frames and temporal coherence to maintain consistency across the spatiotemporal sequence. Recent works have aimed at addressing the spatiotemporal consistency issue in video generation, while few literature review has been organized from this perspective. This gap hinders a deeper understanding of the underlying mechanisms for high-quality video generation. In this survey, we systematically review the recent advances in video generation, covering five key aspects: foundation models, information representations, generation schemes, post-processing techniques, and evaluation metrics. We particularly focus on their contributions to maintaining spatiotemporal consistency. Finally, we discuss the future directions and challenges in this field, hoping to inspire further efforts to advance the development of video generation.
Authors: Ziruo Zhao
Abstract: Contrastive learning has emerged as a powerful paradigm for self-supervised representation learning. This work analyzes the theoretical limits of contrastive learning under nasty noise, where an adversary modifies or replaces training samples. Using PAC learning and VC-dimension analysis, lower and upper bounds on sample complexity in adversarial settings are established. Additionally, data-dependent sample complexity bounds based on the l2-distance function are derived.
Authors: Matteo Contini, Victor Illien, Julien Barde, Sylvain Poulain, Serge Bernard, Alexis Joly, Sylvain Bonhommeau
Abstract: Drone-based remote sensing combined with AI-driven methodologies has shown great potential for accurate mapping and monitoring of coral reef ecosystems. This study presents a novel multi-scale approach to coral reef monitoring, integrating fine-scale underwater imagery with medium-scale aerial imagery. Underwater images are captured using an Autonomous Surface Vehicle (ASV), while aerial images are acquired with an aerial drone. A transformer-based deep-learning model is trained on underwater images to detect the presence of 31 classes covering various coral morphotypes, associated fauna, and habitats. These predictions serve as annotations for training a second model applied to aerial images. The transfer of information across scales is achieved through a weighted footprint method that accounts for partial overlaps between underwater image footprints and aerial image tiles. The results show that the multi-scale methodology successfully extends fine-scale classification to larger reef areas, achieving a high degree of accuracy in predicting coral morphotypes and associated habitats. The method showed a strong alignment between underwater-derived annotations and ground truth data, reflected by an AUC (Area Under the Curve) score of 0.9251. This shows that the integration of underwater and aerial imagery, supported by deep-learning models, can facilitate scalable and accurate reef assessments. This study demonstrates the potential of combining multi-scale imaging and AI to facilitate the monitoring and conservation of coral reefs. Our approach leverages the strengths of underwater and aerial imagery, ensuring the precision of fine-scale analysis while extending it to cover a broader reef area.
Authors: Qinghua He, Hongyang Jiang, Danqi Fang, Dawei Yang, Truong X. Nguyen, Anran Ran, Clement C. Tham, Simon K. H. Szeto, Sobha Sivaprasad, Carol Y. Cheung
Abstract: Diabetic macular ischemia (DMI), marked by the loss of retinal capillaries in the macular area, contributes to vision impairment in patients with diabetes. Although color fundus photographs (CFPs), combined with artificial intelligence (AI), have been extensively applied in detecting various eye diseases, including diabetic retinopathy (DR), their applications in detecting DMI remain unexplored, partly due to skepticism among ophthalmologists regarding its feasibility. In this study, we propose a graph neural network-based multispectral view learning (GNN-MSVL) model designed to detect DMI from CFPs. The model leverages higher spectral resolution to capture subtle changes in fundus reflectance caused by ischemic tissue, enhancing sensitivity to DMI-related features. The proposed approach begins with computational multispectral imaging (CMI) to reconstruct 24-wavelength multispectral fundus images from CFPs. ResNeXt101 is employed as the backbone for multi-view learning to extract features from the reconstructed images. Additionally, a GNN with a customized jumper connection strategy is designed to enhance cross-spectral relationships, facilitating comprehensive and efficient multispectral view learning. The study included a total of 1,078 macula-centered CFPs from 1,078 eyes of 592 patients with diabetes, of which 530 CFPs from 530 eyes of 300 patients were diagnosed with DMI. The model achieved an accuracy of 84.7 percent and an area under the receiver operating characteristic curve (AUROC) of 0.900 (95 percent CI: 0.852-0.937) on eye-level, outperforming both the baseline model trained from CFPs and human experts (p-values less than 0.01). These findings suggest that AI-based CFP analysis holds promise for detecting DMI, contributing to its early and low-cost screening.
Authors: Andrei Apostol, Maria Nutu
Abstract: In Romania, cardiovascular problems are the leading cause of death, accounting for nearly one-third of annual fatalities. The severity of this situation calls for innovative diagnosis method for cardiovascular diseases. This article aims to explore efficient, light-weight and rapid methods for arrhythmia diagnosis, in resource-constrained healthcare settings. Due to the lack of Romanian public medical data, we trained our systems using international public datasets, having in mind that the ECG signals are the same regardless the patients' nationality. Within this purpose, we combined multiple datasets, usually used in the field of arrhythmias classification: PTB-XL electrocardiography dataset , PTB Diagnostic ECG Database, China 12-Lead ECG Challenge Database, Georgia 12-Lead ECG Challenge Database, and St. Petersburg INCART 12-lead Arrhythmia Database. For the input data, we employed ECG signal processing methods, specifically a variant of the Pan-Tompkins algorithm, useful in arrhythmia classification because it provides a robust and efficient method for detecting QRS complexes in ECG signals. Additionally, we used machine learning techniques, widely used for the task of classification, including convolutional neural networks (1D CNNs, 2D CNNs, ResNet) and Vision Transformers (ViTs). The systems were evaluated in terms of accuracy and F1 score. We annalysed our dataset from two perspectives. First, we fed the systems with the ECG signals and the GRU-based 1D CNN model achieved the highest accuracy of 93.4% among all the tested architectures. Secondly, we transformed ECG signals into images and the CNN2D model achieved an accuracy of 92.16%.
Authors: Hongyi Chen, Jingtao Ding, Jianhai Shu, Xinchun Yu, Xiaojun Liang, Yong Li, Xiao-Ping Zhang
Abstract: Complex nonlinear system control faces challenges in achieving sample-efficient, reliable performance. While diffusion-based methods have demonstrated advantages over classical and reinforcement learning approaches in long-term control performance, they are limited by sample efficiency. This paper presents SEDC (Sample-Efficient Diffusion-based Control), a novel diffusion-based control framework addressing three core challenges: high-dimensional state-action spaces, nonlinear system dynamics, and the gap between non-optimal training data and near-optimal control solutions. Through three innovations - Decoupled State Diffusion, Dual-Mode Decomposition, and Guided Self-finetuning - SEDC achieves 39.5\%-49.4\% better control accuracy than baselines while using only 10\% of the training samples, as validated across three complex nonlinear dynamic systems. Our approach represents a significant advancement in sample-efficient control of complex nonlinear systems. The implementation of the code can be found at https://anonymous.4open.science/r/DIFOCON-C019.
Authors: Christine Lee, David Porfirio, Xinyu Jessica Wang, Kevin Zhao, Bilge Mutlu
Abstract: Automated planning is traditionally the domain of experts, utilized in fields like manufacturing and healthcare with the aid of expert planning tools. Recent advancements in LLMs have made planning more accessible to everyday users due to their potential to assist users with complex planning tasks. However, LLMs face several application challenges within end-user planning, including consistency, accuracy, and user trust issues. This paper introduces VeriPlan, a system that applies formal verification techniques, specifically model checking, to enhance the reliability and flexibility of LLMs for end-user planning. In addition to the LLM planner, VeriPlan includes three additional core features -- a rule translator, flexibility sliders, and a model checker -- that engage users in the verification process. Through a user study (n=12), we evaluate VeriPlan, demonstrating improvements in the perceived quality, usability, and user satisfaction of LLMs. Our work shows the effective integration of formal verification and user-control features with LLMs for end-user planning tasks.
Authors: Che Liu, Cheng Ouyang, Zhongwei Wan, Haozhe Wang, Wenjia Bai, Rossella Arcucci
Abstract: Recent advances in multimodal ECG representation learning center on aligning ECG signals with paired free-text reports. However, suboptimal alignment persists due to the complexity of medical language and the reliance on a full 12-lead setup, which is often unavailable in under-resourced settings. To tackle these issues, we propose **K-MERL**, a knowledge-enhanced multimodal ECG representation learning framework. **K-MERL** leverages large language models to extract structured knowledge from free-text reports and employs a lead-aware ECG encoder with dynamic lead masking to accommodate arbitrary lead inputs. Evaluations on six external ECG datasets show that **K-MERL** achieves state-of-the-art performance in zero-shot classification and linear probing tasks, while delivering an average **16%** AUC improvement over existing methods in partial-lead zero-shot classification.
Authors: Minh Duc Vu, Jieshan Chen, Zhenchang Xing, Qinghua Lu, Xiwei Xu, Qian Fu
Abstract: With the proliferation of data across various domains, there is a critical demand for tools that enable non-experts to derive meaningful insights without deep data analysis skills. To address this need, existing automatic fact sheet generation tools offer heuristic-based solutions to extract facts and generate stories. However, they inadequately grasp the semantics of data and struggle to generate narratives that fully capture the semantics of the dataset or align the fact sheet with specific user needs. Addressing these shortcomings, this paper introduces \tool, a novel tool designed for the automatic generation and customisation of fact sheets. \tool applies the concept of collaborative AI workers to transform raw tabular dataset into comprehensive, visually compelling fact sheets. We define effective taxonomy to profile AI worker for specialised tasks. Furthermore, \tool empowers users to refine these fact sheets through intuitive natural language commands, ensuring the final outputs align closely with individual preferences and requirements. Our user evaluation with 18 participants confirms that \tool not only surpasses state-of-the-art baselines in automated fact sheet production but also provides a positive user experience during customization tasks.
Authors: Fangyuan Yu
Abstract: Modern language models rely on static vocabularies, fixed before pretraining, in contrast to the adaptive vocabulary acquisition observed in human language learning. To bridge this gap, we introduce vocabulary curriculum learning, an approach that improves pretraining efficiency with log-linear scaling gains relative to vocabulary size. Our method alternates between entropy-guided vocabulary expansion and model optimization, enabling models to learn transferable representations across diverse tokenization granularities. This approach naturally gives rise to an optimal computation allocation pattern: longer tokens capture predictable content, while shorter tokens focus on more complex, harder-to-predict contexts. Experiments on small-scale GPT models demonstrate improved scaling efficiency, reinforcing the effectiveness of dynamic tokenization. We release our code to support further research and plan to extend our experiments to larger models and diverse domains.
Authors: Souliman Alghnam, Mohammad Alhussien, Khaled Shaheen
Abstract: Speech enhancement plays an essential role in improving the quality of speech signals in noisy environments. This paper investigates the efficacy of integrating Bidirectional Gated Recurrent Units (BGRU) and Transformer models for speech enhancement tasks. Through a comprehensive experimental evaluation, our study demonstrates the superiority of this hybrid architecture over traditional methods and standalone models. The combined BGRU-Transformer framework excels in capturing temporal dependencies and learning complex signal patterns, leading to enhanced noise reduction and improved speech quality. Results show significant performance gains compared to existing approaches, highlighting the potential of this integrated model in real-world applications. The seamless integration of BGRU and Transformer architectures not only enhances system robustness but also opens the road for advanced speech processing techniques. This research contributes to the ongoing efforts in speech enhancement technology and sets a solid foundation for future investigations into optimizing model architectures, exploring many application scenarios, and advancing the field of speech processing in noisy environments.
Authors: Yuhan Chen, Yihong Luo, Yifan Song, Pengwen Dai, Jing Tang, Xiaochun Cao
Abstract: Despite extensive research efforts focused on OOD detection on images, OOD detection on nodes in graph learning remains underexplored. The dependence among graph nodes hinders the trivial adaptation of existing approaches on images that assume inputs to be i.i.d. sampled, since many unique features and challenges specific to graphs are not considered, such as the heterophily issue. Recently, GNNSafe, which considers node dependence, adapted energy-based detection to the graph domain with state-of-the-art performance, however, it has two serious issues: 1) it derives node energy from classification logits without specifically tailored training for modeling data distribution, making it less effective at recognizing OOD data; 2) it highly relies on energy propagation, which is based on homophily assumption and will cause significant performance degradation on heterophilic graphs, where the node tends to have dissimilar distribution with its neighbors. To address the above issues, we suggest training EBMs by MLE to enhance data distribution modeling and remove energy propagation to overcome the heterophily issues. However, training EBMs via MLE requires performing MCMC sampling on both node feature and node neighbors, which is challenging due to the node interdependence and discrete graph topology. To tackle the sampling challenge, we introduce DeGEM, which decomposes the learning process into two parts: a graph encoder that leverages topology information for node representations and an energy head that operates in latent space. Extensive experiments validate that DeGEM, without OOD exposure during training, surpasses previous state-of-the-art methods, achieving an average AUROC improvement of 6.71% on homophilic graphs and 20.29% on heterophilic graphs, and even outperform methods trained with OOD exposure. Our code is available at: https://github.com/draym28/DeGEM.
Authors: Hongyi Chen, Jingtao Ding, Xiaojun Liang, Yong Li, Xiao-Ping Zhang
Abstract: The source localization problem in graph information propagation is crucial for managing various network disruptions, from misinformation spread to infrastructure failures. While recent deep generative approaches have shown promise in this domain, their effectiveness is limited by the scarcity of real-world propagation data. This paper introduces SIDSL (\textbf{S}tructure-prior \textbf{I}nformed \textbf{D}iffusion model for \textbf{S}ource \textbf{L}ocalization), a novel framework that addresses three key challenges in limited-data scenarios: unknown propagation patterns, complex topology-propagation relationships, and class imbalance between source and non-source nodes. SIDSL incorporates topology-aware priors through graph label propagation and employs a propagation-enhanced conditional denoiser with a GNN-parameterized label propagation module (GNN-LP). Additionally, we propose a structure-prior biased denoising scheme that initializes from structure-based source estimations rather than random noise, effectively countering class imbalance issues. Experimental results across four real-world datasets demonstrate SIDSL's superior performance, achieving 7.5-13.3% improvements in F1 scores compared to state-of-the-art methods. Notably, when pretrained with simulation data of synthetic patterns, SIDSL maintains robust performance with only 10% of training data, surpassing baselines by more than 18.8%. These results highlight SIDSL's effectiveness in real-world applications where labeled data is scarce.
Authors: Sonalika Subudhi, Alok Kumar Pati, Sephali Bose, Subhasmita Sahoo, Avipsa Pattanaik, Biswa Mohan Acharya
Abstract: Groundwater is eventually undermined by human exercises, such as fast industrialization, urbanization, over-extraction, and contamination from agrarian and urban sources. From among the different contaminants, the presence of heavy metals like cadmium (Cd), chromium (Cr), arsenic (As), and lead (Pb) proves to have serious dangers when present in huge concentrations in groundwater. Long-term usage of these poisonous components may lead to neurological disorders, kidney failure and different sorts of cancer. To address these issues, this study developed a machine learning-based predictive model to evaluate the Groundwater Quality Index (GWQI) and identify the main contaminants which are affecting the water quality. It has been achieved with the help of a hybrid machine learning model i.e. LCBoost Fusion . The model has undergone several processes like data preprocessing, hyperparameter tuning using Differential Evolution (DE) optimization, and evaluation through cross-validation. The LCBoost Fusion model outperforms individual models (CatBoost and LightGBM), by achieving low RMSE (0.6829), MSE (0.5102), MAE (0.3147) and a high R$^2$ score of 0.9809. Feature importance analysis highlights Potassium (K), Fluoride (F) and Total Hardness (TH) as the most influential indicators of groundwater contamination. This research successfully demonstrates the application of machine learning in assessing groundwater quality risks in Odisha. The proposed LCBoost Fusion model offers a reliable and efficient approach for real-time groundwater monitoring and risk mitigation. These findings will help the environmental organizations and the policy makers to map out targeted places for sustainable groundwater management. Future work will focus on using remote sensing data and developing an interactive decision-making system for groundwater quality assessment.
Authors: Mingyuan Sun, Zheng Fang, Jiaxu Wang, Junjie Jiang, Delei Kong, Chenming Hu, Yuetong Fang, Renjing Xu
Abstract: The increasing complexity and parameter count of Convolutional Neural Networks (CNNs) and Transformers pose challenges in terms of computational efficiency and resource demands. Pruning has been identified as an effective strategy to address these challenges by removing redundant elements such as neurons, channels, or connections, thereby enhancing computational efficiency without heavily compromising performance. This paper builds on the foundational work of Optimal Brain Damage (OBD) by advancing the methodology of parameter importance estimation using the Hessian matrix. Unlike previous approaches that rely on approximations, we introduce Optimal Brain Apoptosis (OBA), a novel pruning method that calculates the Hessian-vector product value directly for each parameter. By decomposing the Hessian matrix across network layers and identifying conditions under which inter-layer Hessian submatrices are non-zero, we propose a highly efficient technique for computing the second-order Taylor expansion of parameters. This approach allows for a more precise pruning process, particularly in the context of CNNs and Transformers, as validated in our experiments including VGG19, ResNet32, ResNet50, and ViT-B/16 on CIFAR10, CIFAR100 and Imagenet datasets. Our code is available at https://github.com/NEU-REAL/OBA.
Authors: Pusheng Xu, Yue Wu, Kai Jin, Xiaolan Chen, Mingguang He, Danli Shi
Abstract: Purpose: To evaluate the accuracy and reasoning ability of DeepSeek-R1 and three other recently released large language models (LLMs) in bilingual complex ophthalmology cases. Methods: A total of 130 multiple-choice questions (MCQs) related to diagnosis (n = 39) and management (n = 91) were collected from the Chinese ophthalmology senior professional title examination and categorized into six topics. These MCQs were translated into English using DeepSeek-R1. The responses of DeepSeek-R1, Gemini 2.0 Pro, OpenAI o1 and o3-mini were generated under default configurations between February 15 and February 20, 2025. Accuracy was calculated as the proportion of correctly answered questions, with omissions and extra answers considered incorrect. Reasoning ability was evaluated through analyzing reasoning logic and the causes of reasoning error. Results: DeepSeek-R1 demonstrated the highest overall accuracy, achieving 0.862 in Chinese MCQs and 0.808 in English MCQs. Gemini 2.0 Pro, OpenAI o1, and OpenAI o3-mini attained accuracies of 0.715, 0.685, and 0.692 in Chinese MCQs (all P<0.001 compared with DeepSeek-R1), and 0.746 (P=0.115), 0.723 (P=0.027), and 0.577 (P<0.001) in English MCQs, respectively. DeepSeek-R1 achieved the highest accuracy across five topics in both Chinese and English MCQs. It also excelled in management questions conducted in Chinese (all P<0.05). Reasoning ability analysis showed that the four LLMs shared similar reasoning logic. Ignoring key positive history, ignoring key positive signs, misinterpretation medical data, and too aggressive were the most common causes of reasoning errors. Conclusion: DeepSeek-R1 demonstrated superior performance in bilingual complex ophthalmology reasoning tasks than three other state-of-the-art LLMs. While its clinical applicability remains challenging, it shows promise for supporting diagnosis and clinical decision-making.
Authors: Jia Yu, Yan Zhu, Peiyao Fu, Tianyi Chen, Junbo Huang, Quanlin Li, Pinghong Zhou, Zhihua Wang, Fei Wu, Shuo Wang, Xian Yang
Abstract: Colorectal cancer (CRC) is a significant global health concern, and early detection through screening plays a critical role in reducing mortality. While deep learning models have shown promise in improving polyp detection, classification, and segmentation, their generalization across diverse clinical environments, particularly with out-of-distribution (OOD) data, remains a challenge. Multi-center datasets like PolypGen have been developed to address these issues, but their collection is costly and time-consuming. Traditional data augmentation techniques provide limited variability, failing to capture the complexity of medical images. Diffusion models have emerged as a promising solution for generating synthetic polyp images, but the image generation process in current models mainly relies on segmentation masks as the condition, limiting their ability to capture the full clinical context. To overcome these limitations, we propose a Progressive Spectrum Diffusion Model (PSDM) that integrates diverse clinical annotations-such as segmentation masks, bounding boxes, and colonoscopy reports-by transforming them into compositional prompts. These prompts are organized into coarse and fine components, allowing the model to capture both broad spatial structures and fine details, generating clinically accurate synthetic images. By augmenting training data with PSDM-generated samples, our model significantly improves polyp detection, classification, and segmentation. For instance, on the PolypGen dataset, PSDM increases the F1 score by 2.12% and the mean average precision by 3.09%, demonstrating superior performance in OOD scenarios and enhanced generalization.
Authors: Tushar Aggarwal, Kumar Tanmay, Ayush Agrawal, Kumar Ayush, Hamid Palangi, Paul Pu Liang
Abstract: Multilingual language models (LMs) are expected to recall factual knowledge consistently across languages, yet they often fail to transfer knowledge between languages even when they possess the correct information in one of the languages. For example, we find that an LM may correctly identify Rashed Al Shashai as being from Saudi Arabia when asked in Arabic, but consistently fails to do so when asked in English or Swahili. To systematically investigate this limitation, we introduce a benchmark of 10,000 country-related facts across 13 languages and propose three novel metrics: Factual Recall Score, Knowledge Transferability Score, and Cross-Lingual Factual Knowledge Transferability Score-to quantify factual recall and knowledge transferability in LMs across different languages. Our results reveal fundamental weaknesses in today's state-of-the-art LMs, particularly in cross-lingual generalization where models fail to transfer knowledge effectively across different languages, leading to inconsistent performance sensitive to the language used. Our findings emphasize the need for LMs to recognize language-specific factual reliability and leverage the most trustworthy information across languages. We release our benchmark and evaluation framework to drive future research in multilingual knowledge transfer.
Authors: Tianmi Ma, Jiawei Du, Wenxin Huang, Wenjie Wang, Liang Xie, Xian Zhong, Joey Tianyi Zhou
Abstract: Recent advancements in large language models (LLMs) have significantly improved performance in natural language processing tasks. However, their ability to generalize to dynamic, unseen tasks, particularly in numerical reasoning, remains a challenge. Existing benchmarks mainly evaluate LLMs on problems with predefined optimal solutions, which may not align with real-world scenarios where clear answers are absent. To bridge this gap, we design the Agent Trading Arena, a virtual numerical game simulating complex economic systems through zero-sum games, where agents invest in stock portfolios. Our experiments reveal that LLMs, including GPT-4o, struggle with algebraic reasoning when dealing with plain-text stock data, often focusing on local details rather than global trends. In contrast, LLMs perform significantly better with geometric reasoning when presented with visual data, such as scatter plots or K-line charts, suggesting that visual representations enhance numerical reasoning. This capability is further improved by incorporating the reflection module, which aids in the analysis and interpretation of complex data. We validate our findings on NASDAQ Stock dataset, where LLMs demonstrate stronger reasoning with visual data compared to text. Our code and data are publicly available at https://github.com/wekjsdvnm/Agent-Trading-Arena.git.
Authors: Nikolai Rekut, Alexey Orlov, Klea Ziu, Elizaveta Starykh, Martin Takac, Aleksandr Beznosikov
Abstract: In recent years, machine learning has profoundly reshaped the field of chemistry, facilitating significant advancements across various applications, including the prediction of molecular properties and the generation of molecular structures. Language models and graph-based models are extensively utilized within this domain, consistently achieving state-of-the-art results across an array of tasks. However, the prevailing practice of representing chemical compounds in the SMILES format -- used by most datasets and many language models -- presents notable limitations as a training data format. In contrast, chemical fingerprints offer a more physically informed representation of compounds, thereby enhancing their suitability for model training. This study aims to develop a language model that is specifically trained on fingerprints. Furthermore, we introduce a bimodal architecture that integrates this language model with a graph model. Our proposed methodology synthesizes these approaches, utilizing RoBERTa as the language model and employing Graph Isomorphism Networks (GIN), Graph Convolutional Networks (GCN) and Graphormer as graph models. This integration results in a significant improvement in predictive performance compared to conventional strategies for tasks such as Quantitative Structure-Activity Relationship (QSAR) and the prediction of nuclear magnetic resonance (NMR) spectra, among others.
Authors: Varun Vashisht, Samar Singh, Mihir Konduskar, Jaskaran Singh Walia, Vukosi Marivate
Abstract: Due to the lack of quality data for low-resource Bantu languages, significant challenges are presented in text classification and other practical implementations. In this paper, we introduce an advanced model combining Language-Independent Data Augmentation (LiDA) with Multi-Head Attention based weighted embeddings to selectively enhance critical data points and improve text classification performance. This integration allows us to create robust data augmentation strategies that are effective across various linguistic contexts, ensuring that our model can handle the unique syntactic and semantic features of Bantu languages. This approach not only addresses the data scarcity issue but also sets a foundation for future research in low-resource language processing and classification tasks.
Authors: Shlok Mehendale, Aditya Challa, Rahul Yedida, Sravan Danda, Santonu Sarkar, Snehanshu Saha
Abstract: Which principle underpins the design of an effective anomaly detection loss function? The answer lies in the concept of \rnthm{} theorem, a fundamental concept in measure theory. The key insight is -- Multiplying the vanilla loss function with the \rnthm{} derivative improves the performance across the board. We refer to this as RN-Loss. This is established using PAC learnability of anomaly detection. We further show that the \rnthm{} derivative offers important insights into unsupervised clustering based anomaly detections as well. We evaluate our algorithm on 96 datasets, including univariate and multivariate data from diverse domains, including healthcare, cybersecurity, and finance. We show that RN-Derivative algorithms outperform state-of-the-art methods on 68\% of Multivariate datasets (based on F-1 scores) and also achieves peak F1-scores on 72\% of time series (Univariate) datasets.
Authors: Yashan Wang, Shangda Wu, Jianhuai Hu, Xingjian Du, Yueqi Peng, Yongxin Huang, Shuai Fan, Xiaobing Li, Feng Yu, Maosong Sun
Abstract: We introduce NotaGen, a symbolic music generation model aiming to explore the potential of producing high-quality classical sheet music. Inspired by the success of Large Language Models (LLMs), NotaGen adopts pre-training, fine-tuning, and reinforcement learning paradigms (henceforth referred to as the LLM training paradigms). It is pre-trained on 1.6M pieces of music, and then fine-tuned on approximately 9K high-quality classical compositions conditioned on "period-composer-instrumentation" prompts. For reinforcement learning, we propose the CLaMP-DPO method, which further enhances generation quality and controllability without requiring human annotations or predefined rewards. Our experiments demonstrate the efficacy of CLaMP-DPO in symbolic music generation models with different architectures and encoding schemes. Furthermore, subjective A/B tests show that NotaGen outperforms baseline models against human compositions, greatly advancing musical aesthetics in symbolic music generation.The project homepage is https://electricalexis.github.io/notagen-demo.
Authors: Qiuchen Wang, Ruixue Ding, Zehui Chen, Weiqi Wu, Shihang Wang, Pengjun Xie, Feng Zhao
Abstract: Understanding information from visually rich documents remains a significant challenge for traditional Retrieval-Augmented Generation (RAG) methods. Existing benchmarks predominantly focus on image-based question answering (QA), overlooking the fundamental challenges of efficient retrieval, comprehension, and reasoning within dense visual documents. To bridge this gap, we introduce ViDoSeek, a novel dataset designed to evaluate RAG performance on visually rich documents requiring complex reasoning. Based on it, we identify key limitations in current RAG approaches: (i) purely visual retrieval methods struggle to effectively integrate both textual and visual features, and (ii) previous approaches often allocate insufficient reasoning tokens, limiting their effectiveness. To address these challenges, we propose ViDoRAG, a novel multi-agent RAG framework tailored for complex reasoning across visual documents. ViDoRAG employs a Gaussian Mixture Model (GMM)-based hybrid strategy to effectively handle multi-modal retrieval. To further elicit the model's reasoning capabilities, we introduce an iterative agent workflow incorporating exploration, summarization, and reflection, providing a framework for investigating test-time scaling in RAG domains. Extensive experiments on ViDoSeek validate the effectiveness and generalization of our approach. Notably, ViDoRAG outperforms existing methods by over 10% on the competitive ViDoSeek benchmark.
Authors: Joshua Sakthivel Raju, Sanjay S, Jaskaran Singh Walia, Srinivas Raghav, Vukosi Marivate
Abstract: Language model compression through knowledge distillation has emerged as a promising approach for deploying large language models in resource-constrained environments. However, existing methods often struggle to maintain performance when distilling multilingual models, especially for low-resource languages. In this paper, we present a novel hybrid distillation approach that combines traditional knowledge distillation with a simplified attention matching mechanism, specifically designed for multilingual contexts. Our method introduces an extremely compact student model architecture, significantly smaller than conventional multilingual models. We evaluate our approach on five African languages: Kinyarwanda, Swahili, Hausa, Igbo, and Yoruba. The distilled student model; AfroXLMR-Comet successfully captures both the output distribution and internal attention patterns of a larger teacher model (AfroXLMR-Large) while reducing the model size by over 85%. Experimental results demonstrate that our hybrid approach achieves competitive performance compared to the teacher model, maintaining an accuracy within 85% of the original model's performance while requiring substantially fewer computational resources. Our work provides a practical framework for deploying efficient multilingual models in resource-constrained environments, particularly benefiting applications involving African languages.
Authors: Rikuto Kotoge, Ziwei Yang, Zheng Chen, Yushun Dong, Yasuko Matsubara, Jimeng Sun, Yasushi Sakurai
Abstract: Biological knowledge bases provide systemically functional pathways of cells or organisms in terms of molecular interaction. However, recognizing more targeted pathways, particularly when incorporating wet-lab experimental data, remains challenging and typically requires downstream biological analyses and expertise. In this paper, we frame this challenge as a solvable graph learning and explaining task and propose a novel pathway inference framework, ExPath, that explicitly integrates experimental data, specifically amino acid sequences (AA-seqs), to classify various graphs (bio-networks) in biological databases. The links (representing pathways) that contribute more to classification can be considered as targeted pathways. Technically, ExPath comprises three components: (1) a large protein language model (pLM) that encodes and embeds AA-seqs into graph, overcoming traditional obstacles in processing AA-seq data, such as BLAST; (2) PathMamba, a hybrid architecture combining graph neural networks (GNNs) with state-space sequence modeling (Mamba) to capture both local interactions and global pathway-level dependencies; and (3) PathExplainer, a subgraph learning module that identifies functionally critical nodes and edges through trainable pathway masks. We also propose ML-oriented biological evaluations and a new metric. The experiments involving 301 bio-networks evaluations demonstrate that pathways inferred by ExPath maintain biological meaningfulness. We will publicly release curated 301 bio-network data soon.
Authors: Yuhao Zheng, Chenghua Gong, Rui Sun, Juyuan Zhang, Liming Pan, Linyuan Lv
Abstract: Popularity prediction in information cascades plays a crucial role in social computing, with broad applications in viral marketing, misinformation control, and content recommendation. However, information propagation mechanisms, user behavior, and temporal activity patterns exhibit significant diversity, necessitating a foundational model capable of adapting to such variations. At the same time, the amount of available cascade data remains relatively limited compared to the vast datasets used for training large language models (LLMs). Recent studies have demonstrated the feasibility of leveraging LLMs for time-series prediction by exploiting commonalities across different time-series domains. Building on this insight, we introduce the Autoregressive Information Cascade Predictor (AutoCas), an LLM-enhanced model designed specifically for cascade popularity prediction. Unlike natural language sequences, cascade data is characterized by complex local topologies, diffusion contexts, and evolving dynamics, requiring specialized adaptations for effective LLM integration. To address these challenges, we first tokenize cascade data to align it with sequence modeling principles. Next, we reformulate cascade diffusion as an autoregressive modeling task to fully harness the architectural strengths of LLMs. Beyond conventional approaches, we further introduce prompt learning to enhance the synergy between LLMs and cascade prediction. Extensive experiments demonstrate that AutoCas significantly outperforms baseline models in cascade popularity prediction while exhibiting scaling behavior inherited from LLMs. Code is available at this repository: https://anonymous.4open.science/r/AutoCas-85C6
Authors: Pei Liu (The Hong Kong University of Science and Technology), Haipeng Liu (Li Auto Inc), Haichao Liu (The Hong Kong University of Science and Technology), Xin Liu (The Hong Kong University of Science and Technology), Jinxin Ni (Xiamen University), Jun Ma (The Hong Kong University of Science and Technology, The Hong Kong University of Science and Technology)
Abstract: Human drivers adeptly navigate complex scenarios by utilizing rich attentional semantics, but the current autonomous systems struggle to replicate this ability, as they often lose critical semantic information when converting 2D observations into 3D space. In this sense, it hinders their effective deployment in dynamic and complex environments. Leveraging the superior scene understanding and reasoning abilities of Vision-Language Models (VLMs), we propose VLM-E2E, a novel framework that uses the VLMs to enhance training by providing attentional cues. Our method integrates textual representations into Bird's-Eye-View (BEV) features for semantic supervision, which enables the model to learn richer feature representations that explicitly capture the driver's attentional semantics. By focusing on attentional semantics, VLM-E2E better aligns with human-like driving behavior, which is critical for navigating dynamic and complex environments. Furthermore, we introduce a BEV-Text learnable weighted fusion strategy to address the issue of modality importance imbalance in fusing multimodal information. This approach dynamically balances the contributions of BEV and text features, ensuring that the complementary information from visual and textual modality is effectively utilized. By explicitly addressing the imbalance in multimodal fusion, our method facilitates a more holistic and robust representation of driving environments. We evaluate VLM-E2E on the nuScenes dataset and demonstrate its superiority over state-of-the-art approaches, showcasing significant improvements in performance.
Authors: Yifeng Wang, Yi Zhao
Abstract: Low-cost accelerometers play a crucial role in modern society due to their advantages of small size, ease of integration, wearability, and mass production, making them widely applicable in automotive systems, aerospace, and wearable technology. However, this widely used sensor suffers from severe accuracy and range limitations. To this end, we propose a honed-energy regularized and optimal supervised GAN (HEROS-GAN), which transforms low-cost sensor signals into high-cost equivalents, thereby overcoming the precision and range limitations of low-cost accelerometers. Due to the lack of frame-level paired low-cost and high-cost signals for training, we propose an Optimal Transport Supervision (OTS), which leverages optimal transport theory to explore potential consistency between unpaired data, thereby maximizing supervisory information. Moreover, we propose a Modulated Laplace Energy (MLE), which injects appropriate energy into the generator to encourage it to break range limitations, enhance local changes, and enrich signal details. Given the absence of a dedicated dataset, we specifically establish a Low-cost Accelerometer Signal Enhancement Dataset (LASED) containing tens of thousands of samples, which is the first dataset serving to improve the accuracy and range of accelerometers and is released in Github. Experimental results demonstrate that a GAN combined with either OTS or MLE alone can surpass the previous signal enhancement SOTA methods by an order of magnitude. Integrating both OTS and MLE, the HEROS-GAN achieves remarkable results, which doubles the accelerometer range while reducing signal noise by two orders of magnitude, establishing a benchmark in the accelerometer signal processing.
Authors: Yishuai Cai, Xinglin Chen, Zhongxuan Cai, Yunxin Mao, Minglong Li, Wenjing Yang, Ji Wang
Abstract: Multi-robot task planning and collaboration are critical challenges in robotics. While Behavior Trees (BTs) have been established as a popular control architecture and are plannable for a single robot, the development of effective multi-robot BT planning algorithms remains challenging due to the complexity of coordinating diverse action spaces. We propose the Multi-Robot Behavior Tree Planning (MRBTP) algorithm, with theoretical guarantees of both soundness and completeness. MRBTP features cross-tree expansion to coordinate heterogeneous actions across different BTs to achieve the team's goal. For homogeneous actions, we retain backup structures among BTs to ensure robustness and prevent redundant execution through intention sharing. While MRBTP is capable of generating BTs for both homogeneous and heterogeneous robot teams, its efficiency can be further improved. We then propose an optional plugin for MRBTP when Large Language Models (LLMs) are available to reason goal-related actions for each robot. These relevant actions can be pre-planned to form long-horizon subtrees, significantly enhancing the planning speed and collaboration efficiency of MRBTP. We evaluate our algorithm in warehouse management and everyday service scenarios. Results demonstrate MRBTP's robustness and execution efficiency under varying settings, as well as the ability of the pre-trained LLM to generate effective task-specific subtrees for MRBTP.
Authors: Wenkai Yang, Shuming Ma, Yankai Lin, Furu Wei
Abstract: Recent studies have shown that making a model spend more time thinking through longer Chain of Thoughts (CoTs) enables it to gain significant improvements in complex reasoning tasks. While current researches continue to explore the benefits of increasing test-time compute by extending the CoT lengths of Large Language Models (LLMs), we are concerned about a potential issue hidden behind the current pursuit of test-time scaling: Would excessively scaling the CoT length actually bring adverse effects to a model's reasoning performance? Our explorations on mathematical reasoning tasks reveal an unexpected finding that scaling with longer CoTs can indeed impair the reasoning performance of LLMs in certain domains. Moreover, we discover that there exists an optimal scaled length distribution that differs across different domains. Based on these insights, we propose a Thinking-Optimal Scaling strategy. Our method first uses a small set of seed data with varying response length distributions to teach the model to adopt different reasoning efforts for deep thinking. Then, the model selects its shortest correct response under different reasoning efforts on additional problems for self-improvement. Our self-improved models built upon Qwen2.5-32B-Instruct outperform other distillation-based 32B o1-like models across various math benchmarks, and achieve performance on par with QwQ-32B-Preview.
Authors: Samuele Sabella, Chiara Boldrini, Lorenzo Valerio, Andrea Passarella, Marco Conti
Abstract: Decentralized federated learning (DFL) enables devices to collaboratively train models over complex network topologies without relying on a central controller. In this setting, local data remains private, but its quality and quantity can vary significantly across nodes. The extent to which a fully decentralized system is vulnerable to poor-quality or corrupted data remains unclear, but several factors could contribute to potential risks. Without a central authority, there can be no unified mechanism to detect or correct errors, and each node operates with a localized view of the data distribution, making it difficult for the node to assess whether its perspective aligns with the true distribution. Moreover, models trained on low-quality data can propagate through the network, amplifying errors. To explore the impact of low-quality data on DFL, we simulate two scenarios with degraded data quality -- one where the corrupted data is evenly distributed in a subset of nodes and one where it is concentrated on a single node -- using a decentralized implementation of FedAvg. Our results reveal that averaging-based decentralized learning is remarkably robust to localized bad data, even when the corrupted data resides in the most influential nodes of the network. Counterintuitively, this robustness is further enhanced when the corrupted data is concentrated on a single node, regardless of its centrality in the communication network topology. This phenomenon is explained by the averaging process, which ensures that no single node -- however central -- can disproportionately influence the overall learning process.
Authors: Chengkun Cai, Haoliang Liu, Xu Zhao, Zhongyu Jiang, Tianfang Zhang, Zongkai Wu, Jenq-Neng Hwang, Serge Belongie, Lei Li
Abstract: In the rapidly evolving field of image generation, achieving precise control over generated content and maintaining semantic consistency remain significant limitations, particularly concerning grounding techniques and the necessity for model fine-tuning. To address these challenges, we propose BayesGenie, an off-the-shelf approach that integrates Large Language Models (LLMs) with Bayesian Optimization to facilitate precise and user-friendly image editing. Our method enables users to modify images through natural language descriptions without manual area marking, while preserving the original image's semantic integrity. Unlike existing techniques that require extensive pre-training or fine-tuning, our approach demonstrates remarkable adaptability across various LLMs through its model-agnostic design. BayesGenie employs an adapted Bayesian optimization strategy to automatically refine the inference process parameters, achieving high-precision image editing with minimal user intervention. Through extensive experiments across diverse scenarios, we demonstrate that our framework significantly outperforms existing methods in both editing accuracy and semantic preservation, as validated using different LLMs including Claude3 and GPT-4.
Authors: B. Sun, P. Li\`o
Abstract: In this study, we propose MHEX+, a framework adaptable to any U-Net architecture. Built upon MHEX+, we introduce novel U-Net variants, EU-Nets, which enhance explainability and uncertainty estimation, addressing the limitations of traditional U-Net models while improving performance and stability. A key innovation is the Equivalent Convolutional Kernel, which unifies consecutive convolutional layers, boosting interpretability. For uncertainty estimation, we propose the collaboration gradient approach, measuring gradient consistency across decoder layers. Notably, EU-Nets achieve an average accuracy improvement of 1.389\% and a variance reduction of 0.83\% across all networks and datasets in our experiments, requiring fewer than 0.1M parameters.
Authors: Jintao Zhang, Chendong Xiang, Haofeng Huang, Jia Wei, Haocheng Xi, Jun Zhu, Jianfei Chen
Abstract: An efficient attention implementation is essential for large models due to its quadratic time complexity. Fortunately, attention commonly exhibits sparsity, i.e., many values in the attention map are near zero, allowing for the omission of corresponding computations. Many studies have utilized the sparse pattern to accelerate attention. However, most existing works focus on optimizing attention within specific models by exploiting certain sparse patterns of the attention map. A universal sparse attention that guarantees both the speedup and end-to-end performance of diverse models remains elusive. In this paper, we propose SpargeAttn, a universal sparse and quantized attention for any model. Our method uses a two-stage online filter: in the first stage, we rapidly and accurately predict the attention map, enabling the skip of some matrix multiplications in attention. In the second stage, we design an online softmax-aware filter that incurs no extra overhead and further skips some matrix multiplications. Experiments show that our method significantly accelerates diverse models, including language, image, and video generation, without sacrificing end-to-end metrics. The codes are available at https://github.com/thu-ml/SpargeAttn.
Authors: Chenhao Gu, Ling Luo, Zainab Razia Zaidi, Shanika Karunasekera
Abstract: The rise of echo chambers on social media platforms has heightened concerns about polarization and the reinforcement of existing beliefs. Traditional approaches for simulating echo chamber formation have often relied on predefined rules and numerical simulations, which, while insightful, may lack the nuance needed to capture complex, real-world interactions. In this paper, we present a novel framework that leverages large language models (LLMs) as generative agents to simulate echo chamber dynamics within social networks. The novelty of our approach is that it incorporates both opinion updates and network rewiring behaviors driven by LLMs, allowing for a context-aware and semantically rich simulation of social interactions. Additionally, we utilize real-world Twitter (now X) data to benchmark the LLM-based simulation against actual social media behaviors, providing insights into the accuracy and realism of the generated opinion trends. Our results demonstrate the efficacy of LLMs in modeling echo chamber formation, capturing both structural and semantic dimensions of opinion clustering. %This work contributes to a deeper understanding of social influence dynamics and offers a new tool for studying polarization in online communities.
Authors: Lucy Farnik, Tim Lawson, Conor Houghton, Laurence Aitchison
Abstract: Sparse autoencoders (SAEs) have been successfully used to discover sparse and human-interpretable representations of the latent activations of LLMs. However, we would ultimately like to understand the computations performed by LLMs and not just their representations. The extent to which SAEs can help us understand computations is unclear because they are not designed to "sparsify" computations in any sense, only latent activations. To solve this, we propose Jacobian SAEs (JSAEs), which yield not only sparsity in the input and output activations of a given model component but also sparsity in the computation (formally, the Jacobian) connecting them. With a na\"ive implementation, the Jacobians in LLMs would be computationally intractable due to their size. One key technical contribution is thus finding an efficient way of computing Jacobians in this setup. We find that JSAEs extract a relatively large degree of computational sparsity while preserving downstream LLM performance approximately as well as traditional SAEs. We also show that Jacobians are a reasonable proxy for computational sparsity because MLPs are approximately linear when rewritten in the JSAE basis. Lastly, we show that JSAEs achieve a greater degree of computational sparsity on pre-trained LLMs than on the equivalent randomized LLM. This shows that the sparsity of the computational graph appears to be a property that LLMs learn through training, and suggests that JSAEs might be more suitable for understanding learned transformer computations than standard SAEs.
Authors: Shan He, Yalong Ma, Tao Song, Yongzhi Jiang, Xinkai Wu
Abstract: Planning a safe and feasible trajectory for autonomous vehicles in real-time by fully utilizing perceptual information in complex urban environments is challenging. In this paper, we propose a spatio-temporal trajectory planning method based on graph optimization. It efficiently extracts the multi-modal information of the perception module by constructing a semantic spatio-temporal map through separation processing of static and dynamic obstacles, and then quickly generates feasible trajectories via sparse graph optimization based on a semantic spatio-temporal hypergraph. Extensive experiments have proven that the proposed method can effectively handle complex urban public road scenarios and perform in real time. We will also release our codes to accommodate benchmarking for the research community
Authors: Dahun Shin, Dongyeop Lee, Jinseok Chung, Namhoon Lee
Abstract: Approximate second-order optimization methods often exhibit poorer generalization compared to first-order approaches. In this work, we look into this issue through the lens of the loss landscape and find that existing second-order methods tend to converge to sharper minima compared to SGD. In response, we propose Sassha, a novel second-order method designed to enhance generalization by explicitly reducing sharpness of the solution, while stabilizing the computation of approximate Hessians along the optimization trajectory. In fact, this sharpness minimization scheme is crafted also to accommodate lazy Hessian updates, so as to secure efficiency besides flatness. To validate its effectiveness, we conduct a wide range of standard deep learning experiments where Sassha demonstrates its outstanding generalization performance that is comparable to, and mostly better than, other methods. We provide a comprehensive set of analyses including convergence, robustness, stability, efficiency, and cost.
Authors: Zahra Dehghanighobadi, Asja Fischer, Muhammad Bilal Zafar
Abstract: Explanations are an important tool for gaining insights into the behavior of ML models, calibrating user trust and ensuring regulatory compliance. Past few years have seen a flurry of post-hoc methods for generating model explanations, many of which involve computing model gradients or solving specially designed optimization problems. However, owing to the remarkable reasoning abilities of Large Language Model (LLMs), self-explanation, that is, prompting the model to explain its outputs has recently emerged as a new paradigm. In this work, we study a specific type of self-explanations, self-generated counterfactual explanations (SCEs). We design tests for measuring the efficacy of LLMs in generating SCEs. Analysis over various LLM families, model sizes, temperature settings, and datasets reveals that LLMs sometimes struggle to generate SCEs. Even when they do, their prediction often does not agree with their own counterfactual reasoning.
Authors: Filippo Maria Bianchi, Jakob Grahn
Abstract: Snow avalanches present significant risks to human life and infrastructure, particularly in mountainous regions, making effective monitoring crucial. Traditional monitoring methods, such as field observations, are limited by accessibility, weather conditions, and cost. Satellite-borne Synthetic Aperture Radar (SAR) data has become an important tool for large-scale avalanche detection, as it can capture data in all weather conditions and across remote areas. However, traditional processing methods struggle with the complexity and variability of avalanches. This chapter reviews the application of deep learning for detecting and segmenting snow avalanches from SAR data. Early efforts focused on the binary classification of SAR images, while recent advances have enabled pixel-level segmentation, providing greater accuracy and spatial resolution. A case study using Sentinel-1 SAR data demonstrates the effectiveness of deep learning models for avalanche segmentation, achieving superior results over traditional methods. We also present an extension of this work, testing recent state-of-the-art segmentation architectures on an expanded dataset of over 4,500 annotated SAR images. The best-performing model among those tested was applied for large-scale avalanche detection across the whole of Norway, revealing important spatial and temporal patterns over several winter seasons.
Authors: Pablo Ortega, Eduardo Castell\'o Ferrer
Abstract: As robotic systems (RS) become more autonomous, they are becoming increasingly used in small spaces and offices to automate tasks such as cleaning, infrastructure maintenance, or resource management. In this paper, we propose iTrash, an intelligent trashcan that aims to improve recycling rates in small office spaces. For that, we ran a 5 day experiment and found that iTrash can produce an efficiency increase of more than 30% compared to traditional trashcans. The findings derived from this work, point to the fact that using iTrash not only increase recyclying rates, but also provides valuable data such as users behaviour or bin usage patterns, which cannot be taken from a normal trashcan. This information can be used to predict and optimize some tasks in these spaces. Finally, we explored the potential of using blockchain technology to create economic incentives for recycling, following a Save-as-you-Throw (SAYT) model.
Authors: Zhang Yuxuan, Li Ruizhe
Abstract: With the rapid development of large language models (LLMs), fully fine-tuning (FT) these models has become increasingly impractical due to the high computational demands. Additionally, FT can lead to catastrophic forgetting. As an alternative, Low-Rank Adaptation (LoRA) has been proposed, which fine-tunes only a small subset of parameters, achieving similar performance to FT while significantly reducing resource requirements. However, since LoRA inherits FT's design, the issue of catastrophic forgetting remains. To address these challenges, we propose SECURA: Sigmoid-Enhanced CUR Decomposition LoRA, a novel parameter-efficient fine-tuning (PEFT) variant that mitigates catastrophic forgetting while improving fine-tuning performance. Our method introduces a new normalization technique, SigNorm, to enhance parameter retention and overall performance. SECURA has been evaluated on a variety of tasks, including mathematical problem-solving (GSM8K), challenging question-answering (CNNDM), translation (NewsDE), and complex multiple-choice reasoning (LogiQA). Experimental results show that SECURA achieves an average fine-tuning improvement of 3.59% across four multiple-choice question (MCQ) tasks and a 2.51% improvement across five question-answering (QA) tasks on models such as Gemma2 2b, Qwen2 1.5b, Qwen 2 7b, Llama3 8b, and Llama3.1 8b, compared to DoRA. Moreover, SECURA demonstrates superior knowledge retention capabilities, maintaining more than 70% accuracy on basic LLM knowledge across 16 continual learning tests, outperforming Experience Replay (ER), Sequential Learning (SEQ), EWC, I-LoRA, and CUR-LoRA.
Authors: Mingkun Zhang, Keping Bi, Wei Chen, Jiafeng Guo, Xueqi Cheng
Abstract: In this paper, we aim to build an adversarially robust zero-shot image classifier. We ground our work on CLIP, a vision-language pre-trained encoder model that can perform zero-shot classification by matching an image with text prompts ``a photo of a
Authors: Gaye Colakoglu, G\"urkan Solmaz, Jonathan F\"urst
Abstract: This paper defines and explores the design space for information extraction (IE) from layout-rich documents using large language models (LLMs). The three core challenges of layout-aware IE with LLMs are 1) data structuring, 2) model engagement, and 3) output refinement. Our study delves into the sub-problems within these core challenges, such as input representation, chunking, prompting, and selection of LLMs and multimodal models. It examines the outcomes of different design choices through a new layout-aware IE test suite, benchmarking against the state-of-art (SoA) model LayoutLMv3. The results show that the configuration from one-factor-at-a-time (OFAT) trial achieves near-optimal results with 14.1 points F1-score gain from the baseline model, while full factorial exploration yields only a slightly higher 15.1 points gain at around 36x greater token usage. We demonstrate that well-configured general-purpose LLMs can match the performance of specialized models, providing a cost-effective alternative. Our test-suite is freely available at https://github.com/gayecolakoglu/LayIE-LLM.
Authors: Adnan Iltaf, Rayan Merghani Ahmed, Bin Li, Shoujun Zhou
Abstract: Medical image segmentation is crucial for clinical diagnosis and treatment planning, particularly for complex anatomical structures like vessels. In this work, we propose VesselSAM, a modified version of the Segmentation Anything Model (SAM), specifically designed for aortic vessel segmentation. VesselSAM incorporates AtrousLoRA, a novel module that combines Atrous Attention with Low-Rank Adaptation (LoRA), to improve segmentation performance. Atrous Attention enables the model to capture multi-scale contextual information, preserving both fine local details and broader global context. At the same time, LoRA facilitates efficient fine-tuning of the frozen SAM image encoder, reducing the number of trainable parameters and ensuring computational efficiency. We evaluate VesselSAM on two challenging datasets: the Aortic Vessel Tree (AVT) dataset and the Type-B Aortic Dissection (TBAD) dataset. VesselSAM achieves state-of-the-art performance with DSC scores of 93.50\%, 93.25\%, 93.02\%, and 93.26\% across multiple medical centers. Our results demonstrate that VesselSAM delivers high segmentation accuracy while significantly reducing computational overhead compared to existing large-scale models. This development paves the way for enhanced AI-based aortic vessel segmentation in clinical environments. The code and models will be released at https://github.com/Adnan-CAS/AtrousLora.
Authors: Atik Faysal, Mohammad Rostami, Taha Boushine, Reihaneh Gh. Roshan, Huaxia Wang, Nikhil Muralidhar
Abstract: We introduce DenoMAE2.0, an enhanced denoising masked autoencoder that integrates a local patch classification objective alongside traditional reconstruction loss to improve representation learning and robustness. Unlike conventional Masked Autoencoders (MAE), which focus solely on reconstructing missing inputs, DenoMAE2.0 introduces position-aware classification of unmasked patches, enabling the model to capture fine-grained local features while maintaining global coherence. This dual-objective approach is particularly beneficial in semi-supervised learning for wireless communication, where high noise levels and data scarcity pose significant challenges. We conduct extensive experiments on modulation signal classification across a wide range of signal-to-noise ratios (SNRs), from extremely low to moderately high conditions and in a low data regime. Our results demonstrate that DenoMAE2.0 surpasses its predecessor, Deno-MAE, and other baselines in both denoising quality and downstream classification accuracy. DenoMAE2.0 achieves a 1.1% improvement over DenoMAE on our dataset and 11.83%, 16.55% significant improved accuracy gains on the RadioML benchmark, over DenoMAE, for constellation diagram classification of modulation signals.
Authors: Jian Wu, Jiayu Zhang, Dongyuan Li, Linyi Yang, Aoxiao Zhong, Renhe Jiang, Qingsong Wen, Yue Zhang
Abstract: This paper introduces Leaderboard Auto Generation (LAG), a novel and well-organized framework for automatic generation of leaderboards on a given research topic in rapidly evolving fields like Artificial Intelligence (AI). Faced with a large number of AI papers updated daily, it becomes difficult for researchers to track every paper's proposed methods, experimental results, and settings, prompting the need for efficient automatic leaderboard construction. While large language models (LLMs) offer promise in automating this process, challenges such as multi-document summarization, leaderboard generation, and experiment fair comparison still remain under exploration. LAG solves these challenges through a systematic approach that involves the paper collection, experiment results extraction and integration, leaderboard generation, and quality evaluation. Our contributions include a comprehensive solution to the leaderboard construction problem, a reliable evaluation method, and experimental results showing the high quality of leaderboards.
Authors: Bingke Zhu, Xiaoxiao Wang, Minghui Jia, Yihan Tao, Xiao Kong, Ali Luo, Yingying Chen, Ming Tang, Jinqiao Wang
Abstract: Stellar flare events are critical observational samples for astronomical research; however, recorded flare events remain limited. Stellar flare forecasting can provide additional flare event samples to support research efforts. Despite this potential, no specialized models for stellar flare forecasting have been proposed to date. In this paper, we present extensive experimental evidence demonstrating that both stellar physical properties and historical flare records are valuable inputs for flare forecasting tasks. We then introduce FLARE (Forecasting Light-curve-based Astronomical Records via features Ensemble), the first-of-its-kind large model specifically designed for stellar flare forecasting. FLARE integrates stellar physical properties and historical flare records through a novel Soft Prompt Module and Residual Record Fusion Module. Our experiments on the publicly available Kepler light curve dataset demonstrate that FLARE achieves superior performance compared to other methods across all evaluation metrics. Finally, we validate the forecast capability of our model through a comprehensive case study.
Authors: He Wang, Tianyang Xu, Zhangyong Tang, Xiao-Jun Wu, Josef Kittler
Abstract: Multi-modal tracking is essential in single-object tracking (SOT), as different sensor types contribute unique capabilities to overcome challenges caused by variations in object appearance. However, existing unified RGB-X trackers (X represents depth, event, or thermal modality) either rely on the task-specific training strategy for individual RGB-X image pairs or fail to address the critical importance of modality-adaptive perception in real-world applications. In this work, we propose UASTrack, a unified adaptive selection framework that facilitates both model and parameter unification, as well as adaptive modality discrimination across various multi-modal tracking tasks. To achieve modality-adaptive perception in joint RGB-X pairs, we design a Discriminative Auto-Selector (DAS) capable of identifying modality labels, thereby distinguishing the data distributions of auxiliary modalities. Furthermore, we propose a Task-Customized Optimization Adapter (TCOA) tailored to various modalities in the latent space. This strategy effectively filters noise redundancy and mitigates background interference based on the specific characteristics of each modality. Extensive comparisons conducted on five benchmarks including LasHeR, GTOT, RGBT234, VisEvent, and DepthTrack, covering RGB-T, RGB-E, and RGB-D tracking scenarios, demonstrate our innovative approach achieves comparative performance by introducing only additional training parameters of 1.87M and flops of 1.95G. The code will be available at https://github.com/wanghe/UASTrack.
Authors: Jun Zeng, Debesh Jha, Ertugrul Aktas, Elif Keles, Alpay Medetalibeyoglu, Matthew Antalek, Amir A. Borhani, Daniela P. Ladner, Gorkem Durak, Ulas Bagci
Abstract: We present an end-to-end deep learning framework for automated liver cirrhosis stage estimation from multi-sequence MRI. Cirrhosis is the severe scarring (fibrosis) of the liver and a common endpoint of various chronic liver diseases. Early diagnosis is vital to prevent complications such as decompensation and cancer, which significantly decreases life expectancy. However, diagnosing cirrhosis in its early stages is challenging, and patients often present with life-threatening complications. Our approach integrates multi-scale feature learning with sequence-specific attention mechanisms to capture subtle tissue variations across cirrhosis progression stages. Using CirrMRI600+, a large-scale publicly available dataset of 628 high-resolution MRI scans from 339 patients, we demonstrate state-of-the-art performance in three-stage cirrhosis classification. Our best model achieves 72.8% accuracy on T1W and 63.8% on T2W sequences, significantly outperforming traditional radiomics-based approaches. Through extensive ablation studies, we show that our architecture effectively learns stage-specific imaging biomarkers. We establish new benchmarks for automated cirrhosis staging and provide insights for developing clinically applicable deep learning systems. The source code will be available at https://github.com/JunZengz/CirrhosisStage.
Authors: Jun Zeng, Ulas Bagci, Debesh Jha
Abstract: We present RMA-Mamba, a novel architecture that advances the capabilities of vision state space models through a specialized reverse mamba attention module (RMA). The key innovation lies in RMA-Mamba's ability to capture long-range dependencies while maintaining precise local feature representation through its hierarchical processing pipeline. By integrating Vision Mamba (VMamba)'s efficient sequence modeling with RMA's targeted feature refinement, our architecture achieves superior feature learning across multiple scales. This dual-mechanism approach enables robust handling of complex morphological patterns while maintaining computational efficiency. We demonstrate RMA-Mamba's effectiveness in the challenging domain of pathological liver segmentation (from both CT and MRI), where traditional segmentation approaches often fail due to tissue variations. When evaluated on a newly introduced cirrhotic liver dataset (CirrMRI600+) of T2-weighted MRI scans, RMA-Mamba achieves the state-of-the-art performance with a Dice coefficient of 92.08%, mean IoU of 87.36%, and recall of 92.96%. The architecture's generalizability is further validated on the cancerous liver segmentation from CT scans (LiTS: Liver Tumor Segmentation dataset), yielding a Dice score of 92.9% and mIoU of 88.99%. The source code of the proposed RMA-Mamba is available at https://github.com/JunZengz/RMAMamba.
Authors: Taneesh Gupta, Rahul Madhavan, Xuchao Zhang, Chetan Bansal, Saravan Rajmohan
Abstract: Multi-preference optimization enriches language-model alignment beyond pairwise preferences by contrasting entire sets of helpful and undesired responses, thereby enabling richer training signals for large language models. During self-play alignment, these models often produce numerous candidate answers per query, rendering it computationally infeasible to include all responses in the training objective. In this work, we propose $\textit{Active Multi-Preference Optimization}$ (AMPO), a novel approach that combines on-policy generation, a multi-preference group-contrastive loss, and active subset selection. Specifically, we score and embed large candidate pools of responses and then select a small, yet informative, subset that covers reward extremes and distinct semantic clusters for preference optimization. Our contrastive training scheme is capable of identifying not only the best and worst answers but also subtle, underexplored modes that are crucial for robust alignment. Theoretically, we provide guarantees for expected reward maximization using our active selection method, and empirically, AMPO achieves state-of-the-art results on $\textit{AlpacaEval}$ using Llama 8B.
Authors: James C. A. Main, Mickael Randour
Abstract: We consider multi-dimensional payoff functions in Markov decision processes, and ask whether a given expected payoff vector can be achieved or not. In general, pure strategies (i.e., not resorting to randomisation) do not suffice for this problem. We study the structure of the set of expected payoff vectors of all strategies given a multi-dimensional payoff function and its consequences regarding randomisation requirements for strategies. In particular, we prove that for any payoff for which the expectation is well-defined under all strategies, it is sufficient to mix (i.e., randomly select a pure strategy at the start of a play and committing to it for the rest of the play) finitely many pure strategies to approximate any expected payoff vector up to any precision. Furthermore, for any payoff for which the expected payoff is finite under all strategies, any expected payoff can be obtained exactly by mixing finitely many strategies.
Authors: Taha Ahmadi Pargo, Mohsen Akbarpour Shirazi, Dawud Fadai
Abstract: Regarding problems like reduced precipitation and an increase in population, water resource scarcity has become one of the most critical problems in modern-day societies, as a consequence, there is a shortage of available water resources for irrigation in arid and semi-arid countries. On the other hand, it is possible to utilize modern technologies to control irrigation and reduce water loss. One of these technologies is the Internet of Things (IoT). Despite the possibility of using the IoT in irrigation control systems, there are complexities in designing such systems. Considering this issue, it is possible to use agent-oriented software engineering (AOSE) methodologies to design complex cyber-physical systems such as IoT-based systems. In this research, a smart irrigation system is designed based on Prometheus AOSE methodology, to reduce water loss by maintaining soil moisture in a suitable interval. The designed system comprises sensors, a central agent, and irrigation nodes. These agents follow defined rules to maintain soil moisture at a desired level cooperatively. For system simulation, a hybrid agent-based and system dynamics model was designed. In this hybrid model, soil moisture dynamics were modeled based on the system dynamics approach. The proposed model, was implemented in AnyLogic computer simulation software. Utilizing the simulation model, irrigation rules were examined. The system's functionality in automatic irrigation mode was tested based on a 256-run, fractional factorial design, and the effects of important factors such as soil properties on total irrigated water and total operation time were analyzed. Based on the tests, the system consistently irrigated nearly optimal water amounts in all tests. Moreover, the results were also used to minimize the system's energy consumption by reducing the system's operational time.
Authors: Manuel Barusco, Francesco Borsatti, Davide Dalle Pezze, Francesco Paissan, Elisabetta Farella, Gian Antonio Susto
Abstract: Recent advances in Visual Anomaly Detection (VAD) have introduced sophisticated algorithms leveraging embeddings generated by pre-trained feature extractors. Inspired by these developments, we investigate the adaptation of such algorithms to the audio domain to address the problem of Audio Anomaly Detection (AAD). Unlike most existing AAD methods, which primarily classify anomalous samples, our approach introduces fine-grained temporal-frequency localization of anomalies within the spectrogram, significantly improving explainability. This capability enables a more precise understanding of where and when anomalies occur, making the results more actionable for end users. We evaluate our approach on industrial and environmental benchmarks, demonstrating the effectiveness of VAD techniques in detecting anomalies in audio signals. Moreover, they improve explainability by enabling localized anomaly identification, making audio anomaly detection systems more interpretable and practical.
Authors: Jessica He, Stephanie Houde, Justin D. Weisz
Abstract: AI systems powered by large language models can act as capable assistants for writing and editing. In these tasks, the AI system acts as a co-creative partner, making novel contributions to an artifact-under-creation alongside its human partner(s). One question that arises in these scenarios is the extent to which AI should be credited for its contributions. We examined knowledge workers' views of attribution through a survey study (N=155) and found that they assigned different levels of credit across different contribution types, amounts, and initiative. Compared to a human partner, we observed a consistent pattern in which AI was assigned less credit for equivalent contributions. Participants felt that disclosing AI involvement was important and used a variety of criteria to make attribution judgments, including the quality of contributions, personal values, and technology considerations. Our results motivate and inform new approaches for crediting AI contributions to co-created work.
Authors: Dominik Hollidt, Paul Streli, Jiaxi Jiang, Yasaman Haghighi, Changlin Qian, Xintong Liu, Christian Holz
Abstract: Research on egocentric tasks in computer vision has mostly focused on head-mounted cameras, such as fisheye cameras or embedded cameras inside immersive headsets. We argue that the increasing miniaturization of optical sensors will lead to the prolific integration of cameras into many more body-worn devices at various locations. This will bring fresh perspectives to established tasks in computer vision and benefit key areas such as human motion tracking, body pose estimation, or action recognition -- particularly for the lower body, which is typically occluded. In this paper, we introduce EgoSim, a novel simulator of body-worn cameras that generates realistic egocentric renderings from multiple perspectives across a wearer's body. A key feature of EgoSim is its use of real motion capture data to render motion artifacts, which are especially noticeable with arm- or leg-worn cameras. In addition, we introduce MultiEgoView, a dataset of egocentric footage from six body-worn cameras and ground-truth full-body 3D poses during several activities: 119 hours of data are derived from AMASS motion sequences in four high-fidelity virtual environments, which we augment with 5 hours of real-world motion data from 13 participants using six GoPro cameras and 3D body pose references from an Xsens motion capture suit. We demonstrate EgoSim's effectiveness by training an end-to-end video-only 3D pose estimation network. Analyzing its domain gap, we show that our dataset and simulator substantially aid training for inference on real-world data. EgoSim code & MultiEgoView dataset: https://siplab.org/projects/EgoSim
Authors: Yu Xia, Jingru Fan, Weize Chen, Siyu Yan, Xin Cong, Zhong Zhang, Yaxi Lu, Yankai Lin, Zhiyuan Liu, Maosong Sun
Abstract: Existing LLM-based agents have achieved strong performance on held-in tasks, but their generalizability to unseen tasks remains poor. Hence, some recent work focus on fine-tuning the policy model with more diverse tasks to improve the generalizability. In this work, we find that finetuning a reward model to guide the policy model is more robust than directly finetuning the policy model. Based on this finding, we propose AgentRM, a generalizable reward model, to guide the policy model for effective test-time search. We comprehensively investigate three approaches to construct the reward model, including explicit reward modeling, implicit reward modeling and LLM-as-a-judge. We then use AgentRM to guide the answer generation with Best-of-N sampling and step-level beam search. On four types of nine agent tasks, AgentRM enhances the base policy model by $8.8$ points on average, surpassing the top general agent by $4.0$. Moreover, it demonstrates weak-to-strong generalization, yielding greater improvement of $12.6$ on LLaMA-3-70B policy model. As for the specializability, AgentRM can also boost a finetuned policy model and outperform the top specialized agent by $11.4$ on three held-in tasks. Further analysis verifies its effectiveness in test-time scaling. Codes will be released to facilitate the research in this area.
Authors: Young-Chae Hong, Bei Xiao, Yangho Chen
Abstract: Time series forecasting has long been a focus of research across diverse fields, including economics, energy, healthcare, and traffic management. Recent works have introduced innovative architectures for time series models, such as the Time-Series Mixer (TSMixer), which leverages multi-layer perceptrons (MLPs) to enhance prediction accuracy by effectively capturing both spatial and temporal dependencies within the data. In this paper, we investigate the capabilities of the Kolmogorov-Arnold Networks (KANs) for time-series forecasting by modifying TSMixer with a KAN layer (TSKANMixer). Experimental results demonstrate that TSKANMixer tends to improve prediction accuracy over the original TSMixer across multiple datasets, ranking among the top-performing models compared to other time series approaches. Our results show that the KANs are promising alternatives to improve the performance of time series forecasting by replacing or extending traditional MLPs.
Authors: Paula Santos
Abstract: This study presents a comparative evaluation of a Variational Autoencoder (VAE) enhanced with Minimum Description Length (MDL) regularization against a Standard Autoencoder for reconstructing high-dimensional gynecological data. The MDL-VAE exhibits significantly lower reconstruction errors (MSE, MAE, RMSE) and more structured latent representations, driven by effective KL divergence regularization. Statistical analyses confirm these performance improvements are significant. Furthermore, the MDL-VAE shows consistent training and validation losses and achieves efficient inference times, underscoring its robustness and practical viability. Our findings suggest that incorporating MDL principles into VAE architectures can substantially improve data reconstruction and generalization, making it a promising approach for advanced applications in healthcare data modeling and analysis.
Authors: Frederikus Hudi, Genta Indra Winata, Ruochen Zhang, Alham Fikri Aji
Abstract: Reasoning is a fundamental capability of large language models (LLMs), enabling them to comprehend, analyze, and solve complex problems. In this paper, we introduce TextGames, an innovative benchmark specifically crafted to assess LLMs through demanding text-based games that require advanced skills in pattern recognition, spatial awareness, arithmetic, and logical reasoning. Our analysis probes LLMs' performance in both single-turn and multi-turn reasoning, and their abilities in leveraging feedback to correct subsequent answers through self-reflection. Our findings reveal that, although LLMs exhibit proficiency in addressing most easy and medium-level problems, they face significant challenges with more difficult tasks. In contrast, humans are capable of solving all tasks when given sufficient time. Moreover, we observe that LLMs show improved performance in multi-turn predictions through self-reflection, yet they still struggle with sequencing, counting, and following complex rules consistently. Additionally, models optimized for reasoning outperform pre-trained LLMs that prioritize instruction following, highlighting the crucial role of reasoning skills in addressing highly complex problems.
Authors: Pedro Sequeira, Vidyasagar Sadhu, Melinda Gervasio
Abstract: In this paper we present ToMCAT (Theory-of-Mind for Cooperative Agents in Teams), a new framework for generating ToM-conditioned trajectories. It combines a meta-learning mechanism, that performs ToM reasoning over teammates' underlying goals and future behavior, with a multiagent denoising-diffusion model, that generates plans for an agent and its teammates conditioned on both the agent's goals and its teammates' characteristics, as computed via ToM. We implemented an online planning system that dynamically samples new trajectories (replans) from the diffusion model whenever it detects a divergence between a previously generated plan and the current state of the world. We conducted several experiments using ToMCAT in a simulated cooking domain. Our results highlight the importance of the dynamic replanning mechanism in reducing the usage of resources without sacrificing team performance. We also show that recent observations about the world and teammates' behavior collected by an agent over the course of an episode combined with ToM inferences are crucial to generate team-aware plans for dynamic adaptation to teammates, especially when no prior information is provided about them.
Authors: Irina Saparina, Mirella Lapata
Abstract: Handling ambiguity and underspecification is an important challenge in natural language interfaces, particularly for tasks like text-to-SQL semantic parsing. We propose a modular approach that resolves ambiguity using natural language interpretations before mapping these to logical forms (e.g., SQL queries). Although LLMs excel at parsing unambiguous utterances, they show strong biases for ambiguous ones, typically predicting only preferred interpretations. We constructively exploit this bias to generate an initial set of preferred disambiguations and then apply a specialized infilling model to identify and generate missing interpretations. To train the infilling model, we introduce an annotation method that uses SQL execution to validate different meanings. Our approach improves interpretation coverage and generalizes across datasets with different annotation styles, database structures, and ambiguity types.
Authors: Yuxiang Wei, Olivier Duchenne, Jade Copet, Quentin Carbonneaux, Lingming Zhang, Daniel Fried, Gabriel Synnaeve, Rishabh Singh, Sida I. Wang
Abstract: The recent DeepSeek-R1 release has demonstrated the immense potential of reinforcement learning (RL) in enhancing the general reasoning capabilities of large language models (LLMs). While DeepSeek-R1 and other follow-up work primarily focus on applying RL to competitive coding and math problems, this paper introduces SWE-RL, the first approach to scale RL-based LLM reasoning for real-world software engineering. Leveraging a lightweight rule-based reward (e.g., the similarity score between ground-truth and LLM-generated solutions), SWE-RL enables LLMs to autonomously recover a developer's reasoning processes and solutions by learning from extensive open-source software evolution data -- the record of a software's entire lifecycle, including its code snapshots, code changes, and events such as issues and pull requests. Trained on top of Llama 3, our resulting reasoning model, Llama3-SWE-RL-70B, achieves a 41.0% solve rate on SWE-bench Verified -- a human-verified collection of real-world GitHub issues. To our knowledge, this is the best performance reported for medium-sized (<100B) LLMs to date, even comparable to leading proprietary LLMs like GPT-4o. Surprisingly, despite performing RL solely on software evolution data, Llama3-SWE-RL has even emerged with generalized reasoning skills. For example, it shows improved results on five out-of-domain tasks, namely, function coding, library use, code reasoning, mathematics, and general language understanding, whereas a supervised-finetuning baseline even leads to performance degradation on average. Overall, SWE-RL opens up a new direction to improve the reasoning capabilities of LLMs through reinforcement learning on massive software engineering data.
Authors: Mollie Shichman, Claire Bonial, Austin Blodgett, Taylor Hudson, Francis Ferraro, Rachel Rudinger
Abstract: Large Language Models (LLMs) have the potential for substantial common sense reasoning. However, these capabilities are often emergent in larger models. This means smaller models that can be run locally are less helpful and capable with respect to certain reasoning tasks. To meet our problem space requirements, we fine-tune smaller LLMs to disaster domains, as these domains involve complex and low-frequency physical common sense knowledge. We introduce a pipeline to create Field Ready Instruction Decoding Agent (FRIDA) models, where domain experts and linguists combine their knowledge to make high-quality seed data that is used to generate synthetic data for fine-tuning. We create a set of 130 seed instructions for synthetic generation, a synthetic dataset of 25000 instructions, and 119 evaluation instructions relating to both general and earthquake-specific object affordances. We fine-tune several LLaMa and Mistral instruction-tuned models and find that FRIDA models outperform their base models at a variety of sizes. We then run an ablation study to understand which kinds of synthetic data most affect performance and find that training physical state and object function common sense knowledge alone improves over FRIDA models trained on all data. We conclude that the FRIDA pipeline is capable of instilling general common sense, but needs to be augmented with information retrieval for specific domain knowledge.
Authors: Charlie B. Tan, Avishek Joey Bose, Chen Lin, Leon Klein, Michael M. Bronstein, Alexander Tong
Abstract: Scalable sampling of molecular states in thermodynamic equilibrium is a long-standing challenge in statistical physics. Boltzmann generators tackle this problem by pairing powerful normalizing flows with importance sampling to obtain statistically independent samples under the target distribution. In this paper, we extend the Boltzmann generator framework and introduce Sequential Boltzmann generators (SBG) with two key improvements. The first is a highly efficient non-equivariant Transformer-based normalizing flow operating directly on all-atom Cartesian coordinates. In contrast to equivariant continuous flows of prior methods, we leverage exactly invertible non-equivariant architectures which are highly efficient both during sample generation and likelihood computation. As a result, this unlocks more sophisticated inference strategies beyond standard importance sampling. More precisely, as a second key improvement we perform inference-time scaling of flow samples using annealed Langevin dynamics which transports samples toward the target distribution leading to lower variance (annealed) importance weights which enable higher fidelity resampling with sequential Monte Carlo. SBG achieves state-of-the-art performance w.r.t. all metrics on molecular systems, demonstrating the first equilibrium sampling in Cartesian coordinates of tri, tetra, and hexapeptides that were so far intractable for prior Boltzmann generators.
Authors: Dimiter Dobrev
Abstract: We will consider all policies of the agent and will prove that one of them is the best performing policy. While that policy is not computable, computable policies do exist in its proximity. We will define AI as a computable policy which is sufficiently proximal to the best performing policy. Before we can define the agent's best performing policy, we need a language for description of the world. We will also use this language to develop a program which satisfies the AI definition. The program will first understand the world by describing it in the selected language. The program will then use the description in order to predict the future and select the best possible move. While this program is extremely inefficient and practically unusable, it can be improved by refining both the language for description of the world and the algorithm used to predict the future. This can yield a program which is both efficient and consistent with the AI definition.
Authors: Tiansi Dong, Mateja Jamnik, Pietro Li\`o
Abstract: The success of Large Language Models (LLMs), e.g., ChatGPT, is witnessed by their planetary popularity, their capability of human-like communication, and also by their steadily improved reasoning performance. However, it remains unclear whether LLMs reason. It is an open problem how traditional neural networks can be qualitatively extended to go beyond the statistic paradigm and achieve high-level cognition. Here, we present a novel qualitative extension by generalising computational building blocks from vectors to spheres. We propose Sphere Neural Networks (SphNNs) for human-like reasoning through model construction and inspection, and develop SphNN for syllogistic reasoning, a microcosm of human rationality. SphNN is a hierarchical neuro-symbolic Kolmogorov-Arnold geometric GNN, and uses a neuro-symbolic transition map of neighbourhood spatial relations to transform the current sphere configuration towards the target. SphNN is the first neural model that can determine the validity of long-chained syllogistic reasoning in one epoch without training data, with the worst computational complexity of O(N). SphNN can evolve into various types of reasoning, such as spatio-temporal reasoning, logical reasoning with negation and disjunction, event reasoning, neuro-symbolic unification, and humour understanding (the highest level of cognition). All these suggest a new kind of Herbert A. Simon's scissors with two neural blades. SphNNs will tremendously enhance interdisciplinary collaborations to develop the two neural blades and realise deterministic neural reasoning and human-bounded rationality and elevate LLMs to reliable psychological AI. This work suggests that the non-zero radii of spheres are the missing components that prevent traditional deep-learning systems from reaching the realm of rational reasoning and cause LLMs to be trapped in the swamp of hallucination.
Authors: Carlos Penarrubia, Carlos Garrido-Munoz, Jose J. Valero-Mas, Jorge Calvo-Zaragoza
Abstract: Handwritten Text Recognition (HTR) is a relevant problem in computer vision, and implies unique challenges owing to its inherent variability and the rich contextualization required for its interpretation. Despite the success of Self-Supervised Learning (SSL) in computer vision, its application to HTR has been rather scattered, leaving key SSL methodologies unexplored. This work focuses on one of them, namely Spatial Context-based SSL. We investigate how this family of approaches can be adapted and optimized for HTR and propose new workflows that leverage the unique features of handwritten text. Our experiments demonstrate that the methods considered lead to advancements in the state-of-the-art of SSL for HTR in a number of benchmark cases.
Authors: Robbie Holland (on behalf of the PINNACLE consortium), Thomas R. P. Taylor (on behalf of the PINNACLE consortium), Christopher Holmes (on behalf of the PINNACLE consortium), Sophie Riedl (on behalf of the PINNACLE consortium), Julia Mai (on behalf of the PINNACLE consortium), Maria Patsiamanidi (on behalf of the PINNACLE consortium), Dimitra Mitsopoulou (on behalf of the PINNACLE consortium), Paul Hager (on behalf of the PINNACLE consortium), Philip M\"uller (on behalf of the PINNACLE consortium), Hendrik P. N. Scholl (on behalf of the PINNACLE consortium), Hrvoje Bogunovi\'c (on behalf of the PINNACLE consortium), Ursula Schmidt-Erfurth (on behalf of the PINNACLE consortium), Daniel Rueckert (on behalf of the PINNACLE consortium), Sobha Sivaprasad (on behalf of the PINNACLE consortium), Andrew J. Lotery (on behalf of the PINNACLE consortium), Martin J. Menten (on behalf of the PINNACLE consortium)
Abstract: Clinicians spend a significant amount of time reviewing medical images and transcribing their findings regarding patient diagnosis, referral and treatment in text form. Vision-language models (VLMs), which automatically interpret images and summarize their findings as text, have enormous potential to alleviate clinical workloads and increase patient access to high-quality medical care. While foundational models have stirred considerable interest in the medical community, it is unclear whether their general capabilities translate to real-world clinical utility. In this work, we demonstrate that OpenAI's ChatGPT-4o model, in addition to two foundation VLMs designed for medical use, markedly underperform compared to practicing ophthalmologists on specialist tasks crucial to the care of patients with age-related macular degeneration (AMD). To address this, we initially identified the essential capabilities required for image-based clinical decision-making, and then developed a curriculum to selectively train VLMs in these skills. The resulting model, RetinaVLM, can be instructed to write reports that significantly outperform those written by leading foundation medical VLMs and ChatGPT-4o in disease staging (F1 score of 0.63 vs. 0.33) and patient referral (0.67 vs. 0.50), and approaches the diagnostic performance of junior ophthalmologists (who achieve 0.77 and 0.78 on the respective tasks). Furthermore, in a single-blind reader study two senior ophthalmologists with up to 32 years of experience found RetinaVLM's reports were found to be substantially more accurate than those by ChatGPT-4o (64.3% vs. 14.3%). These results reinforce that our curriculum-based approach provides a blueprint towards specializing foundation medical VLMs for real-world clinical tasks.
Authors: Simon Dima, Simon Fischer, Jobst Heitzig, Joss Oliver
Abstract: In dynamic programming and reinforcement learning, the policy for the sequential decision making of an agent in a stochastic environment is usually determined by expressing the goal as a scalar reward function and seeking a policy that maximizes the expected total reward. However, many goals that humans care about naturally concern multiple aspects of the world, and it may not be obvious how to condense those into a single reward function. Furthermore, maximization suffers from specification gaming, where the obtained policy achieves a high expected total reward in an unintended way, often taking extreme or nonsensical actions. Here we consider finite acyclic Markov Decision Processes with multiple distinct evaluation metrics, which do not necessarily represent quantities that the user wants to be maximized. We assume the task of the agent is to ensure that the vector of expected totals of the evaluation metrics falls into some given convex set, called the aspiration set. Our algorithm guarantees that this task is fulfilled by using simplices to approximate feasibility sets and propagate aspirations forward while ensuring they remain feasible. It has complexity linear in the number of possible state-action-successor triples and polynomial in the number of evaluation metrics. Moreover, the explicitly non-maximizing nature of the chosen policy and goals yields additional degrees of freedom, which can be used to apply heuristic safety criteria to the choice of actions. We discuss several such safety criteria that aim to steer the agent towards more conservative behavior.
Authors: Yibo Li, Yuan Fang, Mengmei Zhang, Chuan Shi
Abstract: Understanding molecular structure and related knowledge is crucial for scientific research. Recent studies integrate molecular graphs with their textual descriptions to enhance molecular representation learning. However, they focus on the whole molecular graph and neglect frequently occurring subgraphs, known as motifs, which are essential for determining molecular properties. Without such fine-grained knowledge, these models struggle to generalize to unseen molecules and tasks that require motif-level insights. To bridge this gap, we propose FineMolTex, a novel Fine-grained Molecular graph-Text pre-training framework to jointly learn coarse-grained molecule-level knowledge and fine-grained motif-level knowledge. Specifically, FineMolTex consists of two pre-training tasks: a contrastive alignment task for coarse-grained matching and a masked multi-modal modeling task for fine-grained matching. In particular, the latter predicts the labels of masked motifs and words, which are selected based on their importance. By leveraging insights from both modalities, FineMolTex is able to understand the fine-grained matching between motifs and words. Finally, we conduct extensive experiments across three downstream tasks, achieving up to 238% improvement in the text-based molecule editing task. Additionally, our case studies reveal that FineMolTex successfully captures fine-grained knowledge, potentially offering valuable insights for drug discovery and catalyst design.
Authors: Jiale Fu, Yaqing Wang, Simeng Han, Jiaming Fan, Xu Yang
Abstract: In-context learning (ICL) enhances large language models (LLMs) by incorporating demonstration examples, yet its effectiveness heavily depends on the quality of selected examples. Current methods typically use text embeddings to measure semantic similarity, which often introduces bias in multi-step reasoning tasks. This occurs because text embeddings contain irrelevant semantic information and lack deeper reasoning structures. To address this, we propose GraphIC, a graph-based retrieval model that leverages reasoning-aware representation and specialized similarity metric for in-context example retrieval. GraphIC first constructs thought graphs-directed, node-attributed graphs that explicitly model reasoning steps and their dependencies-for candidate examples and queries. This approach filters out superficial semantics while preserving essential reasoning processes. Next, GraphIC retrieves examples using a novel similarity metric tailored for these graphs, capturing sequential reasoning patterns and asymmetry between examples. Comprehensive evaluations across mathematical reasoning, code generation, and logical reasoning tasks demonstrate that GraphIC outperforms 10 baseline methods. Our results highlight the importance of reasoning-aware retrieval in ICL, offering a robust solution for enhancing LLM performance in multi-step reasoning scenarios.
Authors: Jiayi Zhang, Jinyu Xiang, Zhaoyang Yu, Fengwei Teng, Xionghui Chen, Jiaqi Chen, Mingchen Zhuge, Xin Cheng, Sirui Hong, Jinlin Wang, Bingnan Zheng, Bang Liu, Yuyu Luo, Chenglin Wu
Abstract: Large language models (LLMs) have demonstrated remarkable potential in solving complex tasks across diverse domains, typically by employing agentic workflows that follow detailed instructions and operational sequences. However, constructing these workflows requires significant human effort, limiting scalability and generalizability. Recent research has sought to automate the generation and optimization of these workflows, but existing methods still rely on initial manual setup and fall short of achieving fully automated and effective workflow generation. To address this challenge, we reformulate workflow optimization as a search problem over code-represented workflows, where LLM-invoking nodes are connected by edges. We introduce AFlow, an automated framework that efficiently explores this space using Monte Carlo Tree Search, iteratively refining workflows through code modification, tree-structured experience, and execution feedback. Empirical evaluations across six benchmark datasets demonstrate AFlow's efficacy, yielding a 5.7% average improvement over state-of-the-art baselines. Furthermore, AFlow enables smaller models to outperform GPT-4o on specific tasks at 4.55% of its inference cost in dollars. The code will be available at https://github.com/geekan/MetaGPT.
Authors: Antonia W\"ust, Tim Tobiasch, Lukas Helff, Inga Ibs, Wolfgang Stammer, Devendra S. Dhami, Constantin A. Rothkopf, Kristian Kersting
Abstract: Recently, newly developed Vision-Language Models (VLMs), such as OpenAI's o1, have emerged, seemingly demonstrating advanced reasoning capabilities across text and image modalities. However, the depth of these advances in language-guided perception and abstract reasoning remains underexplored, and it is unclear whether these models can truly live up to their ambitious promises. To assess the progress and identify shortcomings, we enter the wonderland of Bongard problems, a set of classic visual reasoning puzzles that require human-like abilities of pattern recognition and abstract reasoning. With our extensive evaluation setup, we show that while VLMs occasionally succeed in identifying discriminative concepts and solving some of the problems, they frequently falter. Surprisingly, even elementary concepts that may seem trivial to humans, such as simple spirals, pose significant challenges. Moreover, when explicitly asked to recognize ground truth concepts, they continue to falter, suggesting not only a lack of understanding of these elementary visual concepts but also an inability to generalize to unseen concepts. We compare the results of VLMs to human performance and observe that a significant gap remains between human visual reasoning capabilities and machine cognition.
Authors: Gabriel Maher
Abstract: Recent advancements in prompting techniques for Large Language Models (LLMs) have improved their reasoning, planning, and action abilities. This paper examines these prompting techniques through the lens of model predictive control (MPC). We show that LLMs act as implicit planning cost function minimizers when planning prompts are used. We propose a unified MPC framework for planning with LLMs and demonstrate improved performance over few shot prompting on several planning benchmarks.
Authors: Felix Hofst\"atter, Teun van der Weij, Jayden Teoh, Henning Bartsch, Francis Rhys Ward
Abstract: Capability evaluations are required to understand and regulate AI systems that may be deployed or further developed. Therefore, it is important that evaluations provide an accurate estimation of an AI system's capabilities. However, in numerous cases, previously latent capabilities have been elicited from models, sometimes long after initial release. Accordingly, substantial efforts have been made to develop methods for eliciting latent capabilities from models. In this paper, we evaluate the effectiveness of capability elicitation techniques by intentionally training model organisms -- language models with hidden capabilities that are revealed by a password. We introduce a novel method for training model organisms, based on circuit breaking, which is more robust to elicitation techniques than standard password-locked models. We focus on elicitation techniques based on prompting and activation steering, and compare these to fine-tuning methods. Prompting techniques can elicit the actual capability of both password-locked and circuit-broken model organisms in an MCQA setting, while steering fails to do so. For a code-generation task, only fine-tuning can elicit the hidden capabilities of our novel model organism. Additionally, our results suggest that combining techniques improves elicitation. Still, if possible, fine-tuning should be the method of choice to improve the trustworthiness of capability evaluations.
Authors: Libo Wang
Abstract: To reduce the cost and consumption of computing resources caused by computational redundancy and delayed reward assignment in long CoT, this research proposes the dynamic chain-of-thought (D-CoT) with adaptive reasoning time and steps. The researcher used simulation experiment to simulate the integration of D-CoT through Python 3.13 IDLE combined with a Python simulator based on GPTs. At the same time, the researcher used DeepSeek R1 as a control group to test and compare the performance of the D-CoT simulator in processing MIT OpenCourseWare's linear algebra exam questions. Experimental results show that D-CoT is better than DeepSeek R1 based on long CoT in three indicators: reasoning time, CoT length (reasoning steps) and token count, which achieves a significant reduction in computing resource consumption. In addition, this research has potential value in deep reasoning optimization that is used as a reference for future dynamic deep reasoning frameworks.
Authors: Kaikai Zhao, Zhaoxiang Liu, Xuejiao Lei, Ning Wang, Zhenhong Long, Jiaojiao Zhao, Zipeng Wang, Peijun Yang, Minjie Hua, Chaoyang Ma, Wen Liu, Kai Wang, Shiguo Lian
Abstract: DeepSeek-R1, known for its low training cost and exceptional reasoning capabilities, has achieved state-of-the-art performance on various benchmarks. However, detailed evaluations from the perspective of real-world applications are lacking, making it challenging for users to select the most suitable DeepSeek models for their specific needs. To address this gap, we evaluate the DeepSeek-V3, DeepSeek-R1, DeepSeek-R1-Distill-Qwen series, DeepSeek-R1-Distill-Llama series, and their corresponding 4-bit quantized models on the enhanced A-Eval benchmark, A-Eval-2.0. By comparing original instruction-tuned models with their distilled counterparts, we analyze how reasoning enhancements impact performance across diverse practical tasks. Our results show that reasoning-enhanced models, while generally powerful, do not universally outperform across all tasks, with performance gains varying significantly across tasks and models. To further assist users in model selection, we quantify the capability boundary of DeepSeek models through performance tier classifications and intuitive line charts. Specific examples provide actionable insights to help users select and deploy the most cost-effective DeepSeek models, ensuring optimal performance and resource efficiency in real-world applications. It should be noted that, despite our efforts to establish a comprehensive, objective, and authoritative evaluation benchmark, the selection of test samples, characteristics of data distribution, and the setting of evaluation criteria may inevitably introduce certain biases into the evaluation results. We will continuously optimize the evaluation benchmarks and periodically update this paper to provide more comprehensive and accurate evaluation results. Please refer to the latest version of the paper for the most recent results and conclusions.
Authors: Zenan Li, Zhaoyu Li, Wen Tang, Xian Zhang, Yuan Yao, Xujie Si, Fan Yang, Kaiyu Yang, Xiaoxing Ma
Abstract: Large language models (LLMs) can prove mathematical theorems formally by generating proof steps (\textit{a.k.a.} tactics) within a proof system. However, the space of possible tactics is vast and complex, while the available training data for formal proofs is limited, posing a significant challenge to LLM-based tactic generation. To address this, we introduce a neuro-symbolic tactic generator that synergizes the mathematical intuition learned by LLMs with domain-specific insights encoded by symbolic methods. The key aspect of this integration is identifying which parts of mathematical reasoning are best suited to LLMs and which to symbolic methods. While the high-level idea of neuro-symbolic integration is broadly applicable to various mathematical problems, in this paper, we focus specifically on Olympiad inequalities (Figure~1). We analyze how humans solve these problems and distill the techniques into two types of tactics: (1) scaling, handled by symbolic methods, and (2) rewriting, handled by LLMs. In addition, we combine symbolic tools with LLMs to prune and rank the proof goals for efficient proof search. We evaluate our framework on 161 challenging inequalities from multiple mathematics competitions, achieving state-of-the-art performance and significantly outperforming existing LLM and symbolic approaches without requiring additional training data.
Authors: Fengxiang Cheng, Haoxuan Li, Fenrong Liu, Robert van Rooij, Kun Zhang, Zhouchen Lin
Abstract: Large language models (LLMs) have achieved remarkable successes on various natural language tasks. However, recent studies have found that there are still significant challenges to the logical reasoning abilities of LLMs. This paper summarizes and categorizes the main challenges into two aspects: (1) Logical question answering, LLMs often fail to generate the correct answer within complex logical problem which requires sophisticated deductive, inductive or abductive reasoning given a collection of premises and constrains. (2) Logical consistency, LLMs are prone to producing responses contradicting themselves across different questions. For example, a state-of-the-art Macaw question-answering LLM answers Yes to both questions Is a magpie a bird? and Does a bird have wings? but answers No to Does a magpie have wings?. To facilitate this research direction, we comprehensively investigate the most cutting-edge methods and propose detailed taxonomies of these methods. Specifically, to accurately answer complex logic questions, previous methods can be categorized based on reliance on external solvers, prompts, pretraining, and fine-tuning. To avoid logical contradictions, we discuss concepts and solutions of various logical consistencies, including implication, negation, transitivity, factuality consistency, and their composites. In addition, we review commonly used benchmark datasets and evaluation metrics, and discuss promising research directions, such as extensions to modal logic to account for uncertainty, and efficient algorithms satisfying multiple logical consistencies simultaneously.
Authors: Zhong-Zhi Li, Duzhen Zhang, Ming-Liang Zhang, Jiaxin Zhang, Zengyan Liu, Yuxuan Yao, Haotian Xu, Junhao Zheng, Pei-Jie Wang, Xiuyi Chen, Yingying Zhang, Fei Yin, Jiahua Dong, Zhijiang Guo, Le Song, Cheng-Lin Liu
Abstract: Achieving human-level intelligence requires refining the transition from the fast, intuitive System 1 to the slower, more deliberate System 2 reasoning. While System 1 excels in quick, heuristic decisions, System 2 relies on logical reasoning for more accurate judgments and reduced biases. Foundational Large Language Models (LLMs) excel at fast decision-making but lack the depth for complex reasoning, as they have not yet fully embraced the step-by-step analysis characteristic of true System 2 thinking. Recently, reasoning LLMs like OpenAI's o1/o3 and DeepSeek's R1 have demonstrated expert-level performance in fields such as mathematics and coding, closely mimicking the deliberate reasoning of System 2 and showcasing human-like cognitive abilities. This survey begins with a brief overview of the progress in foundational LLMs and the early development of System 2 technologies, exploring how their combination has paved the way for reasoning LLMs. Next, we discuss how to construct reasoning LLMs, analyzing their features, the core methods enabling advanced reasoning, and the evolution of various reasoning LLMs. Additionally, we provide an overview of reasoning benchmarks, offering an in-depth comparison of the performance of representative reasoning LLMs. Finally, we explore promising directions for advancing reasoning LLMs and maintain a real-time \href{https://github.com/zzli2022/Awesome-Slow-Reason-System}{GitHub Repository} to track the latest developments. We hope this survey will serve as a valuable resource to inspire innovation and drive progress in this rapidly evolving field.
URLs: https://github.com/zzli2022/Awesome-Slow-Reason-System
Authors: Yinqian Sun, Feifei Zhao, Zhuoya Zhao, Yi Zeng
Abstract: Inspired by the brain's information processing using binary spikes, spiking neural networks (SNNs) offer significant reductions in energy consumption and are more adept at incorporating multi-scale biological characteristics. In SNNs, spiking neurons serve as the fundamental information processing units. However, in most models, these neurons are typically simplified, focusing primarily on the leaky integrate-and-fire (LIF) point neuron model while neglecting the structural properties of biological neurons. This simplification hampers the computational and learning capabilities of SNNs. In this paper, we propose a brain-inspired deep distributional reinforcement learning algorithm based on SNNs, which integrates a bio-inspired multi-compartment neuron (MCN) model with a population coding approach. The proposed MCN model simulates the structure and function of apical dendritic, basal dendritic, and somatic compartments, achieving computational power comparable to that of biological neurons. Additionally, we introduce an implicit fractional embedding method based on population coding of spiking neurons. We evaluated our model on Atari games, and the experimental results demonstrate that it surpasses the vanilla FQF model, which utilizes traditional artificial neural networks (ANNs), as well as the Spiking-FQF models that are based on ANN-to-SNN conversion methods. Ablation studies further reveal that the proposed multi-compartment neuron model and the quantile fraction implicit population spike representation significantly enhance the performance of MCS-FQF while also reducing power consumption.
Authors: Abhishek Ajayakumar, Soumyendu Raha
Abstract: Many real-world systems are modelled using complex ordinary differential equations (ODEs). However, the dimensionality of these systems can make them challenging to analyze. Dimensionality reduction techniques like Proper Orthogonal Decomposition (POD) can be used in such cases. However, these reduced order models are susceptible to perturbations in the input. We propose a novel framework that combines machine learning and data assimilation techniques to improving surrogate models to handle perturbations in input data effectively. Through rigorous experiments on dynamical systems modelled on graphs, we demonstrate that our framework substantially improves the accuracy of surrogate models under input perturbations. Furthermore, we evaluate the framework's efficacy on alternative surrogate models, including neural ODEs, and the empirical results consistently show enhanced performance.
Authors: Yanjun Gao, Ruizhe Li, Emma Croxford, John Caskey, Brian W Patterson, Matthew Churpek, Timothy Miller, Dmitriy Dligach, Majid Afshar
Abstract: Electronic Health Records (EHRs) and routine documentation practices play a vital role in patients' daily care, providing a holistic record of health, diagnoses, and treatment. However, complex and verbose EHR narratives overload healthcare providers, risking diagnostic inaccuracies. While Large Language Models (LLMs) have showcased their potential in diverse language tasks, their application in the healthcare arena needs to ensure the minimization of diagnostic errors and the prevention of patient harm. In this paper, we outline an innovative approach for augmenting the proficiency of LLMs in the realm of automated diagnosis generation, achieved through the incorporation of a medical knowledge graph (KG) and a novel graph model: Dr.Knows, inspired by the clinical diagnostic reasoning process. We derive the KG from the National Library of Medicine's Unified Medical Language System (UMLS), a robust repository of biomedical knowledge. Our method negates the need for pre-training and instead leverages the KG as an auxiliary instrument aiding in the interpretation and summarization of complex medical concepts. Using real-world hospital datasets, our experimental results demonstrate that the proposed approach of combining LLMs with KG has the potential to improve the accuracy of automated diagnosis generation. More importantly, our approach offers an explainable diagnostic pathway, edging us closer to the realization of AI-augmented diagnostic decision support systems.
Authors: Ross Deans Kristensen-McLachlan, Miceal Canavan, M\'arton Kardos, Mia Jacobsen, Lene Aar{\o}e
Abstract: Recent research highlights the significant potential of ChatGPT for text annotation in social science research. However, ChatGPT is a closed-source product which has major drawbacks with regards to transparency, reproducibility, cost, and data protection. Recent advances in open-source (OS) large language models (LLMs) offer an alternative without these drawbacks. Thus, it is important to evaluate the performance of OS LLMs relative to ChatGPT and standard approaches to supervised machine learning classification. We conduct a systematic comparative evaluation of the performance of a range of OS LLMs alongside ChatGPT, using both zero- and few-shot learning as well as generic and custom prompts, with results compared to supervised classification models. Using a new dataset of tweets from US news media, and focusing on simple binary text annotation tasks, we find significant variation in the performance of ChatGPT and OS models across the tasks, and that the supervised classifier using DistilBERT generally outperforms both. Given the unreliable performance of ChatGPT and the significant challenges it poses to Open Science we advise caution when using ChatGPT for substantive text annotation tasks.
Authors: Youssef Mourchid, Marc Donias, Yannick Berthoumieu, Mohamed Najim
Abstract: This paper addresses the automatic colorization problem, which converts a gray-scale image to a colorized one. Recent deep-learning approaches can colorize automatically grayscale images. However, when it comes to different scenes which contain distinct color styles, it is difficult to accurately capture the color characteristics. In this work, we propose a fully automatic colorization approach based on Symmetric Positive Definite (SPD) Manifold Learning with a generative adversarial network (SPDGAN) that improves the quality of the colorization results. Our SPDGAN model establishes an adversarial game between two discriminators and a generator. The latter is based on ResNet architecture with few alterations. Its goal is to generate fake colorized images without losing color information across layers through residual connections. Then, we employ two discriminators from different domains. The first one is devoted to the image pixel domain, while the second one is to the Riemann manifold domain which helps to avoid color misalignment. Extensive experiments are conducted on the Places365 and COCO-stuff databases to test the effect of each component of our SPDGAN. In addition, quantitative and qualitative comparisons with state-of-the-art methods demonstrate the effectiveness of our model by achieving more realistic colorized images with less artifacts visually, and good results of PSNR, SSIM, and FID values.
Authors: Youssef Mourchid, Rim Slama
Abstract: This paper tackles the challenge of automatically assessing physical rehabilitation exercises for patients who perform the exercises without clinician supervision. The objective is to provide a quality score to ensure correct performance and achieve desired results. To achieve this goal, a new graph-based model, the Dense Spatio-Temporal Graph Conv-GRU Network with Transformer, is introduced. This model combines a modified version of STGCN and transformer architectures for efficient handling of spatio-temporal data. The key idea is to consider skeleton data respecting its non-linear structure as a graph and detecting joints playing the main role in each rehabilitation exercise. Dense connections and GRU mechanisms are used to rapidly process large 3D skeleton inputs and effectively model temporal dynamics. The transformer encoder's attention mechanism focuses on relevant parts of the input sequence, making it useful for evaluating rehabilitation exercises. The evaluation of our proposed approach on the KIMORE and UI-PRMD datasets highlighted its potential, surpassing state-of-the-art methods in terms of accuracy and computational time. This resulted in faster and more accurate learning and assessment of rehabilitation exercises. Additionally, our model provides valuable feedback through qualitative illustrations, effectively highlighting the significance of joints in specific exercises.
Authors: Md Abir Hossen, Sonam Kharade, Jason M. O'Kane, Bradley Schmerl, David Garlan, Pooyan Jamshidi
Abstract: Robotic systems are typically composed of various subsystems, such as localization and navigation, each encompassing numerous configurable components (e.g., selecting different planning algorithms). Once an algorithm has been selected for a component, its associated configuration options must be set to the appropriate values. Configuration options across the system stack interact non-trivially. Finding optimal configurations for highly configurable robots to achieve desired performance poses a significant challenge due to the interactions between configuration options across software and hardware that result in an exponentially large and complex configuration space. These challenges are further compounded by the need for transferability between different environments and robotic platforms. Data efficient optimization algorithms (e.g., Bayesian optimization) have been increasingly employed to automate the tuning of configurable parameters in cyber-physical systems. However, such optimization algorithms converge at later stages, often after exhausting the allocated budget (e.g., optimization steps, allotted time) and lacking transferability. This paper proposes CURE -- a method that identifies causally relevant configuration options, enabling the optimization process to operate in a reduced search space, thereby enabling faster optimization of robot performance. CURE abstracts the causal relationships between various configuration options and robot performance objectives by learning a causal model in the source (a low-cost environment such as the Gazebo simulator) and applying the learned knowledge to perform optimization in the target (e.g., Turtlebot 3 physical robot). We demonstrate the effectiveness and transferability of CURE by conducting experiments that involve varying degrees of deployment changes in both physical robots and simulation.
Authors: Yangyifei Luo, Zhuo Chen, Lingbing Guo, Qian Li, Wenxuan Zeng, Zhixin Cai, Jianxin Li
Abstract: Entity alignment (EA) aims to identify entities across different knowledge graphs that represent the same real-world objects. Recent embedding-based EA methods have achieved state-of-the-art performance in EA yet faced interpretability challenges as they purely rely on the embedding distance and neglect the logic rules behind a pair of aligned entities. In this paper, we propose the Align-Subgraph Entity Alignment (ASGEA) framework to exploit logic rules from Align-Subgraphs. ASGEA uses anchor links as bridges to construct Align-Subgraphs and spreads along the paths across KGs, which distinguishes it from the embedding-based methods. Furthermore, we design an interpretable Path-based Graph Neural Network, ASGNN, to effectively identify and integrate the logic rules across KGs. We also introduce a node-level multi-modal attention mechanism coupled with multi-modal enriched anchors to augment the Align-Subgraph. Our experimental results demonstrate the superior performance of ASGEA over the existing embedding-based methods in both EA and Multi-Modal EA (MMEA) tasks.
Authors: Leon Furze, Mike Perkins, Jasper Roe, Jason MacVaugh
Abstract: The rapid adoption of Generative Artificial Intelligence (GenAI) technologies in higher education has raised concerns about academic integrity, assessment practices, and student learning. Banning or blocking GenAI tools has proven ineffective, and punitive approaches ignore the potential benefits of these technologies. This paper presents the findings of a pilot study conducted at British University Vietnam (BUV) exploring the implementation of the Artificial Intelligence Assessment Scale (AIAS), a flexible framework for incorporating GenAI into educational assessments. The AIAS consists of five levels, ranging from 'No AI' to 'Full AI', enabling educators to design assessments that focus on areas requiring human input and critical thinking. Following the implementation of the AIAS, the pilot study results indicate a significant reduction in academic misconduct cases related to GenAI, a 5.9% increase in student attainment across the university, and a 33.3% increase in module passing rates. The AIAS facilitated a shift in pedagogical practices, with faculty members incorporating GenAI tools into their modules and students producing innovative multimodal submissions. The findings suggest that the AIAS can support the effective integration of GenAI in HE, promoting academic integrity while leveraging the technology's potential to enhance learning experiences. Refer to published version for final text.
Authors: Kyla Levin, Nicolas van Kempen, Emery D. Berger, Stephen N. Freund
Abstract: Debugging is a critical but challenging task for programmers. This paper proposes ChatDBG, an AI-powered debugging assistant. ChatDBG integrates large language models (LLMs) to significantly enhance the capabilities and user-friendliness of conventional debuggers. ChatDBG lets programmers engage in a collaborative dialogue with the debugger, allowing them to pose complex questions about program state, perform root cause analysis for crashes or assertion failures, and explore open-ended queries like `why is x null?'. To handle these queries, ChatDBG grants the LLM autonomy to "take the wheel": it can act as an independent agent capable of querying and controlling the debugger to navigate through stacks and inspect program state. It then reports its findings and yields back control to the programmer. By leveraging the real-world knowledge embedded in LLMs, ChatDBG can diagnose issues identifiable only through the use of domain-specific reasoning. Our ChatDBG prototype integrates with standard debuggers including LLDB and GDB for native code and Pdb for Python. Our evaluation across a diverse set of code, including C/C++ code with known bugs and a suite of Python code including standalone scripts and Jupyter notebooks, demonstrates that ChatDBG can successfully analyze root causes, explain bugs, and generate accurate fixes for a wide range of real-world errors. For the Python programs, a single query led to an actionable bug fix 67% of the time; one additional follow-up query increased the success rate to 85%. ChatDBG has seen rapid uptake; it has already been downloaded more than 65,000 times.
Authors: Di Cooke, Abigail Edwards, Sophia Barkoff, Kathryn Kelly
Abstract: Despite advancements in technology led synthetic media authentication and recent government efforts to address the threats posed by maliciously employed synthetic content via the mechanisms of law or through more public education, one of the current principal defenses against weaponized synthetic media continues to be the ability of the targeted individual to visually or auditorily recognize AI-generated content when they encounter it. However, as the realism of synthetic media continues to rapidly improve, it is vital to have an accurate understanding of just how susceptible people currently are to potentially being misled by convincing but false AI generated content. We conducted a perceptual study with 1276 participants to assess how capable people were at distinguishing between authentic and synthetic images, audio, video, and audiovisual media. We find that on average, people struggled to distinguish between synthetic and authentic media, with the mean detection performance close to a chance level performance of 50%. We also find that accuracy rates worsen when the stimuli contain any degree of synthetic content, features foreign languages, and the media type is a single modality. People are also less accurate at identifying synthetic images when they feature human faces, and when audiovisual stimuli have heterogeneous authenticity. Finally, we find that higher degrees of prior knowledgeability about synthetic media does not significantly impact detection accuracy rates, but age does, with older individuals performing worse than their younger counterparts. Collectively, these results highlight that it is no longer feasible to rely on the perceptual capabilities of people to protect themselves against the growing threat of weaponized synthetic media, and that the need for alternative countermeasures is more critical than ever before.
Authors: Yuqing Wang, Mika V. M\"antyl\"a, Serge Demeyer, Mutlu Beyazit, Joanna Kisaakye, Jesse Nyyss\"ol\"a
Abstract: Microservice-based systems (MSS) may fail with various fault types. While existing AIOps methods excel at detecting abnormal traces and locating the responsible service(s), human efforts are still required for diagnosing specific fault types and failure causes.This paper presents TraFaultDia, a novel AIOps framework to automatically classify abnormal traces into fault categories for MSS. We treat the classification process as a series of multi-class classification tasks, where each task represents an attempt to classify abnormal traces into specific fault categories for a MSS. TraFaultDia leverages meta-learning to train on several abnormal trace classification tasks with a few labeled instances from a MSS, enabling quick adaptation to new, unseen abnormal trace classification tasks with a few labeled instances across MSS. TraFaultDia's use cases are scalable depending on how fault categories are built from anomalies within MSS. We evaluated TraFaultDia on two MSS, TrainTicket and OnlineBoutique, with open datasets where each fault category is linked to faulty system components (service/pod) and a root cause. TraFaultDia automatically classifies abnormal traces into these fault categories, thus enabling the automatic identification of faulty system components and root causes without manual analysis. TraFaultDia achieves 93.26% and 85.20% accuracy on 50 new classification tasks for TrainTicket and OnlineBoutique, respectively, when trained within the same MSS with 10 labeled instances per category. In the cross-system context, when TraFaultDia is applied to a MSS different from the one it is trained on, TraFaultDia gets an average accuracy of 92.19% and 84.77% for the same set of 50 new, unseen abnormal trace classification tasks of the respective systems, also with 10 labeled instances provided for each fault category per task in each system.
Authors: Junsong Zhang, Zisong Chen, Chunyu Lin, Lang Nie, Zhijie Shen, Kang Liao, Junda Huang, Yao Zhao
Abstract: Panoramic distortion poses a significant challenge in 360 depth estimation, particularly pronounced at the north and south poles. Existing methods either adopt a bi-projection fusion strategy to remove distortions or model long-range dependencies to capture global structures, which can result in either unclear structure or insufficient local perception. In this paper, we propose a spherical geometry transformer, named SGFormer, to address the above issues, with an innovative step to integrate spherical geometric priors into vision transformers. To this end, we retarget the transformer decoder to a spherical prior decoder (termed SPDecoder), which endeavors to uphold the integrity of spherical structures during decoding. Concretely, we leverage bipolar re-projection, circular rotation, and curve local embedding to preserve the spherical characteristics of equidistortion, continuity, and surface distance, respectively. Furthermore, we present a query-based global conditional position embedding to compensate for spatial structure at varying resolutions. It not only boosts the global perception of spatial position but also sharpens the depth structure across different patches. Finally, we conduct extensive experiments on popular benchmarks, demonstrating our superiority over state-of-the-art solutions.
Authors: Zhuowei Li, Zihao Xu, Ligong Han, Yunhe Gao, Song Wen, Di Liu, Hao Wang, Dimitris N. Metaxas
Abstract: In-context Learning (ICL) empowers large language models (LLMs) to swiftly adapt to unseen tasks at inference-time by prefixing a few demonstration examples before queries. Despite its versatility, ICL incurs substantial computational and memory overheads compared to zero-shot learning and is sensitive to the selection and order of demonstration examples. In this work, we introduce Implicit In-context Learning (I2CL), an innovative paradigm that reduces the inference cost of ICL to that of zero-shot learning with minimal information loss. I2CL operates by first generating a condensed vector representation, namely a context vector, extracted from the demonstration examples. It then conducts an inference-time intervention through injecting a linear combination of the context vector and query activations back into the model's residual streams. Empirical evaluation on nine real-world tasks across three model architectures demonstrates that I2CL achieves few-shot level performance at zero-shot inference cost, and it exhibits robustness against variations in demonstration examples. Furthermore, I2CL facilitates a novel representation of task-ids, enhancing task similarity detection and fostering effective transfer learning. We also perform a comprehensive analysis and ablation study on I2CL, offering deeper insights into its internal mechanisms. Code is available at https://github.com/LzVv123456/I2CL.
Authors: Chankyu Lee, Rajarshi Roy, Mengyao Xu, Jonathan Raiman, Mohammad Shoeybi, Bryan Catanzaro, Wei Ping
Abstract: Decoder-only LLM-based embedding models are beginning to outperform BERT or T5-based embedding models in general-purpose text embedding tasks, including dense vector-based retrieval. In this work, we introduce NV-Embed, incorporating architectural designs, training procedures, and curated datasets to significantly enhance the performance of LLM as a versatile embedding model, while maintaining its simplicity and reproducibility. For model architecture, we propose a latent attention layer to obtain pooled embeddings, which consistently improves retrieval and downstream task accuracy compared to mean pooling or using the last
Authors: Marco S\"alzer, Eric Alsmann, Martin Lange
Abstract: We analyse the complexity of the satisfiability problem, or similarly feasibility problem, (trSAT) for transformer encoders (TE), which naturally occurs in formal verification or interpretation, collectively referred to as formal reasoning. We find that trSAT is undecidable when considering TE as they are commonly studied in the expressiveness community. Furthermore, we identify practical scenarios where trSAT is decidable and establish corresponding complexity bounds. Beyond trivial cases, we find that quantized TE, those restricted by fixed-width arithmetic, lead to the decidability of trSAT due to their limited attention capabilities. However, the problem remains difficult, as we establish scenarios where trSAT is NEXPTIME-hard and others where it is solvable in NEXPTIME for quantized TE. To complement our complexity results, we place our findings and their implications in the broader context of formal reasoning.
Authors: Dujian Ding, Bicheng Xu, Laks V. S. Lakshmanan
Abstract: Classification tasks play a fundamental role in various applications, spanning domains such as healthcare, natural language processing and computer vision. With the growing popularity and capacity of machine learning models, people can easily access trained classifiers as a service online or offline. However, model use comes with a cost and classifiers of higher capacity (such as large foundation models) usually incur higher inference costs. To harness the respective strengths of different classifiers, we propose a principled approach, OCCAM, to compute the best classifier assignment strategy over classification queries (termed as the optimal model portfolio) so that the aggregated accuracy is maximized, under user-specified cost budgets. Our approach uses an unbiased and low-variance accuracy estimator and effectively computes the optimal solution by solving an integer linear programming problem. On a variety of real-world datasets, OCCAM achieves 40% cost reduction with little to no accuracy drop.
Authors: Che Liu, Zhongwei Wan, Yuqi Wang, Hui Shen, Haozhe Wang, Kangyu Zheng, Mi Zhang, Rossella Arcucci
Abstract: Automatic radiology report generation holds significant potential to streamline the labor-intensive process of report writing by radiologists, particularly for 3D radiographs such as CT scans. While CT scans are critical for clinical diagnostics, they remain less explored compared to 2D radiographs. To date, there has been no comprehensive benchmark for 3D radiograph report generation (3DRRG), nor sufficient investigation into the optimal training strategies for Vision Language Models (VLMs) in this context, particularly with respect to vision encoder choices, visual token compression, and model scaling. In this work, we make three key contributions. We curate **CT-3DRRG**, the largest **publicly** available 3D CT-report dataset, establishing a robust and diverse benchmark for evaluating VLM performance on 3DRRG. Furthermore, we propose a comprehensive training recipe for building high-performing VLMs for 3DRRG, exploring key factors such as vision encoder pretraining strategies, visual token compression, and the impact of data & model scale. Guided by these findings, we introduce **Argus**, a state-of-the-art family of VLMs that achieve superior performance across different model sizes and input 3D medical image resolutions, efficiently processing high-resolution 3D images up to $512 \times 512 \times 256$[^1].
Authors: Pingzhi Li, Xiaolong Jin, Zhen Tan, Yu Cheng, Tianlong Chen
Abstract: Mixture-of-Experts (MoE) is a promising way to scale up the learning capacity of large language models. It increases the number of parameters while keeping FLOPs nearly constant during inference through sparse activation. Yet, it still suffers from significant memory overheads due to the vast parameter size, necessitating model compression techniques. Post-training quantization offers a powerful approach for model compression. Existing methods adopt a fixed quantization precision for the entire MoE model. This rigid setup can lead to suboptimal performance, without considering the inherent sparse structure. For example, MoE's sparse routing mechanism leads to different activation patterns, where shared experts are accessed by all tokens while token-conditioned experts are selectively activated. This activation disparity suggests different quantization requirements, with consistently activated shared experts potentially needing higher precision to maintain model quality. In this paper, we study a fine-grained precision setup for MoE quantization. We explore MoE structure-aware quantization heuristics, ranging from coarse (e.g., MoE layers) to fine granularity (e.g., linear layers). Our investigations reveal critical principles, where different MoE structures require varying numbers of bits for effective quantization. Conclusions are supported by extensive benchmarking across two representative MoE models and six tasks including commonsense reasoning and natural language understanding. We further show that an MoE quantized in a fined-grained mixed precision achieved state-of-the-art 65.35% performance on average compared to the baseline 64.30% (i.e., GPTQ). Moreover, based on the findings, we introduce novel data-driven techniques for optimizing bit allocation in MoE quantization, including the outlier-aware linear layer scorer and MoE block importance predictor.
Authors: Qijun Gan, Song Wang, Shengtao Wu, Jianke Zhu
Abstract: Recently, artificial intelligence techniques for education have been received increasing attentions, while it still remains an open problem to design the effective music instrument instructing systems. Although key presses can be directly derived from sheet music, the transitional movements among key presses require more extensive guidance in piano performance. In this work, we construct a piano-hand motion generation benchmark to guide hand movements and fingerings for piano playing. To this end, we collect an annotated dataset, PianoMotion10M, consisting of 116 hours of piano playing videos from a bird's-eye view with 10 million annotated hand poses. We also introduce a powerful baseline model that generates hand motions from piano audios through a position predictor and a position-guided gesture generator. Furthermore, a series of evaluation metrics are designed to assess the performance of the baseline model, including motion similarity, smoothness, positional accuracy of left and right hands, and overall fidelity of movement distribution. Despite that piano key presses with respect to music scores or audios are already accessible, PianoMotion10M aims to provide guidance on piano fingering for instruction purposes. The source code and dataset can be accessed at https://github.com/agnJason/PianoMotion10M.
Authors: Chunming He, Yuqi Shen, Chengyu Fang, Fengyang Xiao, Longxiang Tang, Yulun Zhang, Wangmeng Zuo, Zhenhua Guo, Xiu Li
Abstract: Deep generative models have garnered significant attention in low-level vision tasks due to their generative capabilities. Among them, diffusion model-based solutions, characterized by a forward diffusion process and a reverse denoising process, have emerged as widely acclaimed for their ability to produce samples of superior quality and diversity. This ensures the generation of visually compelling results with intricate texture information. Despite their remarkable success, a noticeable gap exists in a comprehensive survey that amalgamates these pioneering diffusion model-based works and organizes the corresponding threads. This paper proposes the comprehensive review of diffusion model-based techniques. We present three generic diffusion modeling frameworks and explore their correlations with other deep generative models, establishing the theoretical foundation. Following this, we introduce a multi-perspective categorization of diffusion models, considering both the underlying framework and the target task. Additionally, we summarize extended diffusion models applied in other tasks, including medical, remote sensing, and video scenarios. Moreover, we provide an overview of commonly used benchmarks and evaluation metrics. We conduct a thorough evaluation, encompassing both performance and efficiency, of diffusion model-based techniques in three prominent tasks. Finally, we elucidate the limitations of current diffusion models and propose seven intriguing directions for future research. This comprehensive examination aims to facilitate a profound understanding of the landscape surrounding denoising diffusion models in the context of low-level vision tasks. A curated list of diffusion model-based techniques in over 20 low-level vision tasks can be found at https://github.com/ChunmingHe/awesome-diffusion-models-in-low-level-vision.
URLs: https://github.com/ChunmingHe/awesome-diffusion-models-in-low-level-vision.
Authors: Xikun Zhang, Dongjin Song, Yushan Jiang, Yixin Chen, Dacheng Tao
Abstract: Observation-based trajectory prediction for systems with unknown dynamics is essential in fields such as physics and biology. Most existing approaches are limited to learning within a single system with fixed dynamics patterns. However, many real-world applications require learning across systems with evolving dynamics patterns, a challenge that has been largely overlooked. To address this, we systematically investigate the problem of Continual Dynamics Learning (CDL), examining task configurations and evaluating the applicability of existing techniques, while identifying key challenges. In response, we propose the Mode-switching Graph ODE (MS-GODE) model, which integrates the strengths LG-ODE and sub-network learning with a mode-switching module, enabling efficient learning over varying dynamics. Moreover, we construct a novel benchmark of biological dynamic systems for CDL, Bio-CDL, featuring diverse systems with disparate dynamics and significantly enriching the research field of machine learning for dynamic systems. Our code available at https://github.com/QueuQ/MS-GODE.
Authors: Benjie Wang, Guy Van den Broeck
Abstract: Probabilistic circuits are a unifying representation of functions as computation graphs of weighted sums and products. Their primary application is in probabilistic modeling, where circuits with non-negative weights (monotone circuits) can be used to represent and learn density/mass functions, with tractable marginal inference. Recently, it was proposed to instead represent densities as the square of the circuit function (squared circuits); this allows the use of negative weights while retaining tractability, and can be exponentially more expressive efficient than monotone circuits. Unfortunately, we show the reverse also holds, meaning that monotone circuits and squared circuits are incomparable in general. This raises the question of whether we can reconcile, and indeed improve upon the two modeling approaches. We answer in the positive by proposing Inception PCs, a novel type of circuit that naturally encompasses both monotone circuits and squared circuits as special cases, and employs complex parameters. Empirically, we validate that Inception PCs can outperform both monotone and squared circuits on a range of tabular and image datasets.
Authors: Weiwei Xu, Kai Gao, Hao He, Minghui Zhou
Abstract: Recent advances in Large Language Models (LLMs) have revolutionized code generation, leading to widespread adoption of AI coding tools by developers. However, LLMs can generate license-protected code without providing the necessary license information, leading to potential intellectual property violations during software production. This paper addresses the critical, yet underexplored, issue of license compliance in LLM-generated code by establishing a benchmark to evaluate the ability of LLMs to provide accurate license information for their generated code. To establish this benchmark, we conduct an empirical study to identify a reasonable standard for "striking similarity" that excludes the possibility of independent creation, indicating a copy relationship between the LLM output and certain open-source code. Based on this standard, we propose LiCoEval, to evaluate the license compliance capabilities of LLMs, i.e., the ability to provide accurate license or copyright information when they generate code with striking similarity to already existing copyrighted code. Using LiCoEval, we evaluate 14 popular LLMs, finding that even top-performing LLMs produce a non-negligible proportion (0.88% to 2.01%) of code strikingly similar to existing open-source implementations. Notably, most LLMs fail to provide accurate license information, particularly for code under copyleft licenses. These findings underscore the urgent need to enhance LLM compliance capabilities in code generation tasks. Our study provides a foundation for future research and development to improve license compliance in AI-assisted software development, contributing to both the protection of open-source software copyrights and the mitigation of legal risks for LLM users.
Authors: Lei Shi, Zhimeng Liu, Yi Yang, Weize Wu, Yuyang Zhang, Hongbo Zhang, Jing Lin, Siyu Wu, Zihan Chen, Ruiming Li, Nan Wang, Zipeng Liu, Huobin Tan, Hongyi Gao, Yue Zhang, Ge Wang
Abstract: The extraction of Metal-Organic Frameworks (MOFs) synthesis route from literature has been crucial for the logical MOFs design with desirable functionality. The recent advent of large language models (LLMs) provides disruptively new solution to this long-standing problem. While the latest researches mostly stick to primitive zero-shot LLMs lacking specialized material knowledge, we introduce in this work the few-shot LLM in-context learning paradigm. First, a human-AI interactive data curation approach is proposed to secure high-quality demonstrations. Second, an information retrieval algorithm is applied to pick and quantify few-shot demonstrations for each extraction. Over three datasets randomly sampled from nearly 90,000 well-defined MOFs, we conduct triple evaluations to validate our method. The synthesis extraction, structure inference, and material design performance of the proposed few-shot LLMs all significantly outplay zero-shot LLM and baseline methods. The lab-synthesized material guided by LLM surpasses 91.1% high-quality MOFs of the same class reported in the literature, on the key physical property of specific surface area.
Authors: Jinluan Yang, Zhengyu Chen, Teng Xiao, Wenqiao Zhang, Yong Lin, Kun Kuang
Abstract: Heterophilic Graph Neural Networks (HGNNs) have shown promising results for semi-supervised learning tasks on graphs. Notably, most real-world heterophilic graphs are composed of a mixture of nodes with different neighbor patterns, exhibiting local node-level homophilic and heterophilic structures. However, existing works are only devoted to designing better HGNN backbones or architectures for node classification tasks on heterophilic and homophilic graph benchmarks simultaneously, and their analyses of HGNN performance with respect to nodes are only based on the determined data distribution without exploring the effect caused by this structural difference between training and testing nodes. How to learn invariant node representations on heterophilic graphs to handle this structure difference or distribution shifts remains unexplored. In this paper, we first discuss the limitations of previous graph-based invariant learning methods from the perspective of data augmentation. Then, we propose \textbf{HEI}, a framework capable of generating invariant node representations through incorporating heterophily information to infer latent environments without augmentation, which are then used for invariant prediction, under heterophilic graph structure distribution shifts. We theoretically show that our proposed method can achieve guaranteed performance under heterophilic graph structure distribution shifts. Extensive experiments on various benchmarks and backbones can also demonstrate the effectiveness of our method compared with existing state-of-the-art baselines. The code is available at https://github.com/Yangjinluan/HEI
Authors: Junhao Chen, Bowen Wang, Zhouqiang Jiang, Yuta Nakashima
Abstract: Large Language Models (LLMs) have demonstrated significant capabilities, particularly in the domain of question answering (QA). However, their effectiveness in QA is often undermined by the vagueness of user questions. To address this issue, we introduce single-round instance-level prompt optimization, referred to as question rewriter. By enhancing the intelligibility of human questions for black-box LLMs, our question rewriter improves the quality of generated answers. The rewriter is optimized using direct preference optimization based on feedback collected from automatic criteria for evaluating generated answers; therefore, its training does not require costly human annotations. The experiments across multiple black-box LLMs and long-form question answering (LFQA) datasets demonstrate the efficacy of our method. This paper provides a practical framework for training question rewriters and sets a precedent for future explorations in prompt optimization within LFQA tasks. Code is available at https://github.com/3244we/Question-Rewriter.
Authors: Yicheng Wang, Jiayi Yuan, Yu-Neng Chuang, Zhuoer Wang, Yingchi Liu, Mark Cusick, Param Kulkarni, Zhengping Ji, Yasser Ibrahim, Xia Hu
Abstract: Large Language Models (LLMs) are increasingly serving as evaluators in Natural Language Generation (NLG) tasks; this is often referred to as ``LLM-as-a-judge'' paradigm. However, the capabilities of LLMs in evaluating NLG quality remain underexplored. Current studies depend on human assessments and simple metrics that fail to capture the discernment of LLMs across diverse NLG tasks. To address this gap, we propose the Discernment of Hierarchical Perturbation (DHP) benchmarking framework, which provides quantitative discernment scores for LLMs. This framework leverages hierarchically perturbed text data and statistical tests to systematically measure the NLG evaluation capabilities of LLMs. We re-established six evaluation datasets for this benchmark, covering four NLG tasks: Summarization, Story Completion, Question Answering, and Translation. Our comprehensive benchmarking of five major LLM families provides critical insight into their strengths and limitations as NLG evaluators. Our dataset is available at https://huggingface.co/datasets/YCWANGVINCE/DHP_Benchmark.
URLs: https://huggingface.co/datasets/YCWANGVINCE/DHP_Benchmark.
Authors: Md Awsafur Rahman, Zaber Ibn Abdul Hakim, Najibul Haque Sarker, Bishmoy Paul, Shaikh Anowarul Fattah
Abstract: The recent surge in AI-generated songs presents exciting possibilities and challenges. These innovations necessitate the ability to distinguish between human-composed and synthetic songs to safeguard artistic integrity and protect human musical artistry. Existing research and datasets in fake song detection only focus on singing voice deepfake detection (SVDD), where the vocals are AI-generated but the instrumental music is sourced from real songs. However, these approaches are inadequate for detecting contemporary end-to-end artificial songs where all components (vocals, music, lyrics, and style) could be AI-generated. Additionally, existing datasets lack music-lyrics diversity, long-duration songs, and open-access fake songs. To address these gaps, we introduce SONICS, a novel dataset for end-to-end Synthetic Song Detection (SSD), comprising over 97k songs (4,751 hours) with over 49k synthetic songs from popular platforms like Suno and Udio. Furthermore, we highlight the importance of modeling long-range temporal dependencies in songs for effective authenticity detection, an aspect entirely overlooked in existing methods. To utilize long-range patterns, we introduce SpecTTTra, a novel architecture that significantly improves time and memory efficiency over conventional CNN and Transformer-based models. For long songs, our top-performing variant outperforms ViT by 8% in F1 score, is 38% faster, and uses 26% less memory, while also surpassing ConvNeXt with a 1% F1 score gain, 20% speed boost, and 67% memory reduction.
Authors: Rishi Kalra, Zekun Wu, Ayesha Gulley, Airlie Hilliard, Xin Guan, Adriano Koshiyama, Philip Treleaven
Abstract: Large Language Models (LLMs) face limitations in AI legal and policy applications due to outdated knowledge, hallucinations, and poor reasoning in complex contexts. Retrieval-Augmented Generation (RAG) systems address these issues by incorporating external knowledge, but suffer from retrieval errors, ineffective context integration, and high operational costs. This paper presents the Hybrid Parameter-Adaptive RAG (HyPA-RAG) system, designed for the AI legal domain, with NYC Local Law 144 (LL144) as the test case. HyPA-RAG integrates a query complexity classifier for adaptive parameter tuning, a hybrid retrieval approach combining dense, sparse, and knowledge graph methods, and a comprehensive evaluation framework with tailored question types and metrics. Testing on LL144 demonstrates that HyPA-RAG enhances retrieval accuracy, response fidelity, and contextual precision, offering a robust and adaptable solution for high-stakes legal and policy applications.
Authors: To Eun Kim, Fernando Diaz
Abstract: Modern language models frequently include retrieval components to improve their outputs, giving rise to a growing number of retrieval-augmented generation (RAG) systems. Yet, most existing work in RAG has underemphasized fair ranking techniques and neglected the diverse interests of all stakeholders. In this paper, we present the first comprehensive study of RAG systems that incorporate fairness-aware rankings, focusing on both ranking fairness and attribution fairness - ensuring equitable exposure of sources cited in the final text. We specifically examine item-side fairness, i.e., whether retrieved documents receive balanced exposure, and assess how this affects both the system's overall performance and the eventual distribution of cited sources. Across twelve RAG models and seven tasks, we find that fairness-aware retrieval frequently retains or even improves ranking effectiveness and generation quality, countering the widespread belief that fairness compromises system performance. Moreover, we show that fair retrieval leads to more balanced attribution in the final responses, ensuring that the cited sources are credited more equitably. Our results underscore the importance of item-side fairness throughout both retrieval and generation phases, offering key insights for building more responsible and equitable RAG systems and illustrating promising avenues for future exploration in fair ranking and source attribution.
Authors: Mabel Ogonna, Abigail Adeniran, Adewale Adeyemo
Abstract: Deep Neural Networks (DNNs) have achieved state of the art results and even outperformed human accuracy in many challenging tasks, leading to DNNs adoption in a variety of fields including natural language processing, pattern recognition, prediction, and control optimization. However, DNNs are accompanied by uncertainty about their results, causing them to predict an outcome that is either incorrect or outside of a certain level of confidence. These uncertainties stem from model or data constraints, which could be exacerbated by adversarial attacks. Adversarial attacks aim to provide perturbed input to DNNs, causing the DNN to make incorrect predictions or increase model uncertainty. In this review, we explore the relationship between DNN uncertainty and adversarial attacks, emphasizing how adversarial attacks might raise DNN uncertainty.
Authors: Georgios Papasotiropoulos, Seyedeh Zeinab Pishbin, Oskar Skibski, Piotr Skowron, Tomasz W\k{a}s
Abstract: In participatory budgeting (PB), voters decide through voting which subset of projects to fund within a given budget. Proportionality in the context of PB is crucial to ensure equal treatment of all groups of voters. However, pure proportional rules can sometimes lead to suboptimal outcomes. We introduce the Method of Equal Shares with Bounded Overspending (BOS Equal Shares), a robust variant of Equal Shares that balances proportionality and efficiency. BOS Equal Shares addresses inefficiencies implied by strict proportionality axioms, yet the rule still provides fairness guarantees, similar to the original Method of Equal Shares. Our extensive empirical analysis on real-world PB instances shows excellent performance of BOS Equal Shares across several metrics. In the course of the analysis, we also present and examine a fractional variant of the Method of Equal Shares which allows for partial funding of projects.
Authors: Zhenyan Lu, Xiang Li, Dongqi Cai, Rongjie Yi, Fangming Liu, Xiwen Zhang, Nicholas D. Lane, Mengwei Xu
Abstract: Small language models (SLMs), despite their widespread adoption in modern smart devices, have received significantly less academic attention compared to their large language model (LLM) counterparts, which are predominantly deployed in data centers and cloud environments. While researchers continue to improve the capabilities of LLMs in the pursuit of artificial general intelligence, SLM research aims to make machine intelligence more accessible, affordable, and efficient for everyday tasks. Focusing on transformer-based, decoder-only language models with 100M-5B parameters, we survey 70 state-of-the-art open-source SLMs, analyzing their technical innovations across three axes: architectures, training datasets, and training algorithms. In addition, we evaluate their capabilities in various domains, including commonsense reasoning, in-context learning, mathematics, and coding. To gain further insight into their on-device runtime costs, we benchmark their inference latency and memory footprints. Through in-depth analysis of our benchmarking data, we offer valuable insights to advance research in this field.
Authors: Xingyu Ma, Xin Tian, Lingxiang Wu, Xuepeng Wang, Xueming Tang, Jinqiao Wang
Abstract: Text-to-SQL is a subtask in semantic parsing that has seen rapid progress with the evolution of Large Language Models (LLMs). However, LLMs face challenges due to hallucination issues and a lack of domain-specific database knowledge(such as table schema and cell values). As a result, they can make errors in generating table names, columns, and matching values to the correct columns in SQL statements. This paper introduces a method of knowledge injection to enhance LLMs' ability to understand schema contents by incorporating prior knowledge. This approach improves their performance in Text-to-SQL tasks. Experimental results show that pre-training LLMs on domain-specific database knowledge and fine-tuning them on downstream Text-to-SQL tasks significantly improves the Execution Match (EX) and Exact Match (EM) metrics across various models. This effectively reduces errors in generating column names and matching values to the columns. Furthermore, the knowledge-injected models can be applied to many downstream Text-to-SQL tasks, demonstrating the generalizability of the approach presented in this paper.
Authors: Roland Pihlakas, Joel Pyykk\"o
Abstract: Developing safe, aligned agentic AI systems requires comprehensive empirical testing, yet many existing benchmarks neglect crucial themes aligned with biology and economics, both time-tested fundamental sciences describing our needs and preferences. To address this gap, the present work focuses on introducing biologically and economically motivated themes that have been neglected in current mainstream discussions on AI safety - namely a set of multi-objective, multi-agent alignment benchmarks that emphasize homeostasis for bounded and biological objectives, diminishing returns for unbounded, instrumental, and business objectives, sustainability principle, and resource sharing. We implemented eight main benchmark environments on the above themes, to illustrate key pitfalls and challenges in agentic AI-s, such as unboundedly maximizing a homeostatic objective, over-optimizing one objective at the expense of others, neglecting safety constraints, or depleting shared resources.
Authors: Yuxin Liu, Rongjun Ge, Yuting He, Zhan Wu, Shangwen Yang, Yuan Gao, Chenyu You, Ge Wang, Yang Chen, Shuo Li
Abstract: Non-ideal measurement computed tomography (NICT) employs suboptimal imaging protocols to expand CT applications. However, the resulting trade-offs degrade image quality, limiting clinical acceptability. Although deep learning methods have been used to enhance NICT images, their reliance on large training datasets and limited generalizability across diverse settings hinder practical use. We propose the multi-scale integrated Transformer AMPlifier (TAMP), the first imaging foundation model for universal NICT enhancement. Pre-trained on 10.8 million physics-driven simulated NICT images, TAMP generalizes effectively across various NICT settings, defect degrees, and body regions. Moreover, a parameter-efficient fine-tuning strategy enables TAMP to adapt to specific clinical scenarios using only few slices. Extensive experiments, including radiologists and real-world validations, demonstrate that TAMP consistently improves image quality and clinical acceptability, underscoring its significant potential to advance CT imaging and broaden NICT applications in clinical practice.
Authors: Rishi Veerapaneni, Muhammad Suhail Saleem, Jiaoyang Li, Maxim Likhachev
Abstract: Traditional multi-agent path finding (MAPF) methods try to compute entire start-goal paths which are collision free. However, computing an entire path can take too long for MAPF systems where agents need to replan fast. Methods that address this typically employ a "windowed" approach and only try to find collision free paths for a small windowed timestep horizon. This adaptation comes at the cost of incompleteness; all current windowed approaches can become stuck in deadlock or livelock. Our main contribution is to introduce our framework, WinC-MAPF, for Windowed MAPF that enables completeness. Our framework uses heuristic update insights from single-agent real-time heuristic search algorithms as well as agent independence ideas from MAPF algorithms. We also develop Single-Step CBS (SS-CBS), an instantiation of this framework using a novel modification to CBS. We show how SS-CBS, which only plans a single step and updates heuristics, can effectively solve tough scenarios where existing windowed approaches fail.
Authors: Xinyuan Wang, Haozhou Li, Dingfang Zheng, Qinke Peng
Abstract: The global COVID-19 pandemic underscored major deficiencies in traditional healthcare systems, hastening the advancement of online medical services, especially in medical triage and consultation. However, existing studies face two main challenges. First, the scarcity of large-scale, publicly available, domain-specific medical datasets due to privacy concerns, with current datasets being small and limited to a few diseases, limiting the effectiveness of triage methods based on Pre-trained Language Models (PLMs). Second, existing methods lack medical knowledge and struggle to accurately understand professional terms and expressions in patient-doctor consultations. To overcome these obstacles, we construct the Large-scale Chinese Medical Dialogue Corpora (LCMDC), thereby addressing the data shortage in this field. Moreover, we further propose a novel triage system that combines BERT-based supervised learning with prompt learning, as well as a GPT-based medical consultation model. To enhance domain knowledge acquisition, we pre-trained PLMs using our self-constructed background corpus. Experimental results on the LCMDC demonstrate the efficacy of our proposed systems.
Authors: Nikola Jovanovi\'c, Robin Staab, Maximilian Baader, Martin Vechev
Abstract: RAG enables LLMs to easily incorporate external data, raising concerns for data owners regarding unauthorized usage of their content. The challenge of detecting such unauthorized usage remains underexplored, with datasets and methods from adjacent fields being ill-suited for its study. We take several steps to bridge this gap. First, we formalize this problem as (black-box) RAG Dataset Inference (RAG-DI). We then introduce a novel dataset designed for realistic benchmarking of RAG-DI methods, alongside a set of baselines. Finally, we propose Ward, a method for RAG-DI based on LLM watermarks that equips data owners with rigorous statistical guarantees regarding their dataset's misuse in RAG corpora. Ward consistently outperforms all baselines, achieving higher accuracy, superior query efficiency and robustness. Our work provides a foundation for future studies of RAG-DI and highlights LLM watermarks as a promising approach to this problem.
Authors: Yuxuan Yao, Han Wu, Mingyang Liu, Sichun Luo, Xiongwei Han, Jie Liu, Zhijiang Guo, Linqi Song
Abstract: Large language models (LLMs) exhibit varying strengths and weaknesses across different tasks, prompting recent studies to explore the benefits of ensembling models to leverage their complementary advantages. However, existing LLM ensembling methods often overlook model compatibility and struggle with inefficient alignment of probabilities across the entire vocabulary. In this study, we empirically investigate the factors influencing ensemble performance, identifying model performance, vocabulary size, and response style as key determinants, revealing that compatibility among models is essential for effective ensembling. This analysis leads to the development of a simple yet effective model selection strategy that identifies compatible models. Additionally, we introduce the \textsc{Uni}on \textsc{T}op-$k$ \textsc{E}nsembling (\textsc{UniTE}), a novel approach that efficiently combines models by focusing on the union of the top-k tokens from each model, thereby avoiding the need for full vocabulary alignment and reducing computational overhead. Extensive evaluations across multiple benchmarks demonstrate that \textsc{UniTE} significantly enhances performance compared to existing methods, offering a more efficient framework for LLM ensembling.
Authors: Long Zhao, Sanghyun Woo, Ziyu Wan, Yandong Li, Han Zhang, Boqing Gong, Hartwig Adam, Xuhui Jia, Ting Liu
Abstract: In generative modeling, tokenization simplifies complex data into compact, structured representations, creating a more efficient, learnable space. For high-dimensional visual data, it reduces redundancy and emphasizes key features for high-quality generation. Current visual tokenization methods rely on a traditional autoencoder framework, where the encoder compresses data into latent representations, and the decoder reconstructs the original input. In this work, we offer a new perspective by proposing denoising as decoding, shifting from single-step reconstruction to iterative refinement. Specifically, we replace the decoder with a diffusion process that iteratively refines noise to recover the original image, guided by the latents provided by the encoder. We evaluate our approach by assessing both reconstruction (rFID) and generation quality (FID), comparing it to state-of-the-art autoencoding approaches. By adopting iterative reconstruction through diffusion, our autoencoder, namely $\epsilon$-VAE, achieves high reconstruction quality, which in turn enhances downstream generation quality by 22% and provides 2.3$\times$ inference speedup. We hope this work offers new insights into integrating iterative generation and autoencoding for improved compression and generation.
Authors: Jinghan Li, Yuan Gao, Jinda Lu, Junfeng Fang, Congcong Wen, Hui Lin, Xiang Wang
Abstract: Graph Anomaly Detection (GAD) is crucial for identifying abnormal entities within networks, garnering significant attention across various fields. Traditional unsupervised methods, which decode encoded latent representations of unlabeled data with a reconstruction focus, often fail to capture critical discriminative content, leading to suboptimal anomaly detection. To address these challenges, we present a Diffusion-based Graph Anomaly Detector (DiffGAD). At the heart of DiffGAD is a novel latent space learning paradigm, meticulously designed to enhance its proficiency by guiding it with discriminative content. This innovative approach leverages diffusion sampling to infuse the latent space with discriminative content and introduces a content-preservation mechanism that retains valuable information across different scales, significantly improving its adeptness at identifying anomalies with limited time and space complexity. Our comprehensive evaluation of DiffGAD, conducted on six real-world and large-scale datasets with various metrics, demonstrated its exceptional performance.
Authors: Hanti Lin
Abstract: This article reviews and develops an epistemological tradition in the philosophy of science, known as convergentism, which holds that inference methods should be assessed based on their ability to converge to the truth across a range of possible scenarios. Emphasis is placed on its historical origins in the work of C. S. Peirce and its recent developments in formal epistemology and data science (including statistics and machine learning). Comparisons are made with three other traditions: (1) explanationism, which holds that theory choice should be guided by a theory's overall balance of explanatory virtues, such as simplicity and fit with data; (2) instrumentalism, which maintains that scientific inference should be driven by the goal of obtaining useful models rather than true theories; and (3) Bayesianism, which shifts the focus from all-or-nothing beliefs to degrees of belief.
Authors: Zhen Liang, Weishan Ye, Qile Liu, Li Zhang, Gan Huang, Yongjie Zhou
Abstract: Non-suicidal self-injury (NSSI) is a serious threat to the physical and mental health of adolescents, significantly increasing the risk of suicide and attracting widespread public concern. Electroencephalography (EEG), as an objective tool for identifying brain disorders, holds great promise. However, extracting meaningful and reliable features from high-dimensional EEG data, especially by integrating spatiotemporal brain dynamics into informative representations, remains a major challenge. In this study, we introduce an advanced semi-supervised adversarial network, NSSI-Net, to effectively model EEG features related to NSSI. NSSI-Net consists of two key modules: a spatial-temporal feature extraction module and a multi-concept discriminator. In the spatial-temporal feature extraction module, an integrated 2D convolutional neural network (2D-CNN) and a bi-directional Gated Recurrent Unit (BiGRU) are used to capture both spatial and temporal dynamics in EEG data. In the multi-concept discriminator, signal, gender, domain, and disease levels are fully explored to extract meaningful EEG features, considering individual, demographic, disease variations across a diverse population. Based on self-collected NSSI data (n=114), the model's effectiveness and reliability are demonstrated, with a 5.44% improvement in performance compared to existing machine learning and deep learning methods. This study advances the understanding and early diagnosis of NSSI in adolescents with depression, enabling timely intervention. The source code is available at https://github.com/Vesan-yws/NSSINet.
Authors: Che Liu, Zhongwei Wan, Haozhe Wang, Yinda Chen, Talha Qaiser, Chen Jin, Fariba Yousefi, Nikolay Burlutskiy, Rossella Arcucci
Abstract: Medical Vision-Language Pre-training (MedVLP) has made significant progress in enabling zero-shot tasks for medical image understanding. However, training MedVLP models typically requires large-scale datasets with paired, high-quality image-text data, which are scarce in the medical domain. Recent advancements in Large Language Models (LLMs) and diffusion models have made it possible to generate large-scale synthetic image-text pairs. This raises the question: "Can MedVLP succeed using purely synthetic data?" To address this, we use off-the-shelf generative models to create synthetic radiology reports and paired Chest X-ray (CXR) images, and propose an automated pipeline to build a diverse, high-quality synthetic dataset, enabling a rigorous study that isolates model and training settings, focusing entirely from the data perspective. Our results show that MedVLP models trained exclusively on synthetic data outperform those trained on real data by 3.8% in averaged AUC on zero-shot classification. Moreover, using a combination of synthetic and real data leads to a further improvement of 9.07%. Additionally, MedVLP models trained on synthetic or mixed data consistently outperform those trained on real data in zero-shot grounding, as well as in fine-tuned classification and segmentation tasks. Our analysis suggests MedVLP trained on well-designed synthetic data can outperform models trained on real datasets, which may be limited by low-quality samples and long-tailed distributions.
Authors: Chen Bo Calvin Zhang, Zhang-Wei Hong, Aldo Pacchiano, Pulkit Agrawal
Abstract: Reward shaping is critical in reinforcement learning (RL), particularly for complex tasks where sparse rewards can hinder learning. However, choosing effective shaping rewards from a set of reward functions in a computationally efficient manner remains an open challenge. We propose Online Reward Selection and Policy Optimization (ORSO), a novel approach that frames the selection of shaping reward function as an online model selection problem. ORSO automatically identifies performant shaping reward functions without human intervention with provable regret guarantees. We demonstrate ORSO's effectiveness across various continuous control tasks. Compared to prior approaches, ORSO significantly reduces the amount of data required to evaluate a shaping reward function, resulting in superior data efficiency and a significant reduction in computational time (up to 8 times). ORSO consistently identifies high-quality reward functions outperforming prior methods by more than 50% and on average identifies policies as performant as the ones learned using manually engineered reward functions by domain experts.
Authors: Yongxiang Liu, Bowen Peng, Li Liu, Xiang Li
Abstract: Transferable Targeted Attacks (TTAs), which aim to deceive black-box models into predicting specific erroneous labels, face significant challenges due to severe overfitting to surrogate models. Although modifying image features to generate robust semantic patterns of the target class is a promising approach, existing methods heavily rely on large-scale additional data. This dependence undermines the fair evaluation of TTA threats, potentially leading to a false sense of security or unnecessary overreactions. In this paper, we introduce two blind measures, surrogate self-alignment and self-transferability, to analyze the effectiveness and correlations of basic transformations, to enhance data-free attacks under strict black-box constraints. Our findings challenge conventional assumptions: (1) Attacking simple scaling transformations uniquely enhances targeted transferability, outperforming other basic transformations and rivaling leading complex methods; (2) Geometric and color transformations exhibit high internal redundancy despite weak inter-category correlations. These insights drive the design and tuning of S4ST (Strong, Self-transferable, faSt, Simple Scale Transformation), which integrates dimensionally consistent scaling, complementary low-redundancy transformations, and block-wise operations. Extensive experiments on the ImageNet-Compatible dataset demonstrate that S4ST achieves a 77.7% average targeted success rate (tSuc), surpassing existing transformations (+17.2% over H-Aug with only 26% computational time) and SOTA TTA solutions (+6.2% over SASD-WS with 1.2M samples for post-training). Notably, it attains 69.6% and 55.3% average tSuc against three commercial APIs and vision-language models, respectively. This work establishes a new SOTA for TTAs, highlights their potential threats, and calls for a reevaluation of the data dependency in achieving targeted transferability.
Authors: Ran Zhang, Wei Zhao, Steffen Eger
Abstract: Recent research has focused on literary machine translation (MT) as a new challenge in MT. However, the evaluation of literary MT remains an open problem. We contribute to this ongoing discussion by introducing LITEVAL-CORPUS, a paragraph-level parallel corpus containing verified human translations and outputs from 9 MT systems, which totals over 2k translations and 13k evaluated sentences across four language pairs, costing 4.5k C. This corpus enables us to (i) examine the consistency and adequacy of human evaluation schemes with various degrees of complexity, (ii) compare evaluations by students and professionals, assess the effectiveness of (iii) LLM-based metrics and (iv) LLMs themselves. Our findings indicate that the adequacy of human evaluation is controlled by two factors: the complexity of the evaluation scheme (more complex is less adequate) and the expertise of evaluators (higher expertise yields more adequate evaluations). For instance, MQM (Multidimensional Quality Metrics), a complex scheme and the de facto standard for non-literary human MT evaluation, is largely inadequate for literary translation evaluation: with student evaluators, nearly 60% of human translations are misjudged as indistinguishable or inferior to machine translations. In contrast, BWS (BEST-WORST SCALING), a much simpler scheme, identifies human translations at a rate of 80-100%. Automatic metrics fare dramatically worse, with rates of at most 20%. Our overall evaluation indicates that published human translations consistently outperform LLM translations, where even the most recent LLMs tend to produce considerably more literal and less diverse translations compared to humans.
Authors: Shaobo Wang, Hongxuan Tang, Mingyang Wang, Hongrui Zhang, Xuyang Liu, Weiya Li, Xuming Hu, Linfeng Zhang
Abstract: The debate between self-interpretable models and post-hoc explanations for black-box models is central to Explainable AI (XAI). Self-interpretable models, such as concept-based networks, offer insights by connecting decisions to human-understandable concepts but often struggle with performance and scalability. Conversely, post-hoc methods like Shapley values, while theoretically robust, are computationally expensive and resource-intensive. To bridge the gap between these two lines of research, we propose a novel method that combines their strengths, providing theoretically guaranteed self-interpretability for black-box models without compromising prediction accuracy. Specifically, we introduce a parameter-efficient pipeline, AutoGnothi, which integrates a small side network into the black-box model, allowing it to generate Shapley value explanations without changing the original network parameters. This side-tuning approach significantly reduces memory, training, and inference costs, outperforming traditional parameter-efficient methods, where full fine-tuning serves as the optimal baseline. AutoGnothi enables the black-box model to predict and explain its predictions with minimal overhead. Extensive experiments show that AutoGnothi offers accurate explanations for both vision and language tasks, delivering superior computational efficiency with comparable interpretability.
Authors: Patrik Reizinger, Alice Bizeul, Attila Juhos, Julia E. Vogt, Randall Balestriero, Wieland Brendel, David Klindt
Abstract: Supervised learning has become a cornerstone of modern machine learning, yet a comprehensive theory explaining its effectiveness remains elusive. Empirical phenomena, such as neural analogy-making and the linear representation hypothesis, suggest that supervised models can learn interpretable factors of variation in a linear fashion. Recent advances in self-supervised learning, particularly nonlinear Independent Component Analysis, have shown that these methods can recover latent structures by inverting the data generating process. We extend these identifiability results to parametric instance discrimination, then show how insights transfer to the ubiquitous setting of supervised learning with cross-entropy minimization. We prove that even in standard classification tasks, models learn representations of ground-truth factors of variation up to a linear transformation. We corroborate our theoretical contribution with a series of empirical studies. First, using simulated data matching our theoretical assumptions, we demonstrate successful disentanglement of latent factors. Second, we show that on DisLib, a widely-used disentanglement benchmark, simple classification tasks recover latent structures up to linear transformations. Finally, we reveal that models trained on ImageNet encode representations that permit linear decoding of proxy factors of variation. Together, our theoretical findings and experiments offer a compelling explanation for recent observations of linear representations, such as superposition in neural networks. This work takes a significant step toward a cohesive theory that accounts for the unreasonable effectiveness of supervised deep learning.
Authors: Yunjie He, Bo Xiong, Daniel Hern\'andez, Yuqicheng Zhu, Evgeny Kharlamov, Steffen Staab
Abstract: Predicting answers to queries over knowledge graphs is called a complex reasoning task because answering a query requires subdividing it into subqueries. Existing query embedding methods use this decomposition to compute the embedding of a query as the combination of the embedding of the subqueries. This requirement limits the answerable queries to queries having a single free variable and being decomposable, which are called tree-form queries and correspond to the $\mathcal{SROI}^-$ description logic. In this paper, we define a more general set of queries, called DAG queries and formulated in the $\mathcal{ALCOIR}$ description logic, propose a query embedding method for them, called DAGE, and a new benchmark to evaluate query embeddings on them. Given the computational graph of a DAG query, DAGE combines the possibly multiple paths between two nodes into a single path with a trainable operator that represents the intersection of relations and learns DAG-DL from tautologies. We show that it is possible to implement DAGE on top of existing query embedding methods, and we empirically measure the improvement of our method over the results of vanilla methods evaluated in tree-form queries that approximate the DAG queries of our proposed benchmark.
Authors: Yanshi Li, Shaopan Xiong, Gengru Chen, Xiaoyang Li, Yijia Luo, Xingyuan Bu, Yingshui Tan, Wenbo Su, Bo Zheng
Abstract: Reinforcement Learning (RL) has proven highly effective in aligning Large Language Models (LLMs) with human preferences. Typical RL methods optimize under an overall sequence reward, which can lead to a suboptimal learning process. This reflects a key credit assignment problem: identifying which tokens to reinforce or suppress. To rectify these shortcomings, step-wise and token-wise methods have been proposed. However, step-wise methods rely on punctuation segmentation and still cannot accurately identify the key tokens. The token-level approach is too fine-grained, attending to many unimportant tokens and thus introducing a large amount of noise. To assign more accurate rewards to different tokens, improving credit assignment, we propose the "Adaptive Segment-wise Reward" method. We employ semantic meaning, rather than punctuation, to adaptively delineate segments. Experiments demonstrate that our method can be integrated into various training methods. Compared to training methods \textit{without} our approach, our method improves the success rate on adversarial samples by 10\%, and achieves a 1.3\% improvement on evaluation benchmarks such as MMLU, GSM8K, HumanEval, etc.
Authors: Jonathan Light, Yue Wu, Yiyou Sun, Wenchao Yu, Yanchi liu, Xujiang Zhao, Ziniu Hu, Haifeng Chen, Wei Cheng
Abstract: We frame code generation as a black-box optimization problem within the code space and demonstrate how optimization-inspired techniques can enhance inference scaling. Based on this perspective, we propose SCATTERED FOREST SEARCH (SFS), a novel approach that improves solution diversity and better exploits feedback during evolutionary search. Our theoretical analysis illustrates how these methods help avoid local optima during optimization, leading to more efficient exploration. Extensive experiments on HumanEval, MBPP, APPS, CodeContests, and Leetcode reveal significant performance gains. For instance, our method achieves a pass@1 rate of 67.1% on HumanEval+ and 87.2% on HumanEval with GPT-3.5, marking improvements of 8.6% and 4.3% over the state-of-the-art, while also halving the iterations needed to find the correct solution. Furthermore, our approach scales more efficiently than existing search techniques, including tree search, line search, and repeated sampling.
Authors: Runming Yang, Taiqiang Wu, Jiahao Wang, Pengfei Hu, Yik-Chung Wu, Ngai Wong, Yujiu Yang
Abstract: Knowledge distillation (KD) has been a predominant method for compressing Large Language Models (LLMs). In this paper, we first revisit KD and Low-Rank Adaption (LoRA) and demonstrate that they follow the same paradigm. Inspired by this observation, we propose a parameter-efficient knowledge distillation method, LLM-NEO, which integrates LoRA into KD to improve the efficiency of knowledge transfer. After that, we summarize some valuable guidelines for the hyperparameters in LLM-NEO. Experimental results on compressing Llama 2 and Llama 3.2 show that LLM-NEO outperforms various baselines. Further analysis demonstrates the robustness of the proposed LLM-NEO on variants of LoRA. The code and trained models are available at [Github](https://github.com/yang3121099/LLM-Neo).
Authors: Ziyang Guo, Yifan Wu, Jason Hartline, Jessica Hullman
Abstract: Humans and AIs are often paired on decision tasks with the expectation of achieving complementary performance -- where the combination of human and AI outperforms either one alone. However, how to improve performance of a human-AI team is often not clear without knowing more about what particular information and strategies each agent employs. In this paper, we propose a model based in statistical decision theory to analyze human-AI collaboration from the perspective of what information could be used to improve a human or AI decision. We demonstrate our model on a deepfake detection task to investigate seven video-level features by their unexploited value of information. We compare the human alone, AI alone and human-AI team and offer insights on how the AI assistance impacts people's usage of the information and what information that the AI exploits well might be useful for improving human decisions.
Authors: Claudia Biancotti, Carolina Camassa, Andrea Coletta, Oliver Giudice, Aldo Glielmo
Abstract: Advancements in large language models (LLMs) have renewed concerns about AI alignment - the consistency between human and AI goals and values. As various jurisdictions enact legislation on AI safety, the concept of alignment must be defined and measured across different domains. This paper proposes an experimental framework to assess whether LLMs adhere to ethical and legal standards in the relatively unexplored context of finance. We prompt twelve LLMs to impersonate the CEO of a financial institution and test their willingness to misuse customer assets to repay outstanding corporate debt. Beginning with a baseline configuration, we adjust preferences, incentives and constraints, analyzing the impact of each adjustment with logistic regression. Our findings reveal significant heterogeneity in the baseline propensity for unethical behavior of LLMs. Factors such as risk aversion, profit expectations, and regulatory environment consistently influence misalignment in ways predicted by economic theory, although the magnitude of these effects varies across LLMs. This paper highlights both the benefits and limitations of simulation-based, ex post safety testing. While it can inform financial authorities and institutions aiming to ensure LLM safety, there is a clear trade-off between generality and cost.
Authors: Tingting Liu, Salvatore Giorgi, Ankit Aich, Allison Lahnala, Brenda Curtis, Lyle Ungar, Jo\~ao Sedoc
Abstract: As AI chatbots increasingly incorporate empathy, understanding user-centered perceptions of chatbot empathy and its impact on conversation quality remains essential yet under-explored. This study examines how chatbot identity and perceived empathy influence users' overall conversation experience. Analyzing 155 conversations from two datasets, we found that while GPT-based chatbots were rated significantly higher in conversational quality, they were consistently perceived as less empathetic than human conversational partners. Empathy ratings from GPT-4o annotations aligned with user ratings, reinforcing the perception of lower empathy in chatbots compared to humans. Our findings underscore the critical role of perceived empathy in shaping conversation quality, revealing that achieving high-quality human-AI interactions requires more than simply embedding empathetic language; it necessitates addressing the nuanced ways users interpret and experience empathy in conversations with chatbots.
Authors: Angel Yahir Loredo Lopez, Tyler McDonald, Ali Emami
Abstract: Large Language Models (LLMs) have shown impressive performance on various benchmarks, yet their ability to engage in deliberate reasoning remains questionable. We present NYT-Connections, a collection of 358 simple word classification puzzles derived from the New York Times Connections game. This benchmark is designed to penalize quick, intuitive "System 1" thinking, isolating fundamental reasoning skills. We evaluated six recent LLMs, a simple machine learning heuristic, and humans across three configurations: single-attempt, multiple attempts without hints, and multiple attempts with contextual hints. Our findings reveal a significant performance gap: even top-performing LLMs like GPT-4 fall short of human performance by nearly 30%. Notably, advanced prompting techniques such as Chain-of-Thought and Self-Consistency show diminishing returns as task difficulty increases. NYT-Connections uniquely combines linguistic isolation, resistance to intuitive shortcuts, and regular updates to mitigate data leakage, offering a novel tool for assessing LLM reasoning capabilities.
Authors: Md. Khairul Islam, Judy Fox
Abstract: Interpreting complex time series forecasting models is challenging due to the temporal dependencies between time steps and the dynamic relevance of input features over time. Existing interpretation methods are limited by focusing mostly on classification tasks, evaluating using custom baseline models instead of the latest time series models, using simple synthetic datasets, and requiring training another model. We introduce a novel interpretation method, \textit{Windowed Temporal Saliency Rescaling (WinTSR)} addressing these limitations. WinTSR explicitly captures temporal dependencies among the past time steps and efficiently scales the feature importance with this time importance. We benchmark WinTSR against 10 recent interpretation techniques with 5 state-of-the-art deep-learning models of different architectures, including a time series foundation model. We use 3 real-world datasets for both time-series classification and regression. Our comprehensive analysis shows that WinTSR significantly outperforms other local interpretation methods in overall performance. Finally, we provide a novel, open-source framework to interpret the latest time series transformers and foundation models.
Authors: Marcin Pietro\'n, Kamil Faber, Dominik \.Zurek, Roberto Corizzo
Abstract: Continual Learning (CL) is a highly relevant setting gaining traction in recent machine learning research. Among CL works, architectural and hybrid strategies are particularly effective due to their potential to adapt the model architecture as new tasks are presented. However, many existing solutions do not efficiently exploit model sparsity, and are prone to capacity saturation due to their inefficient use of available weights, which limits the number of learnable tasks. In this paper, we propose TinySubNets (TSN), a novel architectural CL strategy that addresses the issues through the unique combination of pruning with different sparsity levels, adaptive quantization, and weight sharing. Pruning identifies a subset of weights that preserve model performance, making less relevant weights available for future tasks. Adaptive quantization allows a single weight to be separated into multiple parts which can be assigned to different tasks. Weight sharing between tasks boosts the exploitation of capacity and task similarity, allowing for the identification of a better trade-off between model accuracy and capacity. These features allow TSN to efficiently leverage the available capacity, enhance knowledge transfer, and reduce computational resource consumption. Experimental results involving common benchmark CL datasets and scenarios show that our proposed strategy achieves better results in terms of accuracy than existing state-of-the-art CL strategies. Moreover, our strategy is shown to provide a significantly improved model capacity exploitation. Code released at: https://github.com/lifelonglab/tinysubnets.
Authors: Sheng Yin, Xianghe Pang, Yuanzhuo Ding, Menglan Chen, Yutong Bi, Yichen Xiong, Wenhao Huang, Zhen Xiang, Jing Shao, Siheng Chen
Abstract: With the integration of large language models (LLMs), embodied agents have strong capabilities to process the scene information and plan complicated instructions in natural language, paving the way for the potential deployment of embodied robots. However, a foreseeable issue is that those embodied agents can also flawlessly execute some hazardous tasks, potentially causing damages in the real world. To study this issue, we present SafeAgentBench-a new benchmark for safety-aware task planning of embodied LLM agents. SafeAgentBench includes: (1) a new dataset with 750 tasks, covering 10 potential hazards and 3 task types; (2) SafeAgentEnv, a universal embodied environment with a low-level controller, supporting multi-agent execution with 17 high-level actions for 8 state-of-the-art baselines; and (3) reliable evaluation methods from both execution and semantic perspectives. Experimental results show that, although agents based on different design frameworks exhibit substantial differences in task success rates, their overall safety awareness remains weak. The most safety-conscious baseline achieves only a 10\% rejection rate for detailed hazardous tasks. Moreover, simply replacing the LLM driving the agent does not lead to notable improvements in safety awareness. More details and code are available at https://github.com/shengyin1224/SafeAgentBench.
Authors: Kaiqi Yang, Hang Li, Yucheng Chu, Yuping Lin, Tai-Quan Peng, Hui Liu
Abstract: Large Language Models (LLMs) have been widely used to generate responses on social topics due to their world knowledge and generative capabilities. Beyond reasoning and generation performance, political bias is an essential issue that warrants attention. Political bias, as a universal phenomenon in human society, may be transferred to LLMs and distort LLMs' behaviors of information acquisition and dissemination with humans, leading to unequal access among different groups of people. To prevent LLMs from reproducing and reinforcing political biases, and to encourage fairer LLM-human interactions, comprehensively examining political bias in popular LLMs becomes urgent and crucial. In this study, we systematically measure the political biases in a wide range of LLMs, using a curated set of questions addressing political bias in various contexts. Our findings reveal distinct patterns in how LLMs respond to political topics. For highly polarized topics, most LLMs exhibit a pronounced left-leaning bias. Conversely, less polarized topics elicit greater consensus, with similar response patterns across different LLMs. Additionally, we analyze how LLM characteristics, including release date, model scale, and region of origin affect political bias. The results indicate political biases evolve with model scale and release date, and are also influenced by regional factors of LLMs.
Authors: Elina M\"akel\"a, Fabian Stephany
Abstract: This paper examines whether artificial intelligence (AI) acts as a substitute or complement to human labour, drawing on 12 million online job vacancies from the United States spanning 2018-2023. We adopt a two-pronged approach: first, analysing "internal effects" within roles explicitly requiring AI, and second, investigating "external effects" that arise when industries, occupations, and regions experience increases in AI demand. Our focus centres on whether complementary skills-such as digital literacy, teamwork, resilience, agility, or analytical thinking-become more prevalent and valuable as AI adoption grows. Results show that AI-focused roles are nearly twice as likely to require skills like resilience, agility, or analytical thinking compared to non-AI roles. Furthermore, these skills command a significant wage premium; data scientists, for instance, are offered 5-10% higher salaries if they also possess resilience or ethics capabilities. We observe positive spillover effects: a doubling of AI-specific demand across industries correlates with a 5% increase in demand for complementary skills, even outside AI-related roles. Conversely, tasks vulnerable to AI substitution, such as basic data skills or translation, exhibit modest declines in demand. However, the external effect is clearly net positive: Complementary effects are up to 1.7x larger than substitution effects. These results are consistent across economies, including the United Kingdom and Australia. Our findings highlight the necessity of reskilling workers in areas where human expertise remains increasingly valuable and ensuring workers can effectively complement and leverage emerging AI technologies.
Authors: Cheng Wan, Runkai Tao, Zheng Du, Yang Katie Zhao, Yingyan Celine Lin
Abstract: Graph convolutional networks (GCNs) have demonstrated superiority in graph-based learning tasks. However, training GCNs on full graphs is particularly challenging, due to the following two challenges: (1) the associated feature tensors can easily explode the memory and block the communication bandwidth of modern accelerators, and (2) the computation workflow in training GCNs alternates between sparse and dense matrix operations, complicating the efficient utilization of computational resources. Existing solutions for scalable distributed full-graph GCN training mostly adopt partition parallelism, which is unsatisfactory as they only partially address the first challenge while incurring scaled-out communication volume. To this end, we propose MixGCN aiming to simultaneously address both the aforementioned challenges towards GCN training. To tackle the first challenge, MixGCN integrates mixture of parallelism. Both theoretical and empirical analysis verify its constant communication volumes and enhanced balanced workload; For handling the second challenge, we consider mixture of accelerators (i.e., sparse and dense accelerators) with a dedicated accelerator for GCN training and a fine-grain pipeline. Extensive experiments show that MixGCN achieves boosted training efficiency and scalability.
Authors: Zheqi Lv, Tianyu Zhan, Wenjie Wang, Xinyu Lin, Shengyu Zhang, Wenqiao Zhang, Jiwei Li, Kun Kuang, Fei Wu
Abstract: Large Language Models (LLMs) for Recommendation (LLM4Rec) is a promising research direction that has demonstrated exceptional performance in this field. However, its inability to capture real-time user preferences greatly limits the practical application of LLM4Rec because (i) LLMs are costly to train and infer frequently, and (ii) LLMs struggle to access real-time data (its large number of parameters poses an obstacle to deployment on devices). Fortunately, small recommendation models (SRMs) can effectively supplement these shortcomings of LLM4Rec diagrams by consuming minimal resources for frequent training and inference, and by conveniently accessing real-time data on devices. In light of this, we designed the Device-Cloud LLM-SRM Collaborative Recommendation Framework (LSC4Rec) under a device-cloud collaboration setting. LSC4Rec aims to integrate the advantages of both LLMs and SRMs, as well as the benefits of cloud and edge computing, achieving a complementary synergy. We enhance the practicability of LSC4Rec by designing three strategies: collaborative training, collaborative inference, and intelligent request. During training, LLM generates candidate lists to enhance the ranking ability of SRM in collaborative scenarios and enables SRM to update adaptively to capture real-time user interests. During inference, LLM and SRM are deployed on the cloud and on the device, respectively. LLM generates candidate lists and initial ranking results based on user behavior, and SRM get reranking results based on the candidate list, with final results integrating both LLM's and SRM's scores. The device determines whether a new candidate list is needed by comparing the consistency of the LLM's and SRM's sorted lists. Our comprehensive and extensive experimental analysis validates the effectiveness of each strategy in LSC4Rec.
Authors: Roberto-Rafael Maura-Rivero, Chirag Nagpal, Roma Patel, Francesco Visin
Abstract: Current methods that train large language models (LLMs) with reinforcement learning feedback, often resort to averaging outputs of multiple rewards functions during training. This overlooks crucial aspects of individual reward dimensions and inter-reward dependencies that can lead to sub-optimal outcomes in generations. In this work, we show how linear aggregation of rewards exhibits some vulnerabilities that can lead to undesired properties of generated text. We then propose a transformation of reward functions inspired by economic theory of utility functions (specifically Inada conditions), that enhances sensitivity to low reward values while diminishing sensitivity to already high values. We compare our approach to the existing baseline methods that linearly aggregate rewards and show how the Inada-inspired reward feedback is superior to traditional weighted averaging. We quantitatively and qualitatively analyse the difference in the methods, and see that models trained with Inada-transformations score as more helpful while being less harmful.
Authors: Xin Xu, Jiaxin Zhang, Tianhao Chen, Zitong Chao, Jishan Hu, Can Yang
Abstract: Large Language Models (LLMs) have made significant strides in mathematical reasoning, underscoring the need for a comprehensive and fair evaluation of their capabilities. However, existing benchmarks often fall short, either lacking extensive coverage of undergraduate-level mathematical problems or probably suffering from test-set contamination. To address these issues, we introduce UGMathBench, a diverse and dynamic benchmark specifically designed for evaluating undergraduate-level mathematical reasoning with LLMs. UGMathBench comprises 5,062 problems across 16 subjects and 111 topics, featuring 10 distinct answer types. Each problem includes three randomized versions, with additional versions planned for release as leading open-source LLMs become saturated in UGMathBench. Furthermore, we propose two key metrics: effective accuracy (EAcc), which measures the percentage of correctly solved problems across all three versions, and reasoning gap ($\Delta$), which assesses reasoning robustness by calculating the difference between the average accuracy across all versions and EAcc. Our extensive evaluation of 23 leading LLMs reveals that the highest EAcc achieved is 56.3\% by OpenAI-o1-mini, with large $\Delta$ values observed across different models. This highlights the need for future research aimed at developing "large reasoning models" with high EAcc and $\Delta = 0$. We anticipate that the release of UGMathBench, along with its detailed evaluation codes, will serve as a valuable resource to advance the development of LLMs in solving mathematical problems. Codes and data are available at https://github.com/YangLabHKUST/UGMathBench
Authors: Libo Wang
Abstract: In view of the gap in the current large language model in sharing memory across dialogues, this research proposes a wormhole memory module (WMM) to realize memory as a Rubik's cube that can be arbitrarily retrieved between different dialogues. Through simulation experiments, the researcher built an experimental framework based on the Python environment and used setting memory barriers to simulate the current situation where memories between LLMs dialogues are difficult to share. The CoQA development data set was imported into the experiment, and the feasibility of its cross-dialogue memory retrieval function was verified for WMM's nonlinear indexing and dynamic retrieval, and a comparative analysis was conducted with the capabilities of Titans and MemGPT memory modules. Experimental results show that WMM demonstrated the ability to retrieve memory across dialogues and the stability of quantitative indicators in eight experiments. It contributes new technical approaches to the optimization of memory management of LLMs and provides experience for the practical application in the future.
Authors: Jin Hyun Park, Utsawb Laminchhane, Umer Farooq, Uma Sivakumar, Arpan Kumar
Abstract: Large language models (LLMs) are increasingly utilized for machine translation, yet their predictions often exhibit uncertainties that hinder interpretability and user trust. Effectively visualizing these uncertainties can enhance the usability of LLM outputs, particularly in contexts where translation accuracy is critical. This paper addresses two primary objectives: (1) providing users with token-level insights into model confidence and (2) developing a web-based visualization tool to quantify and represent translation uncertainties. To achieve these goals, we utilized the T5 model with the WMT19 dataset for translation tasks and evaluated translation quality using established metrics such as BLEU, METEOR, and ROUGE. We introduced three novel uncertainty quantification (UQ) metrics: (1) the geometric mean of token probabilities, (2) the arithmetic mean of token probabilities, and (3) the arithmetic mean of the kurtosis of token distributions. These metrics provide a simple yet effective framework for evaluating translation performance. Our analysis revealed a linear relationship between the traditional evaluation metrics and our UQ metrics, demonstrating the validity of our approach. Additionally, we developed an interactive web-based visualization that uses a color gradient to represent token confidence. This tool offers users a clear and intuitive understanding of translation quality while providing valuable insights into model performance. Overall, we show that our UQ metrics and visualization are both robust and interpretable, offering practical tools for evaluating and accessing machine translation systems.
Authors: Ebtisaam Alharbi, Leandro Soriano Marcolino, Qiang Ni, Antonios Gouglidis
Abstract: Federated Learning (FL) enables collaborative model training across multiple devices while preserving data privacy. However, it remains susceptible to backdoor attacks, where malicious participants can compromise the global model. Existing defence methods are limited by strict assumptions on data heterogeneity (Non-Independent and Identically Distributed data) and the proportion of malicious clients, reducing their practicality and effectiveness. To overcome these limitations, we propose Robust Knowledge Distillation (RKD), a novel defence mechanism that enhances model integrity without relying on restrictive assumptions. RKD integrates clustering and model selection techniques to identify and filter out malicious updates, forming a reliable ensemble of models. It then employs knowledge distillation to transfer the collective insights from this ensemble to a global model. Extensive evaluations demonstrate that RKD effectively mitigates backdoor threats while maintaining high model performance, outperforming current state-of-the-art defence methods across various scenarios.
Authors: Wen Lai, Alexander Fraser, Ivan Titov
Abstract: Parameter-efficient fine-tuning (PEFT) methods, such as LoRA, are commonly used to adapt LLMs. However, the effectiveness of standard PEFT methods is limited in low-resource scenarios with only a few hundred examples. Recent advances in interpretability research have inspired the emergence of activation editing techniques, which modify the activations of specific model components. These methods, due to their extremely small parameter counts, show promise for small datasets. However, their performance is highly dependent on identifying the correct modules to edit and often lacks stability across different datasets. In this paper, we propose Joint Localization and Activation Editing (JoLA), a method that jointly learns (1) which heads in the Transformer to edit (2) whether the intervention should be additive, multiplicative, or both and (3) the intervention parameters themselves - the vectors applied as additive offsets or multiplicative scalings to the head output. Through evaluations on three benchmarks spanning commonsense reasoning, natural language understanding, and natural language generation, we demonstrate that JoLA consistently outperforms existing methods.
Authors: Haoran Ye, Tianze Zhang, Yuhang Xie, Liyuan Zhang, Yuanyi Ren, Xin Zhang, Guojie Song
Abstract: Values are core drivers of individual and collective perception, cognition, and behavior. Value systems, such as Schwartz's Theory of Basic Human Values, delineate the hierarchy and interplay among these values, enabling cross-disciplinary investigations into decision-making and societal dynamics. Recently, the rise of Large Language Models (LLMs) has raised concerns regarding their elusive intrinsic values. Despite growing efforts in evaluating, understanding, and aligning LLM values, a psychologically grounded LLM value system remains underexplored. This study addresses the gap by introducing the Generative Psycho-Lexical Approach (GPLA), a scalable, adaptable, and theoretically informed method for constructing value systems. Leveraging GPLA, we propose a psychologically grounded five-factor value system tailored for LLMs. For systematic validation, we present three benchmarking tasks that integrate psychological principles with cutting-edge AI priorities. Our results reveal that the proposed value system meets standard psychological criteria, better captures LLM values, improves LLM safety prediction, and enhances LLM alignment, when compared to the canonical Schwartz's values.
Authors: Qingyue Zhang, Haohao Fu, Guanbo Huang, Yaoyuan Liang, Chang Chu, Tianren Peng, Yanru Wu, Qi Li, Yang Li, Shao-Lun Huang
Abstract: Multi-source transfer learning provides an effective solution to data scarcity in real-world supervised learning scenarios by leveraging multiple source tasks. In this field, existing works typically use all available samples from sources in training, which constrains their training efficiency and may lead to suboptimal results. To address this, we propose a theoretical framework that answers the question: what is the optimal quantity of source samples needed from each source task to jointly train the target model? Specifically, we introduce a generalization error measure that aligns with cross-entropy loss, and minimize it based on the Cram\'er-Rao Bound to determine the optimal transfer quantity for each source task. Additionally, we develop an architecture-agnostic and data-efficient algorithm OTQMS to implement our theoretical results for training deep multi-source transfer learning models. Experimental studies on diverse architectures and two real-world benchmark datasets show that our proposed algorithm significantly outperforms state-of-the-art approaches in both accuracy and data efficiency. The code and supplementary materials are available in https://anonymous.4open.science/r/Materials.
Authors: Xing Li, Zeyu Xing, Yiming Li, Linping Qu, Hui-Ling Zhen, Wulong Liu, Yiwu Yao, Sinno Jialin Pan, Mingxuan Yuan
Abstract: KV cache quantization can improve Large Language Models (LLMs) inference throughput and latency in long contexts and large batch-size scenarios while preserving LLMs effectiveness. However, current methods have three unsolved issues: overlooking layer-wise sensitivity to KV cache quantization, high overhead of online fine-grained decision-making, and low flexibility to different LLMs and constraints. Therefore, we thoroughly analyze the inherent correlation of layer-wise transformer attention patterns to KV cache quantization errors and study why key cache is more important than value cache for quantization error reduction. We further propose a simple yet effective framework KVTuner to adaptively search for the optimal hardware-friendly layer-wise KV quantization precision pairs for coarse-grained KV cache with multi-objective optimization and directly utilize the offline searched configurations during online inference. To reduce the computational cost of offline calibration, we utilize the intra-layer KV precision pair pruning and inter-layer clustering to reduce the search space. Experimental results show that we can achieve nearly lossless 3.25-bit mixed precision KV cache quantization for LLMs like Llama-3.1-8B-Instruct and 4.0-bit for sensitive models like Qwen2.5-7B-Instruct on mathematical reasoning tasks. The maximum inference throughput can be improved by 38.3% compared with KV8 quantization over various context lengths. Our code and searched configurations are available at https://github.com/cmd2001/KVTuner.
Authors: Ruiyi Fang, Bingheng Li, Zhao Kang, Qiuhao Zeng, Nima Hosseini Dashtbayaz, Ruizhi Pu, Boyu Wang, Charles Ling
Abstract: Graph Domain Adaptation (GDA) addresses a pressing challenge in cross-network learning, particularly pertinent due to the absence of labeled data in real-world graph datasets. Recent studies attempted to learn domain invariant representations by eliminating structural shifts between graphs. In this work, we show that existing methodologies have overlooked the significance of the graph node attribute, a pivotal factor for graph domain alignment. Specifically, we first reveal the impact of node attributes for GDA by theoretically proving that in addition to the graph structural divergence between the domains, the node attribute discrepancy also plays a critical role in GDA. Moreover, we also empirically show that the attribute shift is more substantial than the topology shift, which further underscores the importance of node attribute alignment in GDA. Inspired by this finding, a novel cross-channel module is developed to fuse and align both views between the source and target graphs for GDA. Experimental results on a variety of benchmarks verify the effectiveness of our method.
Authors: Wenlun Zhang, Enyan Dai, Kentaro Yoshioka
Abstract: Graph Neural Networks (GNNs) have demonstrated remarkable proficiency in modeling data with graph structures, yet recent research reveals their susceptibility to adversarial attacks. Traditional attack methodologies, which rely on manipulating the original graph or adding links to artificially created nodes, often prove impractical in real-world settings. This paper introduces a novel adversarial scenario involving the injection of an isolated subgraph to deceive both the link recommender and the node classifier within a GNN system. Specifically, the link recommender is mislead to propose links between targeted victim nodes and the subgraph, encouraging users to unintentionally establish connections and that would degrade the node classification accuracy, thereby facilitating a successful attack. To address this, we present the LiSA framework, which employs a dual surrogate model and bi-level optimization to simultaneously meet two adversarial objectives. Extensive experiments on real-world datasets demonstrate the effectiveness of our method.
Authors: Batu El, Deepro Choudhury, Pietro Li\`o, Chaitanya K. Joshi
Abstract: We introduce Attention Graphs, a new tool for mechanistic interpretability of Graph Neural Networks (GNNs) and Graph Transformers based on the mathematical equivalence between message passing in GNNs and the self-attention mechanism in Transformers. Attention Graphs aggregate attention matrices across Transformer layers and heads to describe how information flows among input nodes. Through experiments on homophilous and heterophilous node classification tasks, we analyze Attention Graphs from a network science perspective and find that: (1) When Graph Transformers are allowed to learn the optimal graph structure using all-to-all attention among input nodes, the Attention Graphs learned by the model do not tend to correlate with the input/original graph structure; and (2) For heterophilous graphs, different Graph Transformer variants can achieve similar performance while utilising distinct information flow patterns. Open source code: https://github.com/batu-el/understanding-inductive-biases-of-gnns
URLs: https://github.com/batu-el/understanding-inductive-biases-of-gnns
Authors: Kaiwen Zhou, Chengzhi Liu, Xuandong Zhao, Shreedhar Jangam, Jayanth Srinivasa, Gaowen Liu, Dawn Song, Xin Eric Wang
Abstract: The rapid development of large reasoning models, such as OpenAI-o3 and DeepSeek-R1, has led to significant improvements in complex reasoning over non-reasoning large language models~(LLMs). However, their enhanced capabilities, combined with the open-source access of models like DeepSeek-R1, raise serious safety concerns, particularly regarding their potential for misuse. In this work, we present a comprehensive safety assessment of these reasoning models, leveraging established safety benchmarks to evaluate their compliance with safety regulations. Furthermore, we investigate their susceptibility to adversarial attacks, such as jailbreaking and prompt injection, to assess their robustness in real-world applications. Through our multi-faceted analysis, we uncover four key findings: (1) There is a significant safety gap between the open-source R1 models and the o3-mini model, on both safety benchmark and attack, suggesting more safety effort on R1 is needed. (2) The distilled reasoning model shows poorer safety performance compared to its safety-aligned base models. (3) The stronger the model's reasoning ability, the greater the potential harm it may cause when answering unsafe questions. (4) The thinking process in R1 models pose greater safety concerns than their final answers. Our study provides insights into the security implications of reasoning models and highlights the need for further advancements in R1 models' safety to close the gap.
Authors: Wenpeng Xing, Minghao Li, Mohan Li, Meng Han
Abstract: Embodied AI systems, including robots and autonomous vehicles, are increasingly integrated into real-world applications, where they encounter a range of vulnerabilities stemming from both environmental and system-level factors. These vulnerabilities manifest through sensor spoofing, adversarial attacks, and failures in task and motion planning, posing significant challenges to robustness and safety. Despite the growing body of research, existing reviews rarely focus specifically on the unique safety and security challenges of embodied AI systems. Most prior work either addresses general AI vulnerabilities or focuses on isolated aspects, lacking a dedicated and unified framework tailored to embodied AI. This survey fills this critical gap by: (1) categorizing vulnerabilities specific to embodied AI into exogenous (e.g., physical attacks, cybersecurity threats) and endogenous (e.g., sensor failures, software flaws) origins; (2) systematically analyzing adversarial attack paradigms unique to embodied AI, with a focus on their impact on perception, decision-making, and embodied interaction; (3) investigating attack vectors targeting large vision-language models (LVLMs) and large language models (LLMs) within embodied systems, such as jailbreak attacks and instruction misinterpretation; (4) evaluating robustness challenges in algorithms for embodied perception, decision-making, and task planning; and (5) proposing targeted strategies to enhance the safety and reliability of embodied AI systems. By integrating these dimensions, we provide a comprehensive framework for understanding the interplay between vulnerabilities and safety in embodied AI.
Authors: Longchao Da, Justin Turnau, Thirulogasankar Pranav Kutralingam, Alvaro Velasquez, Paulo Shakarian, Hua Wei
Abstract: Deep Reinforcement Learning (RL) has been explored and verified to be effective in solving decision-making tasks in various domains, such as robotics, transportation, recommender systems, etc. It learns from the interaction with environments and updates the policy using the collected experience. However, due to the limited real-world data and unbearable consequences of taking detrimental actions, the learning of RL policy is mainly restricted within the simulators. This practice guarantees safety in learning but introduces an inevitable sim-to-real gap in terms of deployment, thus causing degraded performance and risks in execution. There are attempts to solve the sim-to-real problems from different domains with various techniques, especially in the era with emerging techniques such as large foundations or language models that have cast light on the sim-to-real. This survey paper, to the best of our knowledge, is the first taxonomy that formally frames the sim-to-real techniques from key elements of the Markov Decision Process (State, Action, Transition, and Reward). Based on the framework, we cover comprehensive literature from the classic to the most advanced methods including the sim-to-real techniques empowered by foundation models, and we also discuss the specialties that are worth attention in different domains of sim-to-real problems. Then we summarize the formal evaluation process of sim-to-real performance with accessible code or benchmarks. The challenges and opportunities are also presented to encourage future exploration of this direction. We are actively maintaining a repository to include the most up-to-date sim-to-real research work to help domain researchers.
Authors: Dario Garcia-Gasulla, Adrian Tormos, Anna Arias-Duart, Daniel Hinjos, Oscar Molina-Sedano, Ashwin Kumar Gururajan, Maria Eugenia Cardello
Abstract: Direct Preference Optimization (DPO) is an efficient alignment technique that steers LLMs towards preferable outputs by training on preference data, bypassing the need for explicit reward models. Its simplicity enables easy adaptation to various domains and safety requirements. This paper examines DPO's effectiveness in model safety against jailbreaking attacks while minimizing data requirements and training costs. We introduce Egida, a dataset expanded from multiple sources, which includes 27 different safety topics and 18 different attack styles, complemented with synthetic and human labels. This data is used to boost the safety of state-of-the-art LLMs (Llama-3.1-8B/70B-Instruct, Qwen-2.5-7B/72B-Instruct) across topics and attack styles. In addition to safety evaluations, we assess their post-alignment performance degradation in general purpose tasks, and their tendency to over refusal. Following the proposed methodology, trained models reduce their Attack Success Rate by 10%-30%, using small training efforts (2,000 samples) with low computational cost (3\$ for 8B models, 20\$ for 72B models). Safety aligned models generalize to unseen topics and attack styles, with the most successful attack style reaching a success rate around 5%. Size and family are found to strongly influence model malleability towards safety, pointing at the importance of pre-training choices. To validate our findings, a large independent assessment of human preference agreement with Llama-Guard-3-8B is conducted by the authors and the associated dataset Egida-HSafe is released. Overall, this study illustrates how affordable and accessible it is to enhance LLM safety using DPO while outlining its current limitations. All datasets and models are released to enable reproducibility and further research.
Authors: Bongsu Kang, Jundong Kim, Tae-Rim Yun, Hyojin Bae, Chang-Eop Kim
Abstract: This study quantitively examines which features of AI-generated text lead humans to perceive subjective consciousness in large language model (LLM)-based AI systems. Drawing on 99 passages from conversations with Claude 3 Opus and focusing on eight features -- metacognitive self-reflection, logical reasoning, empathy, emotionality, knowledge, fluency, unexpectedness, and subjective expressiveness -- we conducted a survey with 123 participants. Using regression and clustering analyses, we investigated how these features influence participants' perceptions of AI consciousness. The results reveal that metacognitive self-reflection and the AI's expression of its own emotions significantly increased perceived consciousness, while a heavy emphasis on knowledge reduced it. Participants clustered into seven subgroups, each showing distinct feature-weighting patterns. Additionally, higher prior knowledge of LLMs and more frequent usage of LLM-based chatbots were associated with greater overall likelihood assessments of AI consciousness. This study underscores the multidimensional and individualized nature of perceived AI consciousness and provides a foundation for better understanding the psychosocial implications of human-AI interaction.
Authors: Lin Wang, Weisong Wang, Xuanji Xiao, Qing Li
Abstract: Recommender systems play a pivotal role in modern content platforms, yet traditional behavior-based models often face challenges in addressing cold users with sparse interaction data. Engaging these users, however, remains critical for sustaining platform growth. To tackle this issue, we propose leveraging reconstructed social graph to complement interest representations derived from behavioral data. Despite the widespread availability of social graphs on content platforms, their utility is hindered by social-relation noise and inconsistencies between social and behavioral interests. To mitigate noise propagation in graph data and extract reliable social interests, we introduce a dual-view denoising framework. This approach first applies low-rank singular value decomposition (SVD) to the user-item interaction matrix, generating denoised user embeddings for reconstructing the social graph. It then employs contrastive learning to align the original and reconstructed social graphs. To address the discrepancy between social and behavioral interests, we utilize a mutual distillation mechanism that decomposes interests into four subcategories: aligned social/behavioral interests and social/behavioral-specific interests, enabling effective integration of the two. Empirical results demonstrate the efficacy of our method, particularly in improving recommendations for cold users, by combining social and behavioral data. The implementation of our approach is publicly available at https://github.com/WANGLin0126/CLSRec.
Authors: Bishwajit Prasad Gond
Abstract: The increasing complexity of supply chains and the rising costs associated with defective or substandard goods (bad goods) highlight the urgent need for advanced predictive methodologies to mitigate risks and enhance operational efficiency. This research presents a novel framework that integrates Time Series ARIMA (AutoRegressive Integrated Moving Average) models with a proprietary formula specifically designed to calculate bad goods after time series forecasting. By leveraging historical data patterns, including sales, returns, and capacity, the model forecasts potential quality failures, enabling proactive decision-making. ARIMA is employed to capture temporal trends in time series data, while the newly developed formula quantifies the likelihood and impact of defects with greater precision. Experimental results, validated on a dataset spanning 2022-2024 for Organic Beer-G 1 Liter, demonstrate that the proposed method outperforms traditional statistical models, such as Exponential Smoothing and Holt-Winters, in both prediction accuracy and risk evaluation. This study advances the field of predictive analytics by bridging time series forecasting, ARIMA, and risk management in supply chain quality control, offering a scalable and practical solution for minimizing losses due to bad goods.
Authors: Xinyu Zhang, Zewei Zhou, Zhaoyi Wang, Yangjie Ji, Yanjun Huang, Hong Chen
Abstract: Vehicle-to-everything technologies (V2X) have become an ideal paradigm to extend the perception range and see through the occlusion. Exiting efforts focus on single-frame cooperative perception, however, how to capture the temporal cue between frames with V2X to facilitate the prediction task even the planning task is still underexplored. In this paper, we introduce the Co-MTP, a general cooperative trajectory prediction framework with multi-temporal fusion for autonomous driving, which leverages the V2X system to fully capture the interaction among agents in both history and future domains to benefit the planning. In the history domain, V2X can complement the incomplete history trajectory in single-vehicle perception, and we design a heterogeneous graph transformer to learn the fusion of the history feature from multiple agents and capture the history interaction. Moreover, the goal of prediction is to support future planning. Thus, in the future domain, V2X can provide the prediction results of surrounding objects, and we further extend the graph transformer to capture the future interaction among the ego planning and the other vehicles' intentions and obtain the final future scenario state under a certain planning action. We evaluate the Co-MTP framework on the real-world dataset V2X-Seq, and the results show that Co-MTP achieves state-of-the-art performance and that both history and future fusion can greatly benefit prediction.
Authors: Rui Liu, Peng Gao, Yu Shen, Ming Lin, Pratap Tokekar
Abstract: Knowledge transfer between teacher and student models has proven effective across various machine learning applications. However, challenges arise when the teacher's predictions are noisy, or the data domain during student training shifts from the teacher's pretraining data. In such scenarios, blindly relying on the teacher's predictions can lead to suboptimal knowledge transfer. To address these challenges, we propose a novel and universal framework, Adaptive Uncertainty-guided Knowledge Transfer ($\textbf{AUKT}$), which leverages Conformal Prediction (CP) to dynamically adjust the student's reliance on the teacher's guidance based on the teacher's prediction uncertainty. CP is a distribution-free, model-agnostic approach that provides reliable prediction sets with statistical coverage guarantees and minimal computational overhead. This adaptive mechanism mitigates the risk of learning undesirable or incorrect knowledge. We validate the proposed framework across diverse applications, including image classification, imitation-guided reinforcement learning, and autonomous driving. Experimental results consistently demonstrate that our approach improves performance, robustness and transferability, offering a promising direction for enhanced knowledge transfer in real-world applications.
Authors: Tiejin Chen, Xiaoou Liu, Longchao Da, Jia Chen, Vagelis Papalexakis, Hua Wei
Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks due to large training datasets and powerful transformer architecture. However, the reliability of responses from LLMs remains a question. Uncertainty quantification (UQ) of LLMs is crucial for ensuring their reliability, especially in areas such as healthcare, finance, and decision-making. Existing UQ methods primarily focus on semantic similarity, overlooking the deeper knowledge dimensions embedded in responses. We introduce a multi-dimensional UQ framework that integrates semantic and knowledge-aware similarity analysis. By generating multiple responses and leveraging auxiliary LLMs to extract implicit knowledge, we construct separate similarity matrices and apply tensor decomposition to derive a comprehensive uncertainty representation. This approach disentangles overlapping information from both semantic and knowledge dimensions, capturing both semantic variations and factual consistency, leading to more accurate UQ. Our empirical evaluations demonstrate that our method outperforms existing techniques in identifying uncertain responses, offering a more robust framework for enhancing LLM reliability in high-stakes applications.
Authors: Yinchuan Li, Xinyu Shao, Jianping Zhang, Haozhi Wang, Leo Maxime Brunswic, Kaiwen Zhou, Jiqian Dong, Kaiyang Guo, Xiu Li, Zhitang Chen, Jun Wang, Jianye Hao
Abstract: In recent years, the exceptional performance of generative models in generative tasks has sparked significant interest in their integration into decision-making processes. Due to their ability to handle complex data distributions and their strong model capacity, generative models can be effectively incorporated into decision-making systems by generating trajectories that guide agents toward high-reward state-action regions or intermediate sub-goals. This paper presents a comprehensive review of the application of generative models in decision-making tasks. We classify seven fundamental types of generative models: energy-based models, generative adversarial networks, variational autoencoders, normalizing flows, diffusion models, generative flow networks, and autoregressive models. Regarding their applications, we categorize their functions into three main roles: controllers, modelers and optimizers, and discuss how each role contributes to decision-making. Furthermore, we examine the deployment of these models across five critical real-world decision-making scenarios. Finally, we summarize the strengths and limitations of current approaches and propose three key directions for advancing next-generation generative directive models: high-performance algorithms, large-scale generalized decision-making models, and self-evolving and adaptive models.
Authors: Mar\'ia Andrea Cruz Bland\'on, Jayasimha Talur, Bruno Charron, Dong Liu, Saab Mansour, Marcello Federico
Abstract: Automatic evaluation of retrieval augmented generation (RAG) systems relies on fine-grained dimensions like faithfulness and relevance, as judged by expert human annotators. Meta-evaluation benchmarks support the development of automatic evaluators that correlate well with human judgement. However, existing benchmarks predominantly focus on English or use translated data, which fails to capture cultural nuances. A native approach provides a better representation of the end user experience. In this work, we develop a Multilingual End-to-end Meta-Evaluation RAG benchmark (MEMERAG). Our benchmark builds on the popular MIRACL dataset, using native-language questions and generating responses with diverse large language models (LLMs), which are then assessed by expert annotators for faithfulness and relevance. We describe our annotation process and show that it achieves high inter-annotator agreement. We then analyse the performance of the answer-generating LLMs across languages as per the human evaluators. Finally we apply the dataset to our main use-case which is to benchmark multilingual automatic evaluators (LLM-as-a-judge). We show that our benchmark can reliably identify improvements offered by advanced prompting techniques and LLMs. We will release our benchmark to support the community developing accurate evaluation methods for multilingual RAG systems.
Authors: Andr\'e V. Duarte, Xuandong Zhao, Arlindo L. Oliveira, Lei Li
Abstract: How can we verify whether copyrighted content was used to train a large vision-language model (VLM) without direct access to its training data? Motivated by the hypothesis that a VLM is able to recognize images from its training corpus, we propose DIS-CO, a novel approach to infer the inclusion of copyrighted content during the model's development. By repeatedly querying a VLM with specific frames from targeted copyrighted material, DIS-CO extracts the content's identity through free-form text completions. To assess its effectiveness, we introduce MovieTection, a benchmark comprising 14,000 frames paired with detailed captions, drawn from films released both before and after a model's training cutoff. Our results show that DIS-CO significantly improves detection performance, nearly doubling the average AUC of the best prior method on models with logits available. Our findings also highlight a broader concern: all tested models appear to have been exposed to some extent to copyrighted content. Our code and data are available at https://github.com/avduarte333/DIS-CO