Authors: Theofanis Aravanis, Grigorios Chrimatopoulos, Mohammad Ferdows, Michalis Xenos, Efstratios Em Tzirtzilakis
Abstract: Unlike conventional Machine-Learning (ML) approaches, often criticized as "black boxes", Symbolic Regression (SR) stands out as a powerful tool for revealing interpretable mathematical relationships in complex physical systems, requiring no a priori assumptions about models' structures. Motivated by the recognition that, in fluid mechanics, an understanding of the underlying flow physics is as crucial as accurate prediction, this study applies SR to model a fundamental three-dimensional (3D) incompressible flow in a rectangular channel, focusing on the (axial) velocity and pressure fields under laminar conditions. By employing the PySR library, compact symbolic equations were derived directly from numerical simulation data, revealing key characteristics of the flow dynamics. These equations not only approximate the parabolic velocity profile and pressure drop observed in the studied fluid flow, but also perfectly coincide with analytical solutions from the literature. Furthermore, we propose an innovative approach that integrates SR with the knowledge-representation framework of Answer Set Programming (ASP), combining the generative power of SR with the declarative reasoning strengths of ASP. The proposed hybrid SR/ASP framework ensures that the SR-generated symbolic expressions are not only statistically accurate, but also physically plausible, adhering to domain-specific principles. Overall, the study highlights two key contributions: SR's ability to simplify complex flow behaviours into concise, interpretable equations, and the potential of knowledge-representation approaches to improve the reliability and alignment of data-driven SR models with domain principles. Insights from the examined 3D channel flow pave the way for integrating such hybrid approaches into efficient frameworks, [...] where explainable predictions and real-time data analysis are crucial.
Authors: SaiBarath Sundar, Pranav Satheesan, Udayaadithya Avadhanam
Abstract: Recent advances in agentic systems for data analysis have emphasized automation of insight generation through multi-agent frameworks, and orchestration layers. While these systems effectively manage tasks like query translation, data transformation, and visualization, they often overlook the structured reasoning process underlying analytical thinking. Reasoning large language models (LLMs) used for multi-step problem solving are trained as general-purpose problem solvers. As a result, their reasoning or thinking steps do not adhere to fixed processes for specific tasks. Real-world data analysis requires a consistent cognitive workflow: interpreting vague goals, grounding them in contextual knowledge, constructing abstract plans, and adapting execution based on intermediate outcomes. We introduce I2I-STRADA (Information-to-Insight via Structured Reasoning Agent for Data Analysis), an agentic architecture designed to formalize this reasoning process. I2I-STRADA focuses on modeling how analysis unfolds via modular sub-tasks that reflect the cognitive steps of analytical reasoning. Evaluations on the DABstep and DABench benchmarks show that I2I-STRADA outperforms prior systems in planning coherence and insight alignment, highlighting the importance of structured cognitive workflows in agent design for data analysis.
Authors: Timothy Tin Long Yu, Mahdi Mostajabdaveh, Jabo Serge Byusa, Rindra Ramamonjison, Giuseppe Carenini, Kun Mao, Zirui Zhou, Yong Zhang
Abstract: Large language models (LLMs) present intriguing opportunities to enhance user interaction with traditional algorithms and tools in real-world applications. An advanced planning system (APS) is a sophisticated software that leverages optimization to help operations planners create, interpret, and modify an operational plan. While highly beneficial, many customers are priced out of using an APS due to the ongoing costs of consultants responsible for customization and maintenance. To address the need for a more accessible APS expressed by supply chain planners, we present SmartAPS, a conversational system built on a tool-augmented LLM. Our system provides operations planners with an intuitive natural language chat interface, allowing them to query information, perform counterfactual reasoning, receive recommendations, and execute scenario analysis to better manage their operation. A short video demonstrating the system has been released: https://youtu.be/KtIrJjlDbyw
Authors: Dario Della Monica, Angelo Montanari, Pietro Sala
Abstract: Qualitative timeline-based planning models domains as sets of independent, but interacting, components whose behaviors over time, the timelines, are governed by sets of qualitative temporal constraints (ordering relations), called synchronization rules. Its plan-existence problem has been shown to be PSPACE-complete; in particular, PSPACE-membership has been proved via reduction to the nonemptiness problem for nondeterministic finite automata. However, nondeterministic automata cannot be directly used to synthesize planning strategies as a costly determinization step is needed. In this paper, we identify a fragment of qualitative timeline-based planning whose plan-existence problem can be directly mapped into the nonemptiness problem of deterministic finite automata, which can then synthesize strategies. In addition, we identify a maximal subset of Allen's relations that fits into such a deterministic fragment.
Authors: Yusen Peng, Shuhua Mao
Abstract: How can AI move beyond imitation toward genuine creativity? This paper proposes the E.A.R.T.H. framework, a five-stage generative pipeline that transforms model-generated errors into creative assets through Error generation, Amplification, Refine selection, Transform, and Harness feedback. Drawing on cognitive science and generative modeling, we posit that "creative potential hides in failure" and operationalize this via structured prompts, semantic scoring, and human-in-the-loop evaluation. Implemented using LLaMA-2-7B-Chat, SBERT, BERTScore, CLIP, BLIP-2, and Stable Diffusion, the pipeline employs a composite reward function based on novelty, surprise, and relevance. At the Refine stage, creativity scores increase by 52.5% (1.179 to 1.898, t = -5.56, p < 0.001), with final outputs reaching 2.010 - a 70.4% improvement. Refined slogans are 48.4% shorter, 40.7% more novel, with only a 4.0% drop in relevance. Cross-modal tests show strong slogan-to-image alignment (CLIPScore: 0.249; BERTScore F1: 0.816). In human evaluations, 60% of outputs scored >= 4.0, with metaphorical slogans (avg. 4.09) outperforming literal ones (3.99). Feedback highlights stylistic precision and emotional resonance. These results demonstrate that error-centered, feedback-driven generation enhances creativity, offering a scalable path toward self-evolving, human-aligned creative AI.
Authors: Victoria R. Li, Johnathan Sun, Martin Wattenberg
Abstract: Charts and graphs help people analyze data, but can they also be useful to AI systems? To investigate this question, we perform a series of experiments with two commercial vision-language models: GPT 4.1 and Claude 3.5. Across three representative analysis tasks, the two systems describe synthetic datasets more precisely and accurately when raw data is accompanied by a scatterplot, especially as datasets grow in complexity. Comparison with two baselines -- providing a blank chart and a chart with mismatched data -- shows that the improved performance is due to the content of the charts. Our results are initial evidence that AI systems, like humans, can benefit from visualization.
Authors: Yueheng Li, Guangming Xie, Zongqing Lu
Abstract: Due to practical constraints such as partial observability and limited communication, Centralized Training with Decentralized Execution (CTDE) has become the dominant paradigm in cooperative Multi-Agent Reinforcement Learning (MARL). However, existing CTDE methods often underutilize centralized training or lack theoretical guarantees. We propose Multi-Agent Guided Policy Optimization (MAGPO), a novel framework that better leverages centralized training by integrating centralized guidance with decentralized execution. MAGPO uses an auto-regressive joint policy for scalable, coordinated exploration and explicitly aligns it with decentralized policies to ensure deployability under partial observability. We provide theoretical guarantees of monotonic policy improvement and empirically evaluate MAGPO on 43 tasks across 6 diverse environments. Results show that MAGPO consistently outperforms strong CTDE baselines and matches or surpasses fully centralized approaches, offering a principled and practical solution for decentralized multi-agent learning. Our code and experimental data can be found in https://github.com/liyheng/MAGPO.
Authors: Yixiu Liu, Yang Nan, Weixian Xu, Xiangkun Hu, Lyumanshan Ye, Zhen Qin, Pengfei Liu
Abstract: While AI systems demonstrate exponentially improving capabilities, the pace of AI research itself remains linearly bounded by human cognitive capacity, creating an increasingly severe development bottleneck. We present ASI-Arch, the first demonstration of Artificial Superintelligence for AI research (ASI4AI) in the critical domain of neural architecture discovery--a fully autonomous system that shatters this fundamental constraint by enabling AI to conduct its own architectural innovation. Moving beyond traditional Neural Architecture Search (NAS), which is fundamentally limited to exploring human-defined spaces, we introduce a paradigm shift from automated optimization to automated innovation. ASI-Arch can conduct end-to-end scientific research in the domain of architecture discovery, autonomously hypothesizing novel architectural concepts, implementing them as executable code, training and empirically validating their performance through rigorous experimentation and past experience. ASI-Arch conducted 1,773 autonomous experiments over 20,000 GPU hours, culminating in the discovery of 106 innovative, state-of-the-art (SOTA) linear attention architectures. Like AlphaGo's Move 37 that revealed unexpected strategic insights invisible to human players, our AI-discovered architectures demonstrate emergent design principles that systematically surpass human-designed baselines and illuminate previously unknown pathways for architectural innovation. Crucially, we establish the first empirical scaling law for scientific discovery itself--demonstrating that architectural breakthroughs can be scaled computationally, transforming research progress from a human-limited to a computation-scalable process. We provide comprehensive analysis of the emergent design patterns and autonomous research capabilities that enabled these breakthroughs, establishing a blueprint for self-accelerating AI systems.
Authors: Soorya Ram Shimgekar, Shayan Vassef, Abhay Goyal, Navin Kumar, Koustuv Saha
Abstract: Building and deploying machine learning solutions in healthcare remains expensive and labor-intensive due to fragmented preprocessing workflows, model compatibility issues, and stringent data privacy constraints. In this work, we introduce an Agentic AI framework that automates the entire clinical data pipeline, from ingestion to inference, through a system of modular, task-specific agents. These agents handle both structured and unstructured data, enabling automatic feature selection, model selection, and preprocessing recommendation without manual intervention. We evaluate the system on publicly available datasets from geriatrics, palliative care, and colonoscopy imaging. For example, in the case of structured data (anxiety data) and unstructured data (colonoscopy polyps data), the pipeline begins with file-type detection by the Ingestion Identifier Agent, followed by the Data Anonymizer Agent ensuring privacy compliance, where we first identify the data type and then anonymize it. The Feature Extraction Agent identifies features using an embedding-based approach for tabular data, extracting all column names, and a multi-stage MedGemma-based approach for image data, which infers modality and disease name. These features guide the Model-Data Feature Matcher Agent in selecting the best-fit model from a curated repository. The Preprocessing Recommender Agent and Preprocessing Implementor Agent then apply tailored preprocessing based on data type and model requirements. Finally, the ``Model Inference Agent" runs the selected model on the uploaded data and generates interpretable outputs using tools like SHAP, LIME, and DETR attention maps. By automating these high-friction stages of the ML lifecycle, the proposed framework reduces the need for repeated expert intervention, offering a scalable, cost-efficient pathway for operationalizing AI in clinical environments.
Authors: Sedigh Khademi, Christopher Palmer, Muhammad Javed, Hazel Clothier, Jim Buttery, Gerardo Luis Dimaguila, Jim Black
Abstract: The rapid development of COVID-19 vaccines has showcased the global communitys ability to combat infectious diseases. However, the need for post-licensure surveillance systems has grown due to the limited window for safety data collection in clinical trials and early widespread implementation. This study aims to employ Natural Language Processing techniques and Active Learning to rapidly develop a classifier that detects potential vaccine safety issues from emergency department notes. ED triage notes, containing expert, succinct vital patient information at the point of entry to health systems, can significantly contribute to timely vaccine safety signal surveillance. While keyword-based classification can be effective, it may yield false positives and demand extensive keyword modifications. This is exacerbated by the infrequency of vaccination-related ED presentations and their similarity to other reasons for ED visits. NLP offers a more accurate and efficient alternative, albeit requiring annotated data, which is often scarce in the medical field. Active learning optimizes the annotation process and the quality of annotated data, which can result in faster model implementation and improved model performance. This work combines active learning, data augmentation, and active learning and evaluation techniques to create a classifier that is used to enhance vaccine safety surveillance from ED triage notes.
Authors: Moritz Sch\"onherr, Carsten Lutz
Abstract: We study the expressive power of graph neural networks (GNNs) with mean as the aggregation function. In the non-uniform setting, we show that such GNNs have exactly the same expressive power as ratio modal logic, which has modal operators expressing that at least a certain ratio of the successors of a vertex satisfies a specified property. The non-uniform expressive power of mean GNNs is thus higher than that of GNNs with max aggregation, but lower than for sum aggregation--the latter are characterized by modal logic and graded modal logic, respectively. In the uniform setting, we show that the expressive power relative to MSO is exactly that of alternation-free modal logic, under the natural assumptions that combination functions are continuous and classification functions are thresholds. This implies that, relative to MSO and in the uniform setting, mean GNNs are strictly less expressive than sum GNNs and max GNNs. When any of the assumptions is dropped, the expressive power increases.
Authors: Mutian Yang, Jiandong Gao, Ji Wu
Abstract: While large language models (LLMs) leverage both knowledge and reasoning during inference, the capacity to distinguish between them plays a pivotal role in model analysis, interpretability, and development. Inspired by dual-system cognitive theory, we propose a cognition attribution framework to decouple the contribution of knowledge and reasoning. In particular, the cognition of LLMs is decomposed into two distinct yet complementary phases: knowledge retrieval (Phase 1) and reasoning adjustment (Phase 2). To separate these phases, LLMs are prompted to generate answers under two different cognitive modes, fast thinking and slow thinking, respectively. The performance under different cognitive modes is analyzed to quantify the contribution of knowledge and reasoning. This architecture is employed to 15 LLMs across 3 datasets. Results reveal: (1) reasoning adjustment is domain-specific, benefiting reasoning-intensive domains (e.g., mathematics, physics, and chemistry) and potentially imparing knowledge-intensive domains. (2) Parameter scaling improves both knowledge and reasoning, with knowledge improvements being more pronounced. Additionally, parameter scaling make LLMs reasoning significantly more prudent, while moderately more intelligent. (3) Knowledge primarily resides in lower network layers, while reasoning operates in higher layers. Our framework not only helps understand LLMs from a "decoupling" perspective, but also provides new insights into existing research, including scaling laws, hierarchical knowledge editing, and limitations of small-model reasoning.
Authors: Felicidad Aguado, Pedro Cabalar, Brais Mu\~niz, Gilberto P\'erez, Concepci\'on Vidal
Abstract: In this paper, we compare four different semantics for disjunction in Answer Set Programming that, unlike stable models, do not adhere to the principle of model minimality. Two of these approaches, Cabalar and Mu\~niz' \emph{Justified Models} and Doherty and Szalas' \emph{Strongly Supported Models}, directly provide an alternative non-minimal semantics for disjunction. The other two, Aguado et al's \emph{Forks} and Shen and Eiter's \emph{Determining Inference} (DI) semantics, actually introduce a new disjunction connective, but are compared here as if they constituted new semantics for the standard disjunction operator. We are able to prove that three of these approaches (Forks, Justified Models and a reasonable relaxation of the DI semantics) actually coincide, constituting a common single approach under different definitions. Moreover, this common semantics always provides a superset of the stable models of a program (in fact, modulo any context) and is strictly stronger than the fourth approach (Strongly Supported Models), that actually treats disjunctions as in classical logic.
Authors: Antonino Rotolo, Beatrice Ferrigno, Jose Miguel Angel Garcia Godinez, Claudio Novelli, Giovanni Sartor
Abstract: This chapter introduces a conceptual framework for qualitative risk assessment of AI, particularly in the context of the EU AI Act. The framework addresses the complexities of legal compliance and fundamental rights protection by itegrating definitional balancing and defeasible reasoning. Definitional balancing employs proportionality analysis to resolve conflicts between competing rights, while defeasible reasoning accommodates the dynamic nature of legal decision-making. Our approach stresses the need for an analysis of AI deployment scenarios and for identifying potential legal violations and multi-layered impacts on fundamental rights. On the basis of this analysis, we provide philosophical foundations for a logical account of AI risk analysis. In particular, we consider the basic building blocks for conceptually grasping the interaction between AI deployment scenarios and fundamental rights, incorporating in defeasible reasoning definitional balancing and arguments about the contextual promotion or demotion of rights. This layered approach allows for more operative models of assessment of both high-risk AI systems and General Purpose AI (GPAI) systems, emphasizing the broader applicability of the latter. Future work aims to develop a formal model and effective algorithms to enhance AI risk assessment, bridging theoretical insights with practical applications to support responsible AI governance.
Authors: Peter Baumgartner, Lachlan McGinness
Abstract: We present our method for automatically marking Physics exams. The marking problem consists in assessing typed student answers for correctness with respect to a ground truth solution. This is a challenging problem that we seek to tackle using a combination of a computer algebra system, an SMT solver and a term rewriting system. A Large Language Model is used to interpret and remove errors from student responses and rewrite these in a machine readable format. Once formalized and language-aligned, the next step then consists in applying automated reasoning techniques for assessing student solution correctness. We consider two methods of automated theorem proving: off-the-shelf SMT solving and term rewriting systems tailored for physics problems involving trigonometric expressions. The development of the term rewrite system and establishing termination and confluence properties was not trivial, and we describe it in some detail in the paper. We evaluate our system on a rich pool of over 1500 real-world student exam responses from the 2023 Australian Physics Olympiad.
Authors: Zhuang Qiang Bok, Watson Wei Khong Chua
Abstract: Most reasoning benchmarks for LLMs emphasize factual accuracy or step-by-step logic. In finance, however, professionals must not only converge on optimal decisions but also generate creative, plausible futures under uncertainty. We introduce ConDiFi, a benchmark that jointly evaluates divergent and convergent thinking in LLMs for financial tasks. ConDiFi features 607 macro-financial prompts for divergent reasoning and 990 multi-hop adversarial MCQs for convergent reasoning. Using this benchmark, we evaluated 14 leading models and uncovered striking differences. Despite high fluency, GPT-4o underperforms on Novelty and Actionability. In contrast, models like DeepSeek-R1 and Cohere Command R+ rank among the top for generating actionable, insights suitable for investment decisions. ConDiFi provides a new perspective to assess reasoning capabilities essential to safe and strategic deployment of LLMs in finance.
Authors: Shiye Lei, Zhihao Cheng, Kai Jia, Dacheng Tao
Abstract: Large language models (LLMs) have recently demonstrated remarkable progress in reasoning capabilities through reinforcement learning with verifiable rewards (RLVR). By leveraging simple rule-based rewards, RL effectively incentivizes LLMs to produce extended chain-of-thought (CoT) reasoning trajectories, progressively guiding them toward correct answers. However, existing approaches remain largely heuristic and intuition-driven, limiting the development of principled methodologies. In this paper, we present a theoretical characterization of LLM reasoning grounded in information bottleneck (IB) principle, introducing IB-aware reasoning optimization (IBRO), a framework that encourages reasoning trajectories to be both informative about the final correct answer and generalizable across diverse prompts. We derive a practical token-level surrogate objective and propose an efficient approximation, resulting in the lightweight IB regularization method. This technique integrates seamlessly into existing RL-based post-training frameworks without additional computational overhead, requiring only a one-line code modification. Empirically, we validate IB regularization across multiple mathematical reasoning benchmarks and RL algorithms, demonstrating consistent improvements in LLM reasoning performance.
Authors: Kwong Ho Li, Wathsala Karunarathne
Abstract: This paper investigates the application of Reinforcement Learning (RL) to optimise call routing in call centres to minimise client waiting time and staff idle time. Two methods are compared: a model-based approach using Value Iteration (VI) under known system dynamics, and a model-free approach using Proximal Policy Optimisation (PPO) that learns from experience. For the model-based approach, a theoretical model is used, while a simulation model combining Discrete Event Simulation (DES) with the OpenAI Gym environment is developed for model-free learning. Both models frame the problem as a Markov Decision Process (MDP) within a Skills-Based Routing (SBR) framework, with Poisson client arrivals and exponentially distributed service and abandonment times. For policy evaluation, random, VI, and PPO policies are evaluated using the simulation model. After 1,000 test episodes, PPO consistently achives the highest rewards, along with the lowest client waiting time and staff idle time, despite requiring longer training time.
Authors: Enrico Santi, Fabio Tardivo, Agostino Dovier, Andrea Formisano
Abstract: Constraint Programming developed within Logic Programming in the Eighties; nowadays all Prolog systems encompass modules capable of handling constraint programming on finite domains demanding their solution to a constraint solver. This work focuses on a specific form of constraint, the so-called table constraint, used to specify conditions on the values of variables as an enumeration of alternative options. Since every condition on a set of finite domain variables can be ultimately expressed as a finite set of cases, Table can, in principle, simulate any other constraint. These characteristics make Table one of the most studied constraints ever, leading to a series of increasingly efficient propagation algorithms. Despite this, it is not uncommon to encounter real-world problems with hundreds or thousands of valid cases that are simply too many to be handled effectively with standard CPU-based approaches. In this paper, we deal with the Compact-Table (CT) algorithm, the state-of-the-art propagation algorithms for Table. We describe how CT can be enhanced by exploiting the massive computational power offered by modern GPUs to handle large Table constraints. In particular, we report on the design and implementation of GPU-accelerated CT, on its integration into an existing constraint solver, and on an experimental validation performed on a significant set of instances.
Authors: Manuel de Sousa Ribeiro, Afonso Leote, Jo\~ao Leite
Abstract: Concept probing has recently garnered increasing interest as a way to help interpret artificial neural networks, dealing both with their typically large size and their subsymbolic nature, which ultimately renders them unfeasible for direct human interpretation. Concept probing works by training additional classifiers to map the internal representations of a model into human-defined concepts of interest, thus allowing humans to peek inside artificial neural networks. Research on concept probing has mainly focused on the model being probed or the probing model itself, paying limited attention to the data required to train such probing models. In this paper, we address this gap. Focusing on concept probing in the context of image classification tasks, we investigate the effect of the data used to train probing models on their performance. We also make available concept labels for two widely used datasets.
Authors: Shanghai AI Lab, :, Yicheng Bao, Guanxu Chen, Mingkang Chen, Yunhao Chen, Chiyu Chen, Lingjie Chen, Sirui Chen, Xinquan Chen, Jie Cheng, Yu Cheng, Dengke Deng, Yizhuo Ding, Dan Ding, Xiaoshan Ding, Yi Ding, Zhichen Dong, Lingxiao Du, Yuyu Fan, Xinshun Feng, Yanwei Fu, Yuxuan Gao, Ruijun Ge, Tianle Gu, Lujun Gui, Jiaxuan Guo, Qianxi He, Yuenan Hou, Xuhao Hu, Hong Huang, Kaichen Huang, Shiyang Huang, Yuxian Jiang, Shanzhe Lei, Jie Li, Lijun Li, Hao Li, Juncheng Li, Xiangtian Li, Yafu Li, Lingyu Li, Xueyan Li, Haotian Liang, Dongrui Liu, Qihua Liu, Zhixuan Liu, Bangwei Liu, Huacan Liu, Yuexiao Liu, Zongkai Liu, Chaochao Lu, Yudong Lu, Xiaoya Lu, Zhenghao Lu, Qitan Lv, Caoyuan Ma, Jiachen Ma, Xiaoya Ma, Zhongtian Ma, Lingyu Meng, Ziqi Miao, Yazhe Niu, Yuezhang Peng, Yuan Pu, Han Qi, Chen Qian, Xingge Qiao, Jingjing Qu, Jiashu Qu, Wanying Qu, Wenwen Qu, Xiaoye Qu, Qihan Ren, Qingnan Ren, Qingyu Ren, Jing Shao, Wenqi Shao, Shuai Shao, Dongxing Shi, Xin Song, Xinhao Song, Yan Teng, Xuan Tong, Yingchun Wang, Xuhong Wang, Shujie Wang, Xin Wang, Yige Wang, Yixu Wang, Yuanfu Wang, Futing Wang, Ruofan Wang, Wenjie Wang, Yajie Wang, Muhao Wei, Xiaoyu Wen, Fenghua Weng, Yuqi Wu, Yingtong Xiong, Xingcheng Xu, Chao Yang, Yue Yang, Yang Yao, Yulei Ye, Zhenyun Yin, Yi Yu, Bo Zhang, Qiaosheng Zhang, Jinxuan Zhang, Yexin Zhang, Yinqiang Zheng, Hefeng Zhou, Zhanhui Zhou, Pengyu Zhu, Qingzi Zhu, Yubo Zhu, Bowen Zhou
Abstract: We introduce SafeWork-R1, a cutting-edge multimodal reasoning model that demonstrates the coevolution of capabilities and safety. It is developed by our proposed SafeLadder framework, which incorporates large-scale, progressive, safety-oriented reinforcement learning post-training, supported by a suite of multi-principled verifiers. Unlike previous alignment methods such as RLHF that simply learn human preferences, SafeLadder enables SafeWork-R1 to develop intrinsic safety reasoning and self-reflection abilities, giving rise to safety `aha' moments. Notably, SafeWork-R1 achieves an average improvement of $46.54\%$ over its base model Qwen2.5-VL-72B on safety-related benchmarks without compromising general capabilities, and delivers state-of-the-art safety performance compared to leading proprietary models such as GPT-4.1 and Claude Opus 4. To further bolster its reliability, we implement two distinct inference-time intervention methods and a deliberative search mechanism, enforcing step-level verification. Finally, we further develop SafeWork-R1-InternVL3-78B, SafeWork-R1-DeepSeek-70B, and SafeWork-R1-Qwen2.5VL-7B. All resulting models demonstrate that safety and capability can co-evolve synergistically, highlighting the generalizability of our framework in building robust, reliable, and trustworthy general-purpose AI.
Authors: Feng-Qi Cui, Anyang Tong, Jinyang Huang, Jie Zhang, Dan Guo, Zhi Liu, Meng Wang
Abstract: Dynamic Facial Expression Recognition (DFER) plays a critical role in affective computing and human-computer interaction. Although existing methods achieve comparable performance, they inevitably suffer from performance degradation under sample heterogeneity caused by multi-source data and individual expression variability. To address these challenges, we propose a novel framework, called Heterogeneity-aware Distributional Framework (HDF), and design two plug-and-play modules to enhance time-frequency modeling and mitigate optimization imbalance caused by hard samples. Specifically, the Time-Frequency Distributional Attention Module (DAM) captures both temporal consistency and frequency robustness through a dual-branch attention design, improving tolerance to sequence inconsistency and visual style shifts. Then, based on gradient sensitivity and information bottleneck principles, an adaptive optimization module Distribution-aware Scaling Module (DSM) is introduced to dynamically balance classification and contrastive losses, enabling more stable and discriminative representation learning. Extensive experiments on two widely used datasets, DFEW and FERV39k, demonstrate that HDF significantly improves both recognition accuracy and robustness. Our method achieves superior weighted average recall (WAR) and unweighted average recall (UAR) while maintaining strong generalization across diverse and imbalanced scenarios. Codes are released at https://github.com/QIcita/HDF_DFER.
Authors: Liang Zhang, Xiaoming Zhai, Jionghao Lin, Jionghao Lin, Jennifer Kleiman, Diego Zapata-Rivera, Carol Forsyth, Yang Jiang, Xiangen Hu, Arthur C. Graesser
Abstract: Large Language Model (LLM) agents are increasingly utilized in AI-aided education to support tutoring and learning. Effective communication strategies among LLM agents improve collaborative problem-solving efficiency and facilitate cost-effective adoption in education. However, little research has systematically evaluated the impact of different communication strategies on agents' problem-solving. Our study examines four communication modes, \textit{teacher-student interaction}, \textit{peer-to-peer collaboration}, \textit{reciprocal peer teaching}, and \textit{critical debate}, in a dual-agent, chat-based mathematical problem-solving environment using the OpenAI GPT-4o model. Evaluated on the MATH dataset, our results show that dual-agent setups outperform single agents, with \textit{peer-to-peer collaboration} achieving the highest accuracy. Dialogue acts like statements, acknowledgment, and hints play a key role in collaborative problem-solving. While multi-agent frameworks enhance computational tasks, effective communication strategies are essential for tackling complex problems in AI education.
Authors: Justin Morse, Kurt Gilbert, Kyle Shin, Rick Cooke, Peyton Rose, Jack Sullivan, Angelo Sisante
Abstract: Clinician burnout has motivated the growing adoption of ambient medical scribes in the clinic. In this work, we introduce a custom-built ambient scribe application integrated into the EHR system at Included Health, a personalized all-in-one healthcare company offering telehealth services. The application uses Whisper for transcription and a modular in-context learning pipeline with GPT-4o to automatically generate SOAP notes and patient instructions. Testing on mock visit data shows that the notes generated by the application exceed the quality of expert-written notes as determined by an LLM-as-a-judge. The application has been widely adopted by the clinical practice, with over 540 clinicians at Included Health using the application at least once. 94% (n = 63) of surveyed clinicians report reduced cognitive load during visits and 97% (n = 66) report less documentation burden when using the application. Additionally, we show that post-processing notes with a fine-tuned BART model improves conciseness. These findings highlight the potential for AI systems to ease administrative burdens and support clinicians in delivering efficient, high-quality care.
Authors: Josh Hunter, John McDermid, Simon Burton
Abstract: This study investigates how railway professionals perceive safety as a concept within rail, with the intention to help inform future technological developments within the industry. Through a series of interviews with drivers, route planners,and administrative personnel, the research explores the currentstate of safety practices, the potential for automation and the understanding of the railway as a system of systems. Key findings highlight a cautious attitude towards automation, a preference for assistive technologies, and a complex understanding of safety that integrates human, systematic and technological factors. The study also addresses the limitations of transferring automotive automation technologies to railways and the need for a railway-specific causation model to better evaluate and enhance safety in an evolving technological landscape. This study aims to bridge thegap between contemporary research and practical applications, contributing to the development of more effective safety metrics.
Authors: Fatma Bet\"ul G\"ure\c{s}, Tanya Nazaretsky, Bahar Radmehr, Martina Rau, Tanja K\"aser
Abstract: Supporting students in developing effective diagnostic reasoning is a key challenge in various educational domains. Novices often struggle with cognitive biases such as premature closure and over-reliance on heuristics. Scenario-based learning (SBL) can address these challenges by offering realistic case experiences and iterative practice, but the optimal sequencing of instruction and problem-solving activities remains unclear. This study examines how personalized support can be incorporated into different instructional sequences and whether providing explicit diagnostic strategy instruction before (I-PS) or after problem-solving (PS-I) improves learning and its transfer. We employ a between-groups design in an online SBL environment called PharmaSim, which simulates real-world client interactions for pharmacy technician apprentices. Results indicate that while both instruction types are beneficial, PS-I leads to significantly higher performance in transfer tasks.
Authors: Arindam Ghosh, Mark Fuhs, Bongjun Kim, Anurag Chowdhury, Monika Woszczyna
Abstract: From an application standpoint, speaker-role diarization (RD), such as doctor vs. patient, host vs. guest, etc. is often more useful than traditional speaker diarization (SD), which assigns generic labels like speaker-1, speaker-2 etc. In the context of joint automatic speech recognition (ASR) + SD (who spoke what?), recent end-to-end models employ an auxiliary SD transducer, synchronized with the ASR transducer, to predict speakers per word. In this paper, we extend this framework to RD with three key contributions: (1) we simplify the training via forced alignment and cross-entropy loss instead of RNNT loss, (2) we show that word prediction and role prediction require different amounts of predictor's context, leading to separate task-specific predictors, unlike existing shared-predictor models, and (3) we propose a way to leverage RD posterior activity to influence ASR decoding and reduce small-word deletion errors.
Authors: Ahmad Alhonainy (University of Missouri, USA), Praveen Rao (University of Missouri, USA)
Abstract: Federated Learning (FL) allows multiple distributed devices to jointly train a shared model without centralizing data, but communication cost remains a major bottleneck, especially in resource-constrained environments. This paper introduces caching strategies - FIFO, LRU, and Priority-Based - to reduce unnecessary model update transmissions. By selectively forwarding significant updates, our approach lowers bandwidth usage while maintaining model accuracy. Experiments on CIFAR-10 and medical datasets show reduced communication with minimal accuracy loss. Results confirm that intelligent caching improves scalability, memory efficiency, and supports reliable FL in edge IoT networks, making it practical for deployment in smart cities, healthcare, and other latency-sensitive applications.
Authors: Zhangqi Liu
Abstract: As artificial intelligence (AI) continues to evolve from a back-end computational tool into an interactive, generative collaborator, its integration into early-stage design processes demands a rethinking of traditional workflows in human-centered design. This paper explores the emergent paradigm of human-AI co-creation, where AI is not merely used for automation or efficiency gains, but actively participates in ideation, visual conceptualization, and decision-making. Specifically, we investigate the use of large language models (LLMs) like GPT-4 and multimodal diffusion models such as Stable Diffusion as creative agents that engage designers in iterative cycles of proposal, critique, and revision.
Authors: Alexander D. Kalian, Lennart Otte, Jaewook Lee, Emilio Benfenati, Jean-Lou C. M. Dorne, Claire Potter, Olivia J. Osborne, Miao Guo, Christer Hogstrand
Abstract: Geometric deep learning is an emerging technique in Artificial Intelligence (AI) driven cheminformatics, however the unique implications of different Graph Neural Network (GNN) architectures are poorly explored, for this space. This study compared performances of Graph Convolutional Networks (GCNs), Graph Attention Networks (GATs) and Graph Isomorphism Networks (GINs), applied to 7 different toxicological assay datasets of varying data abundance and endpoint, to perform binary classification of assay activation. Following pre-processing of molecular graphs, enforcement of class-balance and stratification of all datasets across 5 folds, Bayesian optimisations were carried out, for each GNN applied to each assay dataset (resulting in 21 unique Bayesian optimisations). Optimised GNNs performed at Area Under the Curve (AUC) scores ranging from 0.728-0.849 (averaged across all folds), naturally varying between specific assays and GNNs. GINs were found to consistently outperform GCNs and GATs, for the top 5 of 7 most data-abundant toxicological assays. GATs however significantly outperformed over the remaining 2 most data-scarce assays. This indicates that GINs are a more optimal architecture for data-abundant environments, whereas GATs are a more optimal architecture for data-scarce environments. Subsequent analysis of the explored higher-dimensional hyperparameter spaces, as well as optimised hyperparameter states, found that GCNs and GATs reached measurably closer optimised states with each other, compared to GINs, further indicating the unique nature of GINs as a GNN algorithm.
Authors: Jie Fan
Abstract: In a recent paper, Kit Fine presents some striking results concerning the logical properties of (first-order) ignorance, second-order ignorance and Rumsfeld ignorance. However, Rumsfeld ignorance is definable in terms of ignorance, which makes some existing results and the axiomatization problem trivial. A main reason is that the accessibility relations for the implicit knowledge operator contained in the packaged operators of ignorance and Rumsfeld ignorance are the same. In this work, we assume the two accessibility relations to be different so that one of them is an arbitrary subset of the other. This will avoid the definability issue and retain most of the previous validities. The main results are axiomatizations over various proper bi-frame classes. Finally we apply our framework to analyze Fine's results.
Authors: M. Tedeschi, S. Rizwan, C. Shringi, V. Devram Chandgir, S. Belich
Abstract: Contemporary database systems, while effective, suffer severe issues related to complexity and usability, especially among individuals who lack technical expertise but are unfamiliar with query languages like Structured Query Language (SQL). This paper presents a new database system supported by Artificial Intelligence (AI), which is intended to improve the management of data using natural language processing (NLP) - based intuitive interfaces, and automatic creation of structured queries and semi-structured data formats like yet another markup language (YAML), java script object notation (JSON), and application program interface (API) documentation. The system is intended to strengthen the potential of databases through the integration of Large Language Models (LLMs) and advanced machine learning algorithms. The integration is purposed to allow the automation of fundamental tasks such as data modeling, schema creation, query comprehension, and performance optimization. We present in this paper a system that aims to alleviate the main problems with current database technologies. It is meant to reduce the need for technical skills, manual tuning for better performance, and the potential for human error. The AI database employs generative schema inference and format selection to build its schema models and execution formats.
Authors: Randy Davila, Boris Brimkov, Ryan Pepper
Abstract: We present four open conjectures in graph theory generated by the automated conjecturing system \texttt{TxGraffiti}. Each conjecture is concise, grounded in natural graph invariants, and empirically validated across hundreds of graphs. Despite extensive effort, these statements remain unresolved--defying both proof and counterexample. They are not only mathematical challenges but creative expressions--born of symbolic pattern recognition and mathematician-defined heuristics, refined through years of human dialogue, and now offered back to the community as collaborative artifacts. These conjectures invite not only formal proof, but also reflection on how machines can evoke wonder, spark curiosity, and contribute to the raw material of discovery. By highlighting these problems, we aim to inspire both human mathematicians and AI systems to engage with them--not only to solve them, but to reflect on what it means when machines participate meaningfully in the creative process of mathematical thought.
Authors: Neil He, Hiren Madhu, Ngoc Bui, Menglin Yang, Rex Ying
Abstract: Foundation models pre-trained on massive datasets, including large language models (LLMs), vision-language models (VLMs), and large multimodal models, have demonstrated remarkable success in diverse downstream tasks. However, recent studies have shown fundamental limitations of these models: (1) limited representational capacity, (2) lower adaptability, and (3) diminishing scalability. These shortcomings raise a critical question: is Euclidean geometry truly the optimal inductive bias for all foundation models, or could incorporating alternative geometric spaces enable models to better align with the intrinsic structure of real-world data and improve reasoning processes? Hyperbolic spaces, a class of non-Euclidean manifolds characterized by exponential volume growth with respect to distance, offer a mathematically grounded solution. These spaces enable low-distortion embeddings of hierarchical structures (e.g., trees, taxonomies) and power-law distributions with substantially fewer dimensions compared to Euclidean counterparts. Recent advances have leveraged these properties to enhance foundation models, including improving LLMs' complex reasoning ability, VLMs' zero-shot generalization, and cross-modal semantic alignment, while maintaining parameter efficiency. This paper provides a comprehensive review of hyperbolic neural networks and their recent development for foundation models. We further outline key challenges and research directions to advance the field.
Authors: Ali Vardasbi, Gustavo Penha, Claudia Hauff, Hugues Bouchard
Abstract: When using LLMs to rank items based on given criteria, or evaluate answers, the order of candidate items can influence the model's final decision. This sensitivity to item positioning in a LLM's prompt is known as position bias. Prior research shows that this bias exists even in large models, though its severity varies across models and tasks. In addition to position bias, LLMs also exhibit varying degrees of low repetition consistency, where repeating the LLM call with the same candidate ordering can lead to different rankings. To address both inconsistencies, a common approach is to prompt the model multiple times with different candidate orderings and aggregate the results via majority voting. However, this repetition strategy, significantly increases computational costs. Extending prior findings, we observe that both the direction -- favoring either the earlier or later candidate in the prompt -- and magnitude of position bias across instances vary substantially, even within a single dataset. This observation highlights the need for a per-instance mitigation strategy. To this end, we introduce a dynamic early-stopping method that adaptively determines the number of repetitions required for each instance. Evaluating our approach across three LLMs of varying sizes and on two tasks, namely re-ranking and alignment, we demonstrate that transitioning to a dynamic repetition strategy reduces the number of LLM calls by an average of 81%, while preserving the accuracy. Furthermore, we propose a confidence-based adaptation to our early-stopping method, reducing LLM calls by an average of 87% compared to static repetition, with only a slight accuracy trade-off relative to our original early-stopping method.
Authors: Eduardo Aguilar-Bejarano, Daniel Lea, Karthikeyan Sivakumar, Jimiama M. Mase, Reza Omidvar, Ruizhe Li, Troy Kettle, James Mitchell-White, Morgan R Alexander, David A Winkler, Grazziela Figueredo
Abstract: Helix is an open-source, extensible, Python-based software framework to facilitate reproducible and interpretable machine learning workflows for tabular data. It addresses the growing need for transparent experimental data analytics provenance, ensuring that the entire analytical process -- including decisions around data transformation and methodological choices -- is documented, accessible, reproducible, and comprehensible to relevant stakeholders. The platform comprises modules for standardised data preprocessing, visualisation, machine learning model training, evaluation, interpretation, results inspection, and model prediction for unseen data. To further empower researchers without formal training in data science to derive meaningful and actionable insights, Helix features a user-friendly interface that enables the design of computational experiments, inspection of outcomes, including a novel interpretation approach to machine learning decisions using linguistic terms all within an integrated environment. Released under the MIT licence, Helix is accessible via GitHub and PyPI, supporting community-driven development and promoting adherence to the FAIR principles.
Authors: Sai Varun Kodathala, Yashwanth Reddy Vutukoori, Rakesh Vunnam
Abstract: This paper addresses the challenge of automated sports video analysis, which has traditionally been limited by computationally intensive models requiring server-side processing and lacking fine-grained understanding of athletic movements. Current approaches struggle to capture the nuanced biomechanical transitions essential for meaningful sports analysis, often missing critical phases like preparation, execution, and follow-through that occur within seconds. To address these limitations, we introduce SV3.3B, a lightweight 3.3B parameter video understanding model that combines novel temporal motion difference sampling with self-supervised learning for efficient on-device deployment. Our approach employs a DWT-VGG16-LDA based keyframe extraction mechanism that intelligently identifies the 16 most representative frames from sports sequences, followed by a V-DWT-JEPA2 encoder pretrained through mask-denoising objectives and an LLM decoder fine-tuned for sports action description generation. Evaluated on a subset of the NSVA basketball dataset, SV3.3B achieves superior performance across both traditional text generation metrics and sports-specific evaluation criteria, outperforming larger closed-source models including GPT-4o variants while maintaining significantly lower computational requirements. Our model demonstrates exceptional capability in generating technically detailed and analytically rich sports descriptions, achieving 29.2% improvement over GPT-4o in ground truth validation metrics, with substantial improvements in information density, action complexity, and measurement precision metrics essential for comprehensive athletic analysis. Model Available at https://huggingface.co/sportsvision/SV3.3B.
Authors: Jonah K\"omen, Edwin D. de Jong, Julius Hense, Hannah Marienwald, Jonas Dippel, Philip Naumann, Eric Marcus, Lukas Ruff, Maximilian Alber, Jonas Teuwen, Frederick Klauschen, Klaus-Robert M\"uller
Abstract: Biomedical Foundation Models (FMs) are rapidly transforming AI-enabled healthcare research and entering clinical validation. However, their susceptibility to learning non-biological technical features -- including variations in surgical/endoscopic techniques, laboratory procedures, and scanner hardware -- poses risks for clinical deployment. We present the first systematic investigation of pathology FM robustness to non-biological features. Our work (i) introduces measures to quantify FM robustness, (ii) demonstrates the consequences of limited robustness, and (iii) proposes a framework for FM robustification to mitigate these issues. Specifically, we developed PathoROB, a robustness benchmark with three novel metrics, including the robustness index, and four datasets covering 28 biological classes from 34 medical centers. Our experiments reveal robustness deficits across all 20 evaluated FMs, and substantial robustness differences between them. We found that non-robust FM representations can cause major diagnostic downstream errors and clinical blunders that prevent safe clinical adoption. Using more robust FMs and post-hoc robustification considerably reduced (but did not yet eliminate) the risk of such errors. This work establishes that robustness evaluation is essential for validating pathology FMs before clinical adoption and demonstrates that future FM development must integrate robustness as a core design principle. PathoROB provides a blueprint for assessing robustness across biomedical domains, guiding FM improvement efforts towards more robust, representative, and clinically deployable AI systems that prioritize biological information over technical artifacts.
Authors: Lijun Wu, Dong Hao, Zhiyi Fan
Abstract: Graph Neural Networks (GNNs) have achieved outstanding performance across a wide range of graph-related tasks. However, their "black-box" nature poses significant challenges to their explainability, and existing methods often fail to effectively capture the intricate interaction patterns among nodes within the network. In this work, we propose a novel explainability framework, GraphEXT, which leverages cooperative game theory and the concept of social externalities. GraphEXT partitions graph nodes into coalitions, decomposing the original graph into independent subgraphs. By integrating graph structure as an externality and incorporating the Shapley value under externalities, GraphEXT quantifies node importance through their marginal contributions to GNN predictions as the nodes transition between coalitions. Unlike traditional Shapley value-based methods that primarily focus on node attributes, our GraphEXT places greater emphasis on the interactions among nodes and the impact of structural changes on GNN predictions. Experimental studies on both synthetic and real-world datasets show that GraphEXT outperforms existing baseline methods in terms of fidelity across diverse GNN architectures , significantly enhancing the explainability of GNN models.
Authors: Rodrigo Moreira, Larissa F. Rodrigues Moreira, Fl\'avio de Oliveira Silva
Abstract: The deployment of large-scale software-based 5G core functions presents significant challenges due to their reliance on optimized and intelligent resource provisioning for their services. Many studies have focused on analyzing the impact of resource allocation for complex deployments using mathematical models, queue theories, or even Artificial Intelligence (AI). This paper elucidates the effects of chaotic workloads, generated by Distributed Denial of Service (DDoS) on different Network Functions (NFs) on User Equipment registration performance. Our findings highlight the necessity of diverse resource profiles to ensure Service-Level Agreement (SLA) compliance in large-scale 5G core deployments. Additionally, our analysis of packet capture approaches demonstrates the potential of kernel-based monitoring for scalable security threat defense. Finally, our empirical evaluation provides insights into the effective deployment of 5G NFs in complex scenarios.
Authors: Yao Fehlis, Charles Crain, Aidan Jensen, Michael Watson, James Juhasz, Paul Mandel, Betty Liu, Shawn Mahon, Daren Wilson, Nick Lynch-Jonely, Ben Leedom, David Fuller
Abstract: Building on the conceptual framework presented in our previous work on agentic AI for pharmaceutical research, this paper provides a comprehensive technical analysis of Tippy's multi-agent system implementation for drug discovery laboratory automation. We present a distributed microservices architecture featuring five specialized agents (Supervisor, Molecule, Lab, Analysis, and Report) that coordinate through OpenAI Agents SDK orchestration and access laboratory tools via the Model Context Protocol (MCP). The system architecture encompasses agent-specific tool integration, asynchronous communication patterns, and comprehensive configuration management through Git-based tracking. Our production deployment strategy utilizes Kubernetes container orchestration with Helm charts, Docker containerization, and CI/CD pipelines for automated testing and deployment. The implementation integrates vector databases for RAG functionality and employs an Envoy reverse proxy for secure external access. This work demonstrates how specialized AI agents can effectively coordinate complex laboratory workflows while maintaining security, scalability, reliability, and integration with existing laboratory infrastructure through standardized protocols.
Authors: Lifeng Chen, Jiner Wang, Zihao Pan, Beier Zhu, Xiaofeng Yang, Chi Zhang
Abstract: Recent advances in text-to-image (T2I) generation have led to impressive visual results. However, these models still face significant challenges when handling complex prompt, particularly those involving multiple subjects with distinct attributes. Inspired by the human drawing process, which first outlines the composition and then incrementally adds details, we propose Detail++, a training-free framework that introduces a novel Progressive Detail Injection (PDI) strategy to address this limitation. Specifically, we decompose a complex prompt into a sequence of simplified sub-prompts, guiding the generation process in stages. This staged generation leverages the inherent layout-controlling capacity of self-attention to first ensure global composition, followed by precise refinement. To achieve accurate binding between attributes and corresponding subjects, we exploit cross-attention mechanisms and further introduce a Centroid Alignment Loss at test time to reduce binding noise and enhance attribute consistency. Extensive experiments on T2I-CompBench and a newly constructed style composition benchmark demonstrate that Detail++ significantly outperforms existing methods, particularly in scenarios involving multiple objects and complex stylistic conditions.
Authors: Ko Watanabe. Stanislav Frolov. Adriano Lucieri. Andreas Dengel
Abstract: Recent advancements in Deep Learning and its application on the edge hold great potential for the revolution of routine screenings for skin cancers like Melanoma. Along with the anticipated benefits of this technology, potential dangers arise from unforseen and inherent biases. Thus, assessing and improving the fairness of such systems is of utmost importance. A key challenge in fairness assessment is to ensure that the evaluation dataset is sufficiently representative of different Personal Identifiable Information (PII) (sex, age, and race) and other minority groups. Against the backdrop of this challenge, this study leverages the state-of-the-art Generative AI (GenAI) LightningDiT model to assess the fairness of publicly available melanoma classifiers. The results suggest that fairness assessment using highly realistic synthetic data is a promising direction. Yet, our findings indicate that verifying fairness becomes difficult when the melanoma-detection model used for evaluation is trained on data that differ from the dataset underpinning the synthetic images. Nonetheless, we propose that our approach offers a valuable new avenue for employing synthetic data to gauge and enhance fairness in medical-imaging GenAI systems.
Authors: Milad Taghipour, Bane Vasic
Abstract: This paper explores the application of reinforcement learning techniques to enhance the performance of decoding of linear block codes based on flipping bits and finding optimal decisions. We describe the methodology for mapping the iterative decoding process into Markov Decision Processes (MDPs) and propose different methods to reduce the number of states in the MDP. A truncated MDP is proposed to reduce the number of states in the MDP by learning a Hamming ball with a specified radius around codewords. We then propose a general scheme for reinforcement learning based decoders applicable to any class of codes to improve the performance of decoders. We call this scheme an action-list decoding. We design an action-list decoder based on the Deep-Q network values that substantially enhance performance. We also get benefit of automorphism group of code to further improve the code performance. Additionally, we propose a feedback-based method to exploit and enhance the performance of existing high-performing decoders by applying reinforcement learning algorithms after the existing decoders. These approaches effectively reduces the complexity of the reinforcement learning block. Finally, we present experimental results for the Low-Density Parity Check (LDPC) codes over the Binary Symmetric Channel (BSC) to demonstrate the efficiency of the proposed methods.
Authors: Shubham Mohole, Sainyam Galhotra
Abstract: Application systems using natural language interfaces to databases (NLIDBs) have democratized data analysis. This positive development has also brought forth an urgent challenge to help users who might use these systems without a background in statistical analysis to formulate bias-free analytical questions. Although significant research has focused on text-to-SQL generation accuracy, addressing cognitive biases in analytical questions remains underexplored. We present VeriMinder, https://veriminder.ai, an interactive system for detecting and mitigating such analytical vulnerabilities. Our approach introduces three key innovations: (1) a contextual semantic mapping framework for biases relevant to specific analysis contexts (2) an analytical framework that operationalizes the Hard-to-Vary principle and guides users in systematic data analysis (3) an optimized LLM-powered system that generates high-quality, task-specific prompts using a structured process involving multiple candidates, critic feedback, and self-reflection. User testing confirms the merits of our approach. In direct user experience evaluation, 82.5% participants reported positively impacting the quality of the analysis. In comparative evaluation, VeriMinder scored significantly higher than alternative approaches, at least 20% better when considered for metrics of the analysis's concreteness, comprehensiveness, and accuracy. Our system, implemented as a web application, is set to help users avoid "wrong question" vulnerability during data analysis. VeriMinder code base with prompts, https://reproducibility.link/veriminder, is available as an MIT-licensed open-source software to facilitate further research and adoption within the community.
URLs: https://veriminder.ai,, https://reproducibility.link/veriminder,
Authors: Phu Thien Nguyen, Yousef Heider, Dennis M. Kochmann, Fadi Aldakheel
Abstract: The ultimate aim of the study is to explore the inverse design of porous metamaterials using a deep learning-based generative framework. Specifically, we develop a property-variational autoencoder (pVAE), a variational autoencoder (VAE) augmented with a regressor, to generate structured metamaterials with tailored hydraulic properties, such as porosity and permeability. While this work uses the lattice Boltzmann method (LBM) to generate intrinsic permeability tensor data for limited porous microstructures, a convolutional neural network (CNN) is trained using a bottom-up approach to predict effective hydraulic properties. This significantly reduces the computational cost compared to direct LBM simulations. The pVAE framework is trained on two datasets: a synthetic dataset of artificial porous microstructures and CT-scan images of volume elements from real open-cell foams. The encoder-decoder architecture of the VAE captures key microstructural features, mapping them into a compact and interpretable latent space for efficient structure-property exploration. The study provides a detailed analysis and interpretation of the latent space, demonstrating its role in structure-property mapping, interpolation, and inverse design. This approach facilitates the generation of new metamaterials with desired properties. The datasets and codes used in this study will be made open-access to support further research.
Authors: Jessica Quaye, Charvi Rastogi, Alicia Parrish, Oana Inel, Minsuk Kahng, Lora Aroyo, Vijay Janapa Reddi
Abstract: Text-to-image (T2I) models have become prevalent across numerous applications, making their robust evaluation against adversarial attacks a critical priority. Continuous access to new and challenging adversarial prompts across diverse domains is essential for stress-testing these models for resilience against novel attacks from multiple vectors. Current techniques for generating such prompts are either entirely authored by humans or synthetically generated. On the one hand, datasets of human-crafted adversarial prompts are often too small in size and imbalanced in their cultural and contextual representation. On the other hand, datasets of synthetically-generated prompts achieve scale, but typically lack the realistic nuances and creative adversarial strategies found in human-crafted prompts. To combine the strengths of both human and machine approaches, we propose Seed2Harvest, a hybrid red-teaming method for guided expansion of culturally diverse, human-crafted adversarial prompt seeds. The resulting prompts preserve the characteristics and attack patterns of human prompts while maintaining comparable average attack success rates (0.31 NudeNet, 0.36 SD NSFW, 0.12 Q16). Our expanded dataset achieves substantially higher diversity with 535 unique geographic locations and a Shannon entropy of 7.48, compared to 58 locations and 5.28 entropy in the original dataset. Our work demonstrates the importance of human-machine collaboration in leveraging human creativity and machine computational capacity to achieve comprehensive, scalable red-teaming for continuous T2I model safety evaluation.
Authors: Hongrong Yang, Markus Schlaepfer
Abstract: Accurate population flow prediction is essential for urban planning, transportation management, and public health. Yet existing methods face key limitations: traditional models rely on static spatial assumptions, deep learning models struggle with cross-city generalization, and Large Language Models (LLMs) incur high computational costs while failing to capture spatial structure. Moreover, many approaches sacrifice resolution by clustering Points of Interest (POIs) or restricting coverage to subregions, limiting their utility for city-wide analytics. We introduce UrbanPulse, a scalable deep learning framework that delivers ultra-fine-grained, city-wide OD flow predictions by treating each POI as an individual node. It combines a temporal graph convolutional encoder with a transformer-based decoder to model multi-scale spatiotemporal dependencies. To ensure robust generalization across urban contexts, UrbanPulse employs a three-stage transfer learning strategy: pretraining on large-scale urban graphs, cold-start adaptation, and reinforcement learning fine-tuning.Evaluated on over 103 million cleaned GPS records from three metropolitan areas in California, UrbanPulse achieves state-of-the-art accuracy and scalability. Through efficient transfer learning, UrbanPulse takes a key step toward making high-resolution, AI-powered urban forecasting deployable in practice across diverse cities.
Authors: Xiaoxu Guo, Siyan Liang, Yachao Cui, Juxiang Zhou, Lei Wang, Han Cao
Abstract: Accurately assessing post quality requires complex relational reasoning to capture nuanced topic-post relationships. However, existing studies face three major limitations: (1) treating the task as unimodal categorization, which fails to leverage multimodal cues and fine-grained quality distinctions; (2) introducing noise during deep multimodal fusion, leading to misleading signals; and (3) lacking the ability to capture complex semantic relationships like relevance and comprehensiveness. To address these issues, we propose the Multimodal Fine-grained Topic-post Relational Reasoning (MFTRR) framework, which mimics human cognitive processes. MFTRR reframes post-quality assessment as a ranking task and incorporates multimodal data to better capture quality variations. It consists of two key modules: (1) the Local-Global Semantic Correlation Reasoning Module, which models fine-grained semantic interactions between posts and topics at both local and global levels, enhanced by a maximum information fusion mechanism to suppress noise; and (2) the Multi-Level Evidential Relational Reasoning Module, which explores macro- and micro-level relational cues to strengthen evidence-based reasoning. We evaluate MFTRR on three newly constructed multimodal topic-post datasets and the public Lazada-Home dataset. Experimental results demonstrate that MFTRR significantly outperforms state-of-the-art baselines, achieving up to 9.52% NDCG@3 improvement over the best unimodal method on the Art History dataset.
Authors: Jaechul Roh, Zachary Novack, Yuefeng Peng, Niloofar Mireshghallah, Taylor Berg-Kirkpatrick, Amir Houmansadr
Abstract: Lyrics-to-Song (LS2) generation models promise end-to-end music synthesis from text, yet their vulnerability to training data memorization remains underexplored. We introduce Adversarial PhoneTic Prompting (APT), a novel attack where lyrics are semantically altered while preserving their acoustic structure through homophonic substitutions (e.g., Eminem's famous "mom's spaghetti" $\rightarrow$ "Bob's confetti"). Despite these distortions, we uncover a powerful form of sub-lexical memorization: models like SUNO and YuE regenerate outputs strikingly similar to known training content, achieving high similarity across audio-domain metrics, including CLAP, AudioJudge, and CoverID. This vulnerability persists across multiple languages and genres. More surprisingly, we discover that phoneme-altered lyrics alone can trigger visual memorization in text-to-video models. When prompted with phonetically modified lyrics from Lose Yourself, Veo 3 reconstructs visual elements from the original music video -- including character appearance and scene composition -- despite no visual cues in the prompt. We term this phenomenon phonetic-to-visual regurgitation. Together, these findings expose a critical vulnerability in transcript-conditioned multimodal generation: phonetic prompting alone can unlock memorized audiovisual content, raising urgent questions about copyright, safety, and content provenance in modern generative systems. Example generations are available on our demo page (jrohsc.github.io/music_attack/).
Authors: Amirarsalan Moatazedian, Yauhen Yakimenka, R\'emi A. Chou, J\"org Kliewer
Abstract: We study a privacy-preserving data-sharing setting where a privatizer transforms private data into a sanitized version observed by an authorized reconstructor and two unauthorized adversaries, each with access to side information correlated with the private data. The reconstructor is evaluated under a distortion function, while each adversary is evaluated using a separate loss function. The privatizer ensures the reconstructor distortion remains below a fixed threshold while maximizing the minimum loss across the two adversaries. This two-adversary setting models cases where individual users cannot reconstruct the data accurately, but their combined side information enables estimation within the distortion threshold. The privatizer maximizes individual loss while permitting accurate reconstruction only through collaboration. This echoes secret-sharing principles, but with lossy rather than perfect recovery. We frame this as a constrained data-driven minimax optimization problem and propose a data-driven training procedure that alternately updates the privatizer, reconstructor, and adversaries. We also analyze the Gaussian and binary cases as special scenarios where optimal solutions can be obtained. These theoretical optimal results are benchmarks for evaluating the proposed minimax training approach.
Authors: Hulayyil Alshammari, Praveen Rao
Abstract: Large language models (LLMs) have rapidly transformed the creation of written materials. LLMs have led to questions about writing integrity, thereby driving the creation of artificial intelligence (AI) detection technologies. Adversarial attacks, such as standard and humanized paraphrasing, inhibit detectors' ability to detect machine-generated text. Previous studies have mainly focused on ChatGPT and other well-known LLMs and have shown varying accuracy across detectors. However, there is a clear gap in the literature about DeepSeek, a recently published LLM. Therefore, in this work, we investigate whether six generally accessible AI detection tools -- AI Text Classifier, Content Detector AI, Copyleaks, QuillBot, GPT-2, and GPTZero -- can consistently recognize text generated by DeepSeek. The detectors were exposed to the aforementioned adversarial attacks. We also considered DeepSeek as a detector by performing few-shot prompting and chain-of-thought reasoning (CoT) for classifying AI and human-written text. We collected 49 human-authored question-answer pairs from before the LLM era and generated matching responses using DeepSeek-v3, producing 49 AI-generated samples. Then, we applied adversarial techniques such as paraphrasing and humanizing to add 196 more samples. These were used to challenge detector robustness and assess accuracy impact. While QuillBot and Copyleaks showed near-perfect performance on original and paraphrased DeepSeek text, others -- particularly AI Text Classifier and GPT-2 -- showed inconsistent results. The most effective attack was humanization, reducing accuracy to 71% for Copyleaks, 58% for QuillBot, and 52% for GPTZero. Few-shot and CoT prompting showed high accuracy, with the best five-shot result misclassifying only one of 49 samples (AI recall 96%, human recall 100%).
Authors: Shubham Mohole, Hongjun Choi, Shusen Liu, Christine Klymko, Shashank Kushwaha, Derek Shi, Wesam Sakla, Sainyam Galhotra, Ruben Glatt
Abstract: Retrieval-augmented generation (RAG) systems are increasingly adopted in clinical decision support, yet they remain methodologically blind-they retrieve evidence but cannot vet its scientific quality. A paper claiming "Antioxidant proteins decreased after alloferon treatment" and a rigorous multi-laboratory replication study will be treated as equally credible, even if the former lacked scientific rigor or was even retracted. To address this challenge, we introduce VERIRAG, a framework that makes three notable contributions: (i) the Veritable, an 11-point checklist that evaluates each source for methodological rigor, including data integrity and statistical validity; (ii) a Hard-to-Vary (HV) Score, a quantitative aggregator that weights evidence by its quality and diversity; and (iii) a Dynamic Acceptance Threshold, which calibrates the required evidence based on how extraordinary a claim is. Across four datasets-comprising retracted, conflicting, comprehensive, and settled science corpora-the VERIRAG approach consistently outperforms all baselines, achieving absolute F1 scores ranging from 0.53 to 0.65, representing a 10 to 14 point improvement over the next-best method in each respective dataset. We will release all materials necessary for reproducing our results.
Authors: Sohaib Imran, Ihor Kendiukhov, Matthew Broerman, Aditya Thomas, Riccardo Campanella, Rob Lamb, Peter M. Atkinson
Abstract: Do larger and more capable language models learn to update their "beliefs" about propositions more consistently with Bayes' theorem when presented with evidence in-context? To test this, we formulate a Bayesian Coherence Coefficient (BCC) metric and generate a dataset with which to measure the BCC. We measure BCC for multiple pre-trained-only language models across five model families, comparing against the number of model parameters, the amount of training data, and model scores on common benchmarks. Our results provide evidence for our hypothesis that larger and more capable pre-trained language models assign credences that are more coherent with Bayes' theorem. These results have important implications for our understanding and governance of LLMs.
Authors: Daniel Carlstrom Schad, Shrey Dixit, Janis Keck, Viktor Studenyak, Aleksandr Shpilevoi, Andrej Bicanski
Abstract: We present VIBE, a two-stage Transformer that fuses multi-modal video, audio, and text features to predict fMRI activity. Representations from open-source models (Qwen2.5, BEATs, Whisper, SlowFast, V-JEPA) are merged by a modality-fusion transformer and temporally decoded by a prediction transformer with rotary embeddings. Trained on 65 hours of movie data from the CNeuroMod dataset and ensembled across 20 seeds, VIBE attains mean parcel-wise Pearson correlations of 32.25 on in-distribution Friends S07 and 21.25 on six out-of-distribution films. An earlier iteration of the same architecture obtained 0.3198 and 0.2096, respectively, winning Phase-1 and placing second overall in the Algonauts 2025 Challenge.
Authors: Fitsum Gaim, Jong C. Park
Abstract: Despite being spoken by millions of people, Tigrinya remains severely underrepresented in Natural Language Processing (NLP) research. This work presents a comprehensive survey of NLP research for Tigrinya, analyzing over 40 studies spanning more than a decade of work from 2011 to 2025. We systematically review the current state of computational resources, models, and applications across ten distinct downstream tasks, including morphological processing, machine translation, speech recognition, and question-answering. Our analysis reveals a clear trajectory from foundational, rule-based systems to modern neural architectures, with progress consistently unlocked by resource creation milestones. We identify key challenges rooted in Tigrinya's morphological complexity and resource scarcity, while highlighting promising research directions, including morphology-aware modeling, cross-lingual transfer, and community-centered resource development. This work serves as both a comprehensive reference for researchers and a roadmap for advancing Tigrinya NLP. A curated metadata of the surveyed studies and resources is made publicly available.\footnote{Tigrinya NLP Anthology: https://github.com/fgaim/tigrinya-nlp-anthology.
Authors: Rui Deng, Ziqi Li, Mingshu Wang
Abstract: Accurate modeling and explaining geospatial tabular data (GTD) are critical for understanding geospatial phenomena and their underlying processes. Recent work has proposed a novel transformer-based deep learning model named GeoAggregator (GA) for this purpose, and has demonstrated that it outperforms other statistical and machine learning approaches. In this short paper, we further improve GA by 1) developing an optimized pipeline that accelerates the dataloading process and streamlines the forward pass of GA to achieve better computational efficiency; and 2) incorporating a model ensembling strategy and a post-hoc model explanation function based on the GeoShapley framework to enhance model explainability. We validate the functionality and efficiency of the proposed strategies by applying the improved GA model to synthetic datasets. Experimental results show that our implementation improves the prediction accuracy and inference speed of GA compared to the original implementation. Moreover, explanation experiments indicate that GA can effectively captures the inherent spatial effects in the designed synthetic dataset. The complete pipeline has been made publicly available for community use (https://github.com/ruid7181/GA-sklearn).
Authors: Paulo Mendes (GECAD, ISEP, Polytechnic of Porto, Portugal), Eva Maia (GECAD, ISEP, Polytechnic of Porto, Portugal), Isabel Pra\c{c}a (GECAD, ISEP, Polytechnic of Porto, Portugal)
Abstract: Phishing emails continue to pose a significant threat to cybersecurity by exploiting human vulnerabilities through deceptive content and malicious payloads. While Machine Learning (ML) models are effective at detecting phishing threats, their performance largely relies on the quality and diversity of the training data. This paper presents MeAJOR (Merged email Assets from Joint Open-source Repositories) Corpus, a novel, multi-source phishing email dataset designed to overcome critical limitations in existing resources. It integrates 135894 samples representing a broad number of phishing tactics and legitimate emails, with a wide spectrum of engineered features. We evaluated the dataset's utility for phishing detection research through systematic experiments with four classification models (RF, XGB, MLP, and CNN) across multiple feature configurations. Results highlight the dataset's effectiveness, achieving 98.34% F1 with XGB. By integrating broad features from multiple categories, our dataset provides a reusable and consistent resource, while addressing common challenges like class imbalance, generalisability and reproducibility.
Authors: Xin Wang (Jeff), R. Tyrrell Rockafellar (Jeff), Xuegang (Jeff), Ban
Abstract: Data-driven traffic state estimation and prediction (TSEP) relies heavily on data sources that contain sensitive information. While the abundance of data has fueled significant breakthroughs, particularly in machine learning-based methods, it also raises concerns regarding privacy, cybersecurity, and data freshness. These issues can erode public trust in intelligent transportation systems. Recently, regulations have introduced the "right to be forgotten", allowing users to request the removal of their private data from models. As machine learning models can remember old data, simply removing it from back-end databases is insufficient in such systems. To address these challenges, this study introduces a novel learning paradigm for TSEP-Machine Unlearning TSEP-which enables a trained TSEP model to selectively forget privacy-sensitive, poisoned, or outdated data. By empowering models to "unlearn," we aim to enhance the trustworthiness and reliability of data-driven traffic TSEP.
Authors: Alex Liu, Lief Esbenshade, Shawon Sarkar, Victor Tian, Zachary Zhang, Kevin He, Min Sun
Abstract: The integration of large language models (LLMs) into educational tools has the potential to substantially impact how teachers plan instruction, support diverse learners, and engage in professional reflection. Yet little is known about how educators actually use these tools in practice and how their interactions with AI can be meaningfully studied at scale. This paper presents a human-AI collaborative methodology for large-scale qualitative analysis of over 140,000 educator-AI messages drawn from a generative AI platform used by K-12 teachers. Through a four-phase coding pipeline, we combined inductive theme discovery, codebook development, structured annotation, and model benchmarking to examine patterns of educator engagement and evaluate the performance of LLMs in qualitative coding tasks. We developed a hierarchical codebook aligned with established teacher evaluation frameworks, capturing educators' instructional goals, contextual needs, and pedagogical strategies. Our findings demonstrate that LLMs, particularly Claude 3.5 Haiku, can reliably support theme identification, extend human recognition in complex scenarios, and outperform open-weight models in both accuracy and structural reliability. The analysis also reveals substantive patterns in how educators inquire AI to enhance instructional practices (79.7 percent of total conversations), create or adapt content (76.1 percent), support assessment and feedback loop (46.9 percent), attend to student needs for tailored instruction (43.3 percent), and assist other professional responsibilities (34.2 percent), highlighting emerging AI-related competencies that have direct implications for teacher preparation and professional development. This study offers a scalable, transparent model for AI-augmented qualitative research and provides foundational insights into the evolving role of generative AI in educational practice.
Authors: Jake R. Patock, Nicole Catherine Lewis, Kevin McCoy, Christina Gomez, Canling Chen, Lorenzo Luzi
Abstract: State-of-the-art (SOTA) image and text generation models are multimodal models that have many similarities to large language models (LLMs). Despite achieving strong performances, leading foundational multimodal model architectures frequently lag behind the architectural sophistication of contemporary LLMs. We propose GRR-CoCa, an improved SOTA Contrastive Captioner (CoCa) model that incorporates Gaussian error gated linear units, root mean squared normalization, and rotary positional embedding into the textual decoders and the vision transformer (ViT) encoder. Each architectural modification has been shown to improve model performance in LLMs, but has yet to be adopted in CoCa. We benchmarked GRR-CoCa against Baseline CoCa, a model with the same modified textual decoders but with CoCa's original ViT encoder. We used standard pretraining and fine-tuning workflows to benchmark the models on contrastive and generative tasks. Our GRR-CoCa significantly outperformed Baseline CoCa on the pretraining dataset and three diverse fine-tuning datasets. Pretraining improvements were 27.25% in contrastive loss, 3.71% in perplexity, and 7.15% in CoCa loss. The average fine-tuning improvements were 13.66% in contrastive loss, 5.18% in perplexity, and 5.55% in CoCa loss. We show that GRR-CoCa's modified architecture improves performance and generalization across vision-language domains.
Authors: Maria Vlachou
Abstract: In Conversational Recommendation Systems (CRS), a user provides feedback on recommended items at each turn, leading the CRS towards improved recommendations. Due to the need for a large amount of data, a user simulator is employed for both training and evaluation. Such user simulators critique the current retrieved item based on knowledge of a single target item. However, system evaluation in offline settings with simulators is limited by the focus on a single target item and their unlimited patience over a large number of turns. To overcome these limitations of existing simulators, we propose Fashion-AlterEval, a new dataset that contains human judgments for a selection of alternative items by adding new annotations in common fashion CRS datasets. Consequently, we propose two novel meta-user simulators that use the collected judgments and allow simulated users not only to express their preferences about alternative items to their original target, but also to change their mind and level of patience. In our experiments using the Shoes and Fashion IQ as the original datasets and three CRS models, we find that using the knowledge of alternatives by the simulator can have a considerable impact on the evaluation of existing CRS models, specifically that the existing single-target evaluation underestimates their effectiveness, and when simulatedusers are allowed to instead consider alternative relevant items, the system can rapidly respond to more quickly satisfy the user.
Authors: Weizhi Fei, Hao Shi, Jing Xu, Jingchen Peng, Jiazheng Li, Jingzhao Zhang, Bo Bai, Wei Han, Zhenyuan Chen, Xueyan Niu
Abstract: Efficiently editing knowledge stored in large language models (LLMs) enables model updates without large-scale training. One possible solution is Locate-and-Edit (L\&E), allowing simultaneous modifications of a massive number of facts. However, such editing may compromise the general abilities of LLMs and even result in forgetting edited facts when scaling up to thousands of edits. In this paper, we model existing linear L\&E methods as querying a Key-Value (KV) database. From this perspective, we then propose NeuralDB, an editing framework that explicitly represents the edited facts as a neural KV database equipped with a non-linear gated retrieval module, % In particular, our gated module only operates when inference involves the edited facts, effectively preserving the general abilities of LLMs. Comprehensive experiments involving the editing of 10,000 facts were conducted on the ZsRE and CounterFacts datasets, using GPT2-XL, GPT-J (6B) and Llama-3 (8B). The results demonstrate that NeuralDB not only excels in editing efficacy, generalization, specificity, fluency, and consistency, but also preserves overall performance across six representative text understanding and generation tasks. Further experiments indicate that NeuralDB maintains its effectiveness even when scaled to 100,000 facts (\textbf{50x} more than in prior work).
Authors: Ahmad ALBarqawi, Mahmoud Nazzal, Issa Khalil, Abdallah Khreishah, NhatHai Phan
Abstract: The rapid rise of deepfake technology, which produces realistic but fraudulent digital content, threatens the authenticity of media. Traditional deepfake detection approaches often struggle with sophisticated, customized deepfakes, especially in terms of generalization and robustness against malicious attacks. This paper introduces ViGText, a novel approach that integrates images with Vision Large Language Model (VLLM) Text explanations within a Graph-based framework to improve deepfake detection. The novelty of ViGText lies in its integration of detailed explanations with visual data, as it provides a more context-aware analysis than captions, which often lack specificity and fail to reveal subtle inconsistencies. ViGText systematically divides images into patches, constructs image and text graphs, and integrates them for analysis using Graph Neural Networks (GNNs) to identify deepfakes. Through the use of multi-level feature extraction across spatial and frequency domains, ViGText captures details that enhance its robustness and accuracy to detect sophisticated deepfakes. Extensive experiments demonstrate that ViGText significantly enhances generalization and achieves a notable performance boost when it detects user-customized deepfakes. Specifically, average F1 scores rise from 72.45% to 98.32% under generalization evaluation, and reflects the model's superior ability to generalize to unseen, fine-tuned variations of stable diffusion models. As for robustness, ViGText achieves an increase of 11.1% in recall compared to other deepfake detection approaches. When facing targeted attacks that exploit its graph-based architecture, ViGText limits classification performance degradation to less than 4%. ViGText uses detailed visual and textual analysis to set a new standard for detecting deepfakes, helping ensure media authenticity and information integrity.
Authors: Mingfeng Yuan, Letian Wang, Steven L. Waslander
Abstract: Pre-trained large language models (LLMs) have demonstrated strong common-sense reasoning abilities, making them promising for robotic navigation and planning tasks. However, despite recent progress, bridging the gap between language descriptions and actual robot actions in the open-world, beyond merely invoking limited predefined motion primitives, remains an open challenge. In this work, we aim to enable robots to interpret and decompose complex language instructions, ultimately synthesizing a sequence of trajectory points to complete diverse navigation tasks given open-set instructions and open-set objects. We observe that multi-modal large language models (MLLMs) exhibit strong cross-modal understanding when processing free-form language instructions, demonstrating robust scene comprehension. More importantly, leveraging their code-generation capability, MLLMs can interact with vision-language perception models to generate compositional 2D bird-eye-view value maps, effectively integrating semantic knowledge from MLLMs with spatial information from maps to reinforce the robot's spatial understanding. To further validate our approach, we effectively leverage large-scale autonomous vehicle datasets (AVDs) to validate our proposed zero-shot vision-language navigation framework in outdoor navigation tasks, demonstrating its capability to execute a diverse range of free-form natural language navigation instructions while maintaining robustness against object detection errors and linguistic ambiguities. Furthermore, we validate our system on a Husky robot in both indoor and outdoor scenes, demonstrating its real-world robustness and applicability. Supplementary videos are available at https://trailab.github.io/OpenNav-website/
Authors: Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal
Abstract: Inference-time steering methods offer a lightweight alternative to fine-tuning large language models (LLMs) and vision-language models (VLMs) by modifying internal activations at test time without updating model weights. However, most existing approaches rely on fixed, global intervention vectors, overlook the causal influence of individual input tokens, and fail to leverage informative gradients from the model's logits, particularly in multimodal settings where visual and textual inputs contribute unevenly. To address these limitations, we introduce GrAInS, an inference-time steering approach that operates across both language-only and vision-language models and tasks. GrAInS uses contrastive, gradient-based attribution via Integrated Gradients to identify the top-k most influential tokens, both positively and negatively attributed based on their contribution to preferred versus dispreferred outputs. These tokens are then used to construct directional steering vectors that capture semantic shifts from undesirable to desirable behavior. During inference, GrAInS adjusts hidden activations at transformer layers guided by token-level attribution signals, and normalizes activations to preserve representational scale. This enables fine-grained, interpretable, and modular control over model behavior, without retraining or auxiliary supervision. Empirically, GrAInS consistently outperforms both fine-tuning and existing steering baselines: it achieves a 13.22% accuracy gain on TruthfulQA using Llama-3.1-8B, reduces hallucination rates on MMHal-Bench from 0.624 to 0.514 with LLaVA-1.6-7B, and improves alignment win rates on SPA-VL by 8.11%, all while preserving the model's fluency and general capabilities.
Authors: Hoyeon Lee, Sejung Son, Ye-Eun Kang, Jong-Hwan Kim
Abstract: Current approaches to phrase break prediction address crucial prosodic aspects of text-to-speech systems but heavily rely on vast human annotations from audio or text, incurring significant manual effort and cost. Inherent variability in the speech domain, driven by phonetic factors, further complicates acquiring consistent, high-quality data. Recently, large language models (LLMs) have shown success in addressing data challenges in NLP by generating tailored synthetic data while reducing manual annotation needs. Motivated by this, we explore leveraging LLM to generate synthetic phrase break annotations, addressing the challenges of both manual annotation and speech-related tasks by comparing with traditional annotations and assessing effectiveness across multiple languages. Our findings suggest that LLM-based synthetic data generation effectively mitigates data challenges in phrase break prediction and highlights the potential of LLMs as a viable solution for the speech domain.
Authors: Hanwen Shen, Jiajie Lu, Yupeng Cao, Xiaonan Yang
Abstract: Recent advances in AI-generated video have shown strong performance on \emph{text-to-video} tasks, particularly for short clips depicting a single scene. However, current models struggle to generate longer videos with coherent scene transitions, primarily because they cannot infer when a transition is needed from the prompt. Most open-source models are trained on datasets consisting of single-scene video clips, which limits their capacity to learn and respond to prompts requiring multiple scenes. Developing scene transition awareness is essential for multi-scene generation, as it allows models to identify and segment videos into distinct clips by accurately detecting transitions. To address this, we propose the \textbf{Transition-Aware Video} (TAV) dataset, which consists of preprocessed video clips with multiple scene transitions. Our experiment shows that post-training on the \textbf{TAV} dataset improves prompt-based scene transition understanding, narrows the gap between required and generated scenes, and maintains image quality.
Authors: Zehan Li, Hongjie Chen, Yuxin Zhang, Jing Zhou, Xuening Wang, Hang Lv, Mengjie Du, Yaodong Song, Jie Lian, Jian Kang, Jie Li, Yongxiang Li, Zhongjiang He, Xuelong Li
Abstract: Spoken language models (SLMs) have seen rapid progress in recent years, along with the development of numerous benchmarks for evaluating their performance. However, most existing benchmarks primarily focus on evaluating whether SLMs can perform complex tasks comparable to those tackled by large language models (LLMs), often failing to align with how users naturally interact in real-world conversational scenarios. In this paper, we propose TELEVAL, a dynamic benchmark specifically designed to evaluate SLMs' effectiveness as conversational agents in realistic Chinese interactive settings. TELEVAL defines three evaluation dimensions: Explicit Semantics, Paralinguistic and Implicit Semantics, and System Abilities. It adopts a dialogue format consistent with real-world usage and evaluates text and audio outputs separately. TELEVAL particularly focuses on the model's ability to extract implicit cues from user speech and respond appropriately without additional instructions. Our experiments demonstrate that despite recent progress, existing SLMs still have considerable room for improvement in natural conversational tasks. We hope that TELEVAL can serve as a user-centered evaluation framework that directly reflects the user experience and contributes to the development of more capable dialogue-oriented SLMs.
Authors: Chujie Zheng, Shixuan Liu, Mingze Li, Xiong-Hui Chen, Bowen Yu, Chang Gao, Kai Dang, Yuqiong Liu, Rui Men, An Yang, Jingren Zhou, Junyang Lin
Abstract: This paper introduces Group Sequence Policy Optimization (GSPO), our stable, efficient, and performant reinforcement learning algorithm for training large language models. Unlike previous algorithms that adopt token-level importance ratios, GSPO defines the importance ratio based on sequence likelihood and performs sequence-level clipping, rewarding, and optimization. We demonstrate that GSPO achieves superior training efficiency and performance compared to the GRPO algorithm, notably stabilizes Mixture-of-Experts (MoE) RL training, and has the potential for simplifying the design of RL infrastructure. These merits of GSPO have contributed to the remarkable improvements in the latest Qwen3 models.
Authors: Pascal Spiegler, Taha Koleilat, Arash Harirpoush, Corey S. Miller, Hassan Rivaz, Marta Kersten-Oertel, Yiming Xiao
Abstract: Pancreatic cancer carries a poor prognosis and relies on endoscopic ultrasound (EUS) for targeted biopsy and radiotherapy. However, the speckle noise, low contrast, and unintuitive appearance of EUS make segmentation of pancreatic tumors with fully supervised deep learning (DL) models both error-prone and dependent on large, expert-curated annotation datasets. To address these challenges, we present TextSAM-EUS, a novel, lightweight, text-driven adaptation of the Segment Anything Model (SAM) that requires no manual geometric prompts at inference. Our approach leverages text prompt learning (context optimization) through the BiomedCLIP text encoder in conjunction with a LoRA-based adaptation of SAM's architecture to enable automatic pancreatic tumor segmentation in EUS, tuning only 0.86% of the total parameters. On the public Endoscopic Ultrasound Database of the Pancreas, TextSAM-EUS with automatic prompts attains 82.69% Dice and 85.28% normalized surface distance (NSD), and with manual geometric prompts reaches 83.10% Dice and 85.70% NSD, outperforming both existing state-of-the-art (SOTA) supervised DL models and foundation models (e.g., SAM and its variants). As the first attempt to incorporate prompt learning in SAM-based medical image segmentation, TextSAM-EUS offers a practical option for efficient and robust automatic EUS segmentation. Our code will be publicly available upon acceptance.
Authors: Ruizhe Chen, Zhiting Fan, Tianze Luo, Heqing Zou, Zhaopeng Feng, Guiyang Xie, Hansheng Zhang, Zhuochen Wang, Zuozhu Liu, Huaijian Zhang
Abstract: Video Temporal Grounding (VTG) aims to localize relevant temporal segments in videos given natural language queries. Despite recent progress with large vision-language models (LVLMs) and instruction-tuning, existing approaches often suffer from limited temporal awareness and poor generalization. In this work, we introduce a two-stage training framework that integrates supervised fine-tuning with reinforcement learning (RL) to improve both the accuracy and robustness of VTG models. Our approach first leverages high-quality curated cold start data for SFT initialization, followed by difficulty-controlled RL to further enhance temporal localization and reasoning abilities. Comprehensive experiments on multiple VTG benchmarks demonstrate that our method consistently outperforms existing models, particularly in challenging and open-domain scenarios. We conduct an in-depth analysis of training strategies and dataset curation, highlighting the importance of both high-quality cold start data and difficulty-controlled RL. To facilitate further research and industrial adoption, we release all intermediate datasets, models, and code to the community.
Authors: JinYoung Kim, DaeUng Jo, Kimin Yun, Jeonghyo Song, Youngjoon Yoo
Abstract: Estimating uncertainty from deep neural networks is a widely used approach for detecting out-of-distribution (OoD) samples, which typically exhibit high predictive uncertainty. However, conventional methods such as Monte Carlo (MC) Dropout often focus solely on either model or data uncertainty, failing to align with the semantic objective of OoD detection. To address this, we propose the Free-Energy Posterior Network, a novel framework that jointly models distributional uncertainty and identifying OoD and misclassified regions using free energy. Our method introduces two key contributions: (1) a free-energy-based density estimator parameterized by a Beta distribution, which enables fine-grained uncertainty estimation near ambiguous or unseen regions; and (2) a loss integrated within a posterior network, allowing direct uncertainty estimation from learned parameters without requiring stochastic sampling. By integrating our approach with the residual prediction branch (RPL) framework, the proposed method goes beyond post-hoc energy thresholding and enables the network to learn OoD regions by leveraging the variance of the Beta distribution, resulting in a semantically meaningful and computationally efficient solution for uncertainty-aware segmentation. We validate the effectiveness of our method on challenging real-world benchmarks, including Fishyscapes, RoadAnomaly, and Segment-Me-If-You-Can.
Authors: Binghua Li, Ziqing Chang, Tong Liang, Chao Li, Toshihisa Tanaka, Shigeki Aoki, Qibin Zhao, Zhe Sun
Abstract: We address the challenge of parameter-efficient fine-tuning (PEFT) for three-dimensional (3D) U-Net-based denoising diffusion probabilistic models (DDPMs) in magnetic resonance imaging (MRI) image generation. Despite its practical significance, research on parameter-efficient representations of 3D convolution operations remains limited. To bridge this gap, we propose Tensor Volumetric Operator (TenVOO), a novel PEFT method specifically designed for fine-tuning DDPMs with 3D convolutional backbones. Leveraging tensor network modeling, TenVOO represents 3D convolution kernels with lower-dimensional tensors, effectively capturing complex spatial dependencies during fine-tuning with few parameters. We evaluate TenVOO on three downstream brain MRI datasets-ADNI, PPMI, and BraTS2021-by fine-tuning a DDPM pretrained on 59,830 T1-weighted brain MRI scans from the UK Biobank. Our results demonstrate that TenVOO achieves state-of-the-art performance in multi-scale structural similarity index measure (MS-SSIM), outperforming existing approaches in capturing spatial dependencies while requiring only 0.3% of the trainable parameters of the original model. Our code is available at: https://github.com/xiaovhua/tenvoo
Authors: Hongjie Chen, Zehan Li, Yaodong Song, Wenming Deng, Yitong Yao, Yuxin Zhang, Hang Lv, Xuechao Zhu, Jian Kang, Jie Lian, Jie Li, Chao Wang, Shuangyong Song, Yongxiang Li, Zhongjiang He
Abstract: Recent advances in end-to-end spoken language models (SLMs) have significantly improved the ability of AI systems to engage in natural spoken interactions. However, most existing models treat speech merely as a vehicle for linguistic content, often overlooking the rich paralinguistic and speaker characteristic cues embedded in human speech, such as dialect, age, emotion, and non-speech vocalizations. In this work, we introduce GOAT-SLM, a novel spoken language model with paralinguistic and speaker characteristic awareness, designed to extend spoken language modeling beyond text semantics. GOAT-SLM adopts a dual-modality head architecture that decouples linguistic modeling from acoustic realization, enabling robust language understanding while supporting expressive and adaptive speech generation. To enhance model efficiency and versatility, we propose a modular, staged training strategy that progressively aligns linguistic, paralinguistic, and speaker characteristic information using large-scale speech-text corpora. Experimental results on TELEVAL, a multi-dimensional evaluation benchmark, demonstrate that GOAT-SLM achieves well-balanced performance across both semantic and non-semantic tasks, and outperforms existing open-source models in handling emotion, dialectal variation, and age-sensitive interactions. This work highlights the importance of modeling beyond linguistic content and advances the development of more natural, adaptive, and socially aware spoken language systems.
Authors: Juexin Zhang, Ying Weng, Ke Chen
Abstract: This paper introduces a novel approach to synthesize healthy 3D brain tissue from masked input images, specifically focusing on the task of 'ASNR-MICCAI BraTS Local Synthesis of Tissue via Inpainting'. Our proposed method employs a U-Net-based architecture, which is designed to effectively reconstruct the missing or corrupted regions of brain MRI scans. To enhance our model's generalization capabilities and robustness, we implement a comprehensive data augmentation strategy that involves randomly masking healthy images during training. Our model is trained on the BraTS-Local-Inpainting dataset and demonstrates the exceptional performance in recovering healthy brain tissue. The evaluation metrics employed, including Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR), and Mean Squared Error (MSE), consistently yields impressive results. On the BraTS-Local-Inpainting validation set, our model achieved an SSIM score of 0.841, a PSNR score of 23.257, and an MSE score of 0.007. Notably, these evaluation metrics exhibit relatively low standard deviations, i.e., 0.103 for SSIM score, 4.213 for PSNR score and 0.007 for MSE score, which indicates that our model's reliability and consistency across various input scenarios. Our method also secured first place in the challenge.
Authors: Juexin Zhang, Ying Weng, Ke Chen
Abstract: Glioblastoma, a highly aggressive brain tumor with diverse molecular and pathological features, poses a diagnostic challenge due to its heterogeneity. Accurate diagnosis and assessment of this heterogeneity are essential for choosing the right treatment and improving patient outcomes. Traditional methods rely on identifying specific features in tissue samples, but deep learning offers a promising approach for improved glioblastoma diagnosis. In this paper, we present our approach to the BraTS-Path Challenge 2024. We leverage a pre-trained model and fine-tune it on the BraTS-Path training dataset. Our model demonstrates poor performance on the challenging BraTS-Path validation set, as rigorously assessed by the Synapse online platform. The model achieves an accuracy of 0.392229, a recall of 0.392229, and a F1-score of 0.392229, indicating a consistent ability to correctly identify instances under the target condition. Notably, our model exhibits perfect specificity of 0.898704, showing an exceptional capacity to correctly classify negative cases. Moreover, a Matthews Correlation Coefficient (MCC) of 0.255267 is calculated, to signify a limited positive correlation between predicted and actual values and highlight our model's overall predictive power. Our solution also achieves the second place during the testing phase.
Authors: Gonzalo Cardenal Antolin, Jacques Fellay, Bashkim Jaha, Roger Kouyos, Niko Beerenwinkel, Diane Duroux
Abstract: Large language models (LLMs) are emerging as valuable tools to support clinicians in routine decision-making. HIV management is a compelling use case due to its complexity, including diverse treatment options, comorbidities, and adherence challenges. However, integrating LLMs into clinical practice raises concerns about accuracy, potential harm, and clinician acceptance. Despite their promise, AI applications in HIV care remain underexplored, and LLM benchmarking studies are scarce. This study evaluates the current capabilities of LLMs in HIV management, highlighting their strengths and limitations. We introduce HIVMedQA, a benchmark designed to assess open-ended medical question answering in HIV care. The dataset consists of curated, clinically relevant questions developed with input from an infectious disease physician. We evaluated seven general-purpose and three medically specialized LLMs, applying prompt engineering to enhance performance. Our evaluation framework incorporates both lexical similarity and an LLM-as-a-judge approach, extended to better reflect clinical relevance. We assessed performance across key dimensions: question comprehension, reasoning, knowledge recall, bias, potential harm, and factual accuracy. Results show that Gemini 2.5 Pro consistently outperformed other models across most dimensions. Notably, two of the top three models were proprietary. Performance declined as question complexity increased. Medically fine-tuned models did not always outperform general-purpose ones, and larger model size was not a reliable predictor of performance. Reasoning and comprehension were more challenging than factual recall, and cognitive biases such as recency and status quo were observed. These findings underscore the need for targeted development and evaluation to ensure safe, effective LLM integration in clinical care.
Authors: Riting Xia, Rucong Wang, Yulin Liu, Anchen Li, Xueyan Liu, Yan Zhang
Abstract: Class-imbalanced graph node classification is a practical yet underexplored research problem. Although recent studies have attempted to address this issue, they typically assume clean and reliable labels when processing class-imbalanced graphs. This assumption often violates the nature of real-world graphs, where labels frequently contain noise. Given this gap, this paper systematically investigates robust node classification for class-imbalanced graphs with noisy labels. We propose GraphALP, a novel Graph Augmentation framework based on Large language models (LLMs) and Pseudo-labeling techniques. Specifically, we design an LLM-based oversampling method to generate synthetic minority nodes, producing label-accurate minority nodes to alleviate class imbalance. Based on the class-balanced graphs, we develop a dynamically weighted pseudo-labeling method to obtain high-confidence pseudo labels to reduce label noise ratio. Additionally, we implement a secondary LLM-guided oversampling mechanism to mitigate potential class distribution skew caused by pseudo labels. Experimental results show that GraphALP achieves superior performance over state-of-the-art methods on class-imbalanced graphs with noisy labels.
Authors: Kexin Chen, Dongxia Wang, Yi Liu, Haonan Zhang, Wenhai Wang
Abstract: Despite the widespread use of Transformer-based text embedding models in NLP tasks, surprising 'sticky tokens' can undermine the reliability of embeddings. These tokens, when repeatedly inserted into sentences, pull sentence similarity toward a certain value, disrupting the normal distribution of embedding distances and degrading downstream performance. In this paper, we systematically investigate such anomalous tokens, formally defining them and introducing an efficient detection method, Sticky Token Detector (STD), based on sentence and token filtering. Applying STD to 40 checkpoints across 14 model families, we discover a total of 868 sticky tokens. Our analysis reveals that these tokens often originate from special or unused entries in the vocabulary, as well as fragmented subwords from multilingual corpora. Notably, their presence does not strictly correlate with model size or vocabulary size. We further evaluate how sticky tokens affect downstream tasks like clustering and retrieval, observing significant performance drops of up to 50%. Through attention-layer analysis, we show that sticky tokens disproportionately dominate the model's internal representations, raising concerns about tokenization robustness. Our findings show the need for better tokenization strategies and model design to mitigate the impact of sticky tokens in future text embedding applications.
Authors: Dhruv Jain, Romain Modzelewski, Romain H\'erault, Clement Chatelain, Eva Torfeh, Sebastien Thureau
Abstract: In data-scarce scenarios, deep learning models often overfit to noise and irrelevant patterns, which limits their ability to generalize to unseen samples. To address these challenges in medical image segmentation, we introduce Diff-UMamba, a novel architecture that combines the UNet framework with the mamba mechanism for modeling long-range dependencies. At the heart of Diff-UMamba is a Noise Reduction Module (NRM), which employs a signal differencing strategy to suppress noisy or irrelevant activations within the encoder. This encourages the model to filter out spurious features and enhance task-relevant representations, thereby improving its focus on clinically meaningful regions. As a result, the architecture achieves improved segmentation accuracy and robustness, particularly in low-data settings. Diff-UMamba is evaluated on multiple public datasets, including MSD (lung and pancreas) and AIIB23, demonstrating consistent performance gains of 1-3% over baseline methods across diverse segmentation tasks. To further assess performance under limited-data conditions, additional experiments are conducted on the BraTS-21 dataset by varying the proportion of available training samples. The approach is also validated on a small internal non-small cell lung cancer (NSCLC) dataset for gross tumor volume (GTV) segmentation in cone beam CT (CBCT), where it achieves a 4-5% improvement over the baseline.
Authors: Wonjun Jeong, Dongseok Kim, Taegkeun Whangbo
Abstract: Large Language Models (LLMs) can achieve inflated scores on multiple-choice tasks by exploiting inherent biases in option positions or labels, rather than demonstrating genuine understanding. This study introduces SCOPE, an evaluation framework designed to measure and mitigate such selection bias in a dataset-independent manner. By repeatedly invoking a null prompt that lacks semantic content, SCOPE estimates each model's unique position-bias distribution. It then redistributes the answer slot according to the inverse-bias distribution, thereby equalizing the lucky-rate, the probability of selecting the correct answer by chance. Furthermore, it prevents semantically similar distractors from being placed adjacent to the answer, thereby blocking near-miss guesses based on superficial proximity cues. Across multiple benchmark experiments, SCOPE consistently outperformed existing debiasing methods in terms of stable performance improvements and showed clearer confidence distributions over correct options. This framework thus offers a new standard for enhancing the fairness and reliability of LLM evaluations.
Authors: San Kim, Jonghwi Kim, Yejin Jeon, Gary Geunbae Lee
Abstract: Retrieval-Augmented Generation (RAG) enhances Large Language Models (LLMs) by providing external knowledge for accurate and up-to-date responses. However, this reliance on external sources exposes a security risk, attackers can inject poisoned documents into the knowledge base to steer the generation process toward harmful or misleading outputs. In this paper, we propose Gradient-based Masked Token Probability (GMTP), a novel defense method to detect and filter out adversarially crafted documents. Specifically, GMTP identifies high-impact tokens by examining gradients of the retriever's similarity function. These key tokens are then masked, and their probabilities are checked via a Masked Language Model (MLM). Since injected tokens typically exhibit markedly low masked-token probabilities, this enables GMTP to easily detect malicious documents and achieve high-precision filtering. Experiments demonstrate that GMTP is able to eliminate over 90% of poisoned content while retaining relevant documents, thus maintaining robust retrieval and generation performance across diverse datasets and adversarial settings.
Authors: Arup Kumar Sahoo, Itzik Klein
Abstract: A fundamental requirement for full autonomy in mobile robots is accurate navigation even in situations where satellite navigation or cameras are unavailable. In such practical situations, relying only on inertial sensors will result in navigation solution drift due to the sensors' inherent noise and error terms. One of the emerging solutions to mitigate drift is to maneuver the robot in a snake-like slithering motion to increase the inertial signal-to-noise ratio, allowing the regression of the mobile robot position. In this work, we propose MoRPI-PINN as a physics-informed neural network framework for accurate inertial-based mobile robot navigation. By embedding physical laws and constraints into the training process, MoRPI-PINN is capable of providing an accurate and robust navigation solution. Using real-world experiments, we show accuracy improvements of over 85% compared to other approaches. MoRPI-PINN is a lightweight approach that can be implemented even on edge devices and used in any typical mobile robot application.
Authors: Chang Gong, Zhongwen Li, Xiaoqi Li
Abstract: Information security is facing increasingly severe challenges, and traditional protection means are difficult to cope with complex and changing threats. In recent years, as an emerging intelligent technology, large language models (LLMs) have shown a broad application prospect in the field of information security. In this paper, we focus on the key role of LLM in information security, systematically review its application progress in malicious behavior prediction, network threat analysis, system vulnerability detection, malicious code identification, and cryptographic algorithm optimization, and explore its potential in enhancing security protection performance. Based on neural networks and Transformer architecture, this paper analyzes the technical basis of large language models and their advantages in natural language processing tasks. It is shown that the introduction of large language modeling helps to improve the detection accuracy and reduce the false alarm rate of security systems. Finally, this paper summarizes the current application results and points out that it still faces challenges in model transparency, interpretability, and scene adaptability, among other issues. It is necessary to explore further the optimization of the model structure and the improvement of the generalization ability to realize a more intelligent and accurate information security protection system.
Authors: Zhongzheng Yuan, Lianshuai Guo, Xunkai Li, Yinlin Zhu, Wenyu Wang, Meixia Qu
Abstract: Federated Graph Learning (FGL) is a distributed learning paradigm that enables collaborative training over large-scale subgraphs located on multiple local systems. However, most existing FGL approaches rely on synchronous communication, which leads to inefficiencies and is often impractical in real-world deployments. Meanwhile, current asynchronous federated learning (AFL) methods are primarily designed for conventional tasks such as image classification and natural language processing, without accounting for the unique topological properties of graph data. Directly applying these methods to graph learning can possibly result in semantic drift and representational inconsistency in the global model. To address these challenges, we propose FedSA-GCL, a semi-asynchronous federated framework that leverages both inter-client label distribution divergence and graph topological characteristics through a novel ClusterCast mechanism for efficient training. We evaluate FedSA-GCL on multiple real-world graph datasets using the Louvain and Metis split algorithms, and compare it against 9 baselines. Extensive experiments demonstrate that our method achieves strong robustness and outstanding efficiency, outperforming the baselines by an average of 2.92% with the Louvain and by 3.4% with the Metis.
Authors: Nenad Petrovic, Fengjunjie Pan, Vahid Zolfaghari, Krzysztof Lebioda, Andre Schamschurko, Alois Knoll
Abstract: This paper introduces a GenAI-empowered approach to automated development of automotive software, with emphasis on autonomous and Advanced Driver Assistance Systems (ADAS) capabilities. The process starts with requirements as input, while the main generated outputs are test scenario code for simulation environment, together with implementation of desired ADAS capabilities targeting hardware platform of the vehicle connected to testbench. Moreover, we introduce additional steps for requirements consistency checking leveraging Model-Driven Engineering (MDE). In the proposed workflow, Large Language Models (LLMs) are used for model-based summarization of requirements (Ecore metamodel, XMI model instance and OCL constraint creation), test scenario generation, simulation code (Python) and target platform code generation (C++). Additionally, Retrieval Augmented Generation (RAG) is adopted to enhance test scenario generation from autonomous driving regulations-related documents. Our approach aims shorter compliance and re-engineering cycles, as well as reduced development and testing time when it comes to ADAS-related capabilities.
Authors: Zeqiang Zhang, Ruxin Chen
Abstract: The application of Reinforcement Learning (RL) to economic modeling reveals a fundamental conflict between the assumptions of equilibrium theory and the emergent behavior of learning agents. While canonical economic models assume atomistic agents act as `takers' of aggregate market conditions, a naive single-agent RL simulation incentivizes the agent to become a `manipulator' of its environment. This paper first demonstrates this discrepancy within a search-and-matching model with concave production, showing that a standard RL agent learns a non-equilibrium, monopsonistic policy. Additionally, we identify a parametric bias arising from the mismatch between economic discounting and RL's treatment of intertemporal costs. To address both issues, we propose a calibrated Mean-Field Reinforcement Learning framework that embeds a representative agent in a fixed macroeconomic field and adjusts the cost function to reflect economic opportunity costs. Our iterative algorithm converges to a self-consistent fixed point where the agent's policy aligns with the competitive equilibrium. This approach provides a tractable and theoretically sound methodology for modeling learning agents in economic systems within the broader domain of computational social science.
Authors: Longjian Zeng, Zunjie Zhu, Rongfeng Lu, Ming Lu, Bolun Zheng, Chenggang Yan, Anke Xue
Abstract: In recent years, foundation models for monocular depth estimation have received increasing attention. Current methods mainly address typical daylight conditions, but their effectiveness notably decreases in low-light environments. There is a lack of robust foundational models for monocular depth estimation specifically designed for low-light scenarios. This largely stems from the absence of large-scale, high-quality paired depth datasets for low-light conditions and the effective parameter-efficient fine-tuning (PEFT) strategy. To address these challenges, we propose DepthDark, a robust foundation model for low-light monocular depth estimation. We first introduce a flare-simulation module and a noise-simulation module to accurately simulate the imaging process under nighttime conditions, producing high-quality paired depth datasets for low-light conditions. Additionally, we present an effective low-light PEFT strategy that utilizes illumination guidance and multiscale feature fusion to enhance the model's capability in low-light environments. Our method achieves state-of-the-art depth estimation performance on the challenging nuScenes-Night and RobotCar-Night datasets, validating its effectiveness using limited training data and computing resources.
Authors: Dongyang Guo, Yasmeen Abdrabou, Enkeleda Thaqi, Enkelejda Kasneci
Abstract: Eye-tracking data reveals valuable insights into users' cognitive states but is difficult to analyze due to its structured, non-linguistic nature. While large language models (LLMs) excel at reasoning over text, they struggle with temporal and numerical data. This paper presents a multimodal human-AI collaborative framework designed to enhance cognitive pattern extraction from eye-tracking signals. The framework includes: (1) a multi-stage pipeline using horizontal and vertical segmentation alongside LLM reasoning to uncover latent gaze patterns; (2) an Expert-Model Co-Scoring Module that integrates expert judgment with LLM output to generate trust scores for behavioral interpretations; and (3) a hybrid anomaly detection module combining LSTM-based temporal modeling with LLM-driven semantic analysis. Our results across several LLMs and prompt strategies show improvements in consistency, interpretability, and performance, with up to 50% accuracy in difficulty prediction tasks. This approach offers a scalable, interpretable solution for cognitive modeling and has broad potential in adaptive learning, human-computer interaction, and educational analytics.
Authors: Junyao Li, Yahao Lu, Xingyuan Guo, Xiaoyu Xian, Tiantian Wang, Yukai Shi
Abstract: Infrared small target detection (ISTD) plays a vital role in numerous practical applications. In pursuit of determining the performance boundaries, researchers employ large and expensive manual-labeling data for representation learning. Nevertheless, this approach renders the state-of-the-art ISTD methods highly fragile in real-world challenges. In this paper, we first study the variation in detection performance across several mainstream methods under various scarcity -- namely, the absence of high-quality infrared data -- that challenge the prevailing theories about practical ISTD. To address this concern, we introduce the Gaussian Agnostic Representation Learning. Specifically, we propose the Gaussian Group Squeezer, leveraging Gaussian sampling and compression for non-uniform quantization. By exploiting a diverse array of training samples, we enhance the resilience of ISTD models against various challenges. Then, we introduce two-stage diffusion models for real-world reconstruction. By aligning quantized signals closely with real-world distributions, we significantly elevate the quality and fidelity of the synthetic samples. Comparative evaluations against state-of-the-art detection methods in various scarcity scenarios demonstrate the efficacy of the proposed approach.
Authors: Chenyu Su, Weiwei Shang, Chen Qian, Fei Zhang, Shuang Cong
Abstract: Semantics-driven 3D spatial constraints align highlevel semantic representations with low-level action spaces, facilitating the unification of task understanding and execution in robotic manipulation. The synergistic reasoning of Multimodal Large Language Models (MLLMs) and Vision Foundation Models (VFMs) enables cross-modal 3D spatial constraint construction. Nevertheless, existing methods have three key limitations: (1) coarse semantic granularity in constraint modeling, (2) lack of real-time closed-loop planning, (3) compromised robustness in semantically diverse environments. To address these challenges, we propose ReSem3D, a unified manipulation framework for semantically diverse environments, leveraging the synergy between VFMs and MLLMs to achieve fine-grained visual grounding and dynamically constructs hierarchical 3D spatial constraints for real-time manipulation. Specifically, the framework is driven by hierarchical recursive reasoning in MLLMs, which interact with VFMs to automatically construct 3D spatial constraints from natural language instructions and RGB-D observations in two stages: part-level extraction and region-level refinement. Subsequently, these constraints are encoded as real-time optimization objectives in joint space, enabling reactive behavior to dynamic disturbances. Extensive simulation and real-world experiments are conducted in semantically rich household and sparse chemical lab environments. The results demonstrate that ReSem3D performs diverse manipulation tasks under zero-shot conditions, exhibiting strong adaptability and generalization. Code and videos at https://resem3d.github.io.
Authors: Suhang Wu, Jialong Tang, Chengyi Yang, Pei Zhang, Baosong Yang, Junhui Li, Junfeng Yao, Min Zhang, Jinsong Su
Abstract: Direct speech translation (ST) has garnered increasing attention nowadays, yet the accurate translation of terminology within utterances remains a great challenge. In this regard, current studies mainly concentrate on leveraging various translation knowledge into ST models. However, these methods often struggle with interference from irrelevant noise and can not fully utilize the translation knowledge. To address these issues, in this paper, we propose a novel Locate-and-Focus method for terminology translation. It first effectively locates the speech clips containing terminologies within the utterance to construct translation knowledge, minimizing irrelevant information for the ST model. Subsequently, it associates the translation knowledge with the utterance and hypothesis from both audio and textual modalities, allowing the ST model to better focus on translation knowledge during translation. Experimental results across various datasets demonstrate that our method effectively locates terminologies within utterances and enhances the success rate of terminology translation, while maintaining robust general translation performance.
Authors: Xuebo Jin, Longfei Gao, Anshuo Tong, Zhengyang Chen, Jianlei Kong, Ning Sun, Huijun Ma, Qiang Wang, Yuting Bai, Tingli Su
Abstract: Traditional Chinese medicine (TCM) tongue diagnosis, while clinically valuable, faces standardization challenges due to subjective interpretation and inconsistent imaging protocols, compounded by the lack of large-scale, annotated datasets for AI development. To address this gap, we present the first specialized dataset for AI-driven TCM tongue diagnosis, comprising 6,719 high-quality images captured under standardized conditions and annotated with 20 pathological symptom categories (averaging 2.54 clinically validated labels per image, all verified by licensed TCM practitioners). The dataset supports multiple annotation formats (COCO, TXT, XML) for broad usability and has been benchmarked using nine deep learning models (YOLOv5/v7/v8 variants, SSD, and MobileNetV2) to demonstrate its utility for AI development. This resource provides a critical foundation for advancing reliable computational tools in TCM, bridging the data shortage that has hindered progress in the field, and facilitating the integration of AI into both research and clinical practice through standardized, high-quality diagnostic data.
Authors: Delong Ran, Xinlei He, Tianshuo Cong, Anyu Wang, Qi Li, Xiaoyun Wang
Abstract: Language Models (LMs) typically adhere to a "pre-training and fine-tuning" paradigm, where a universal pre-trained model can be fine-tuned to cater to various specialized domains. Low-Rank Adaptation (LoRA) has gained the most widespread use in LM fine-tuning due to its lightweight computational cost and remarkable performance. Because the proportion of parameters tuned by LoRA is relatively small, there might be a misleading impression that the LoRA fine-tuning data is invulnerable to Membership Inference Attacks (MIAs). However, we identify that utilizing the pre-trained model can induce more information leakage, which is neglected by existing MIAs. Therefore, we introduce LoRA-Leak, a holistic evaluation framework for MIAs against the fine-tuning datasets of LMs. LoRA-Leak incorporates fifteen membership inference attacks, including ten existing MIAs, and five improved MIAs that leverage the pre-trained model as a reference. In experiments, we apply LoRA-Leak to three advanced LMs across three popular natural language processing tasks, demonstrating that LoRA-based fine-tuned LMs are still vulnerable to MIAs (e.g., 0.775 AUC under conservative fine-tuning settings). We also applied LoRA-Leak to different fine-tuning settings to understand the resulting privacy risks. We further explore four defenses and find that only dropout and excluding specific LM layers during fine-tuning effectively mitigate MIA risks while maintaining utility. We highlight that under the "pre-training and fine-tuning" paradigm, the existence of the pre-trained model makes MIA a more severe risk for LoRA-based LMs. We hope that our findings can provide guidance on data privacy protection for specialized LM providers.
Authors: Minje Park, Jeonghwa Lim, Taehyung Yu, Sunghoon Joo
Abstract: Electrocardiogram (ECG) delineation, the segmentation of meaningful waveform features, is critical for clinical diagnosis. Despite recent advances using deep learning, progress has been limited by the scarcity of publicly available annotated datasets. Semi-supervised learning presents a promising solution by leveraging abundant unlabeled ECG data. In this study, we present the first systematic benchmark for semi-supervised semantic segmentation (SemiSeg) in ECG delineation. We curated and unified multiple public datasets, including previously underused sources, to support robust and diverse evaluation. We adopted five representative SemiSeg algorithms from computer vision, implemented them on two different architectures: the convolutional network and the transformer, and evaluated them in two different settings: in-domain and cross-domain. Additionally, we propose ECG-specific training configurations and augmentation strategies and introduce a standardized evaluation framework. Our results show that the transformer outperforms the convolutional network in semi-supervised ECG delineation. We anticipate that our benchmark will serve as a foundation for advancing semi-supervised ECG delineation methods and will facilitate further research in this domain.
Authors: Lars Ullrich, Michael Buchholz, Jonathan Petit, Klaus Dietmayer, Knut Graichen
Abstract: Efficient scalability of automated driving (AD) is key to reducing costs, enhancing safety, conserving resources, and maximizing impact. However, research focuses on specific vehicles and context, while broad deployment requires scalability across various configurations and environments. Differences in vehicle types, sensors, actuators, but also traffic regulations, legal requirements, cultural dynamics, or even ethical paradigms demand high flexibility of data-driven developed capabilities. In this paper, we address the challenge of scalable adaptation of generic capabilities to desired systems and environments. Our concept follows a two-stage fine-tuning process. In the first stage, fine-tuning to the specific environment takes place through a country-specific reward model that serves as an interface between technological adaptations and socio-political requirements. In the second stage, vehicle-specific transfer learning facilitates system adaptation and governs the validation of design decisions. In sum, our concept offers a data-driven process that integrates both technological and socio-political aspects, enabling effective scalability across technical, legal, cultural, and ethical differences.
Authors: Ezhini Rasendiran R, Chandresh Kumar Maurya
Abstract: We address the problem of classifying bird species using their song recordings, a challenging task due to environmental noise, overlapping vocalizations, and missing labels. Existing models struggle with low-SNR or multi-species recordings. We hypothesize that birds can be classified by visualizing their pitch pattern, speed, and repetition, collectively called motifs. Deep learning models applied to spectrogram images help, but similar motifs across species cause confusion. To mitigate this, we embed frequency information into spectrograms using primary color additives. This enhances species distinction and improves classification accuracy. Our experiments show that the proposed approach achieves statistically significant gains over models without colorization and surpasses the BirdCLEF 2024 winner, improving F1 by 7.3%, ROC-AUC by 6.2%, and CMAP by 6.6%. These results demonstrate the effectiveness of incorporating frequency information via colorization.
Authors: Asaf Yehudai, Lilach Eden, Yotam Perlitz, Roy Bar-Haim, Michal Shmueli-Scheuer
Abstract: The evaluation of Large Language Models (LLMs) increasingly relies on other LLMs acting as judges. However, current evaluation paradigms typically yield a single score or ranking, answering which model is better but not why. While essential for benchmarking, these top-level scores obscure the specific, actionable reasons behind a model's performance. To bridge this gap, we introduce CLEAR, an interactive, open-source package for LLM-based error analysis. CLEAR first generates per-instance textual feedback, then it creates a set of system-level error issues, and quantifies the prevalence of each identified issue. Our package also provides users with an interactive dashboard that allows for a comprehensive error analysis through aggregate visualizations, applies interactive filters to isolate specific issues or score ranges, and drills down to the individual instances that exemplify a particular behavioral pattern. We demonstrate CLEAR analysis for RAG and Math benchmarks, and showcase its utility through a user case study.
Authors: Rana Alshaikh, Israa Alghanmi, Shelan Jeawak
Abstract: The cognitive and reasoning abilities of large language models (LLMs) have enabled remarkable progress in natural language processing. However, their performance in interpreting structured data, especially in tabular formats, remains limited. Although benchmarks for English tabular data are widely available, Arabic is still underrepresented because of the limited availability of public resources and its unique language features. To address this gap, we present AraTable, a novel and comprehensive benchmark designed to evaluate the reasoning and understanding capabilities of LLMs when applied to Arabic tabular data. AraTable consists of various evaluation tasks, such as direct question answering, fact verification, and complex reasoning, involving a wide range of Arabic tabular sources. Our methodology follows a hybrid pipeline, where initial content is generated by LLMs and subsequently filtered and verified by human experts to ensure high dataset quality. Initial analyses using AraTable show that, while LLMs perform adequately on simpler tabular tasks such as direct question answering, they continue to face significant cognitive challenges when tasks require deeper reasoning and fact verification. This indicates that there are substantial opportunities for future work to improve performance on complex tabular reasoning tasks. We also propose a fully automated evaluation framework that uses a self-deliberation mechanism and achieves performance nearly identical to that of human judges. This research provides a valuable, publicly available resource and evaluation framework that can help accelerate the development of foundational models for processing and analysing Arabic structured data.
Authors: Md Obyedullahil Mamun, Md Adyelullahil Mamun, Arif Ahmad, Md. Imran Hossain Emu
Abstract: Punctuation restoration enhances the readability of text and is critical for post-processing tasks in Automatic Speech Recognition (ASR), especially for low-resource languages like Bangla. In this study, we explore the application of transformer-based models, specifically XLM-RoBERTa-large, to automatically restore punctuation in unpunctuated Bangla text. We focus on predicting four punctuation marks: period, comma, question mark, and exclamation mark across diverse text domains. To address the scarcity of annotated resources, we constructed a large, varied training corpus and applied data augmentation techniques. Our best-performing model, trained with an augmentation factor of alpha = 0.20%, achieves an accuracy of 97.1% on the News test set, 91.2% on the Reference set, and 90.2% on the ASR set. Results show strong generalization to reference and ASR transcripts, demonstrating the model's effectiveness in real-world, noisy scenarios. This work establishes a strong baseline for Bangla punctuation restoration and contributes publicly available datasets and code to support future research in low-resource NLP.
Authors: Sizhe Ma, Katherine A. Flanigan, Mario Berg\'es
Abstract: The advancement of the Internet of Things (IoT) and Artificial Intelligence has catalyzed the evolution of Digital Twins (DTs) from conceptual ideas to more implementable realities. Yet, transitioning from academia to industry is complex due to the absence of standardized frameworks. This paper builds upon the authors' previously established functional and informational requirements supporting standardized DT development, focusing on a crucial aspect: transferability. While existing DT research primarily centers on asset transfer, the significance of "sim-to-real transfer" and "real-to-sim transfer"--transferring knowledge between simulations and real-world operations--is vital for comprehensive lifecycle management in DTs. A key challenge in this process is calibrating the "reality gap," the discrepancy between simulated predictions and actual outcomes. Our research investigates the impact of integrating a single Reality Gap Analysis (RGA) module into an existing DT framework to effectively manage both sim-to-real and real-to-sim transfers. This integration is facilitated by data pipelines that connect the RGA module with the existing components of the DT framework, including the historical repository and the simulation model. A case study on a pedestrian bridge at Carnegie Mellon University showcases the performance of different levels of integration of our approach with an existing framework. With full implementation of an RGA module and a complete data pipeline, our approach is capable of bidirectional knowledge transfer between simulations and real-world operations without compromising efficiency.
Authors: Basel Alshaikhdeeb, Ahmed Abdelmonem Hemedan, Soumyabrata Ghosh, Irina Balaur, Venkata Satagopam
Abstract: Generating clinical synthetic text represents an effective solution for common clinical NLP issues like sparsity and privacy. This paper aims to conduct a systematic review on generating synthetic medical free-text by formulating quantitative analysis to three research questions concerning (i) the purpose of generation, (ii) the techniques, and (iii) the evaluation methods. We searched PubMed, ScienceDirect, Web of Science, Scopus, IEEE, Google Scholar, and arXiv databases for publications associated with generating synthetic medical unstructured free-text. We have identified 94 relevant articles out of 1,398 collected ones. A great deal of attention has been given to the generation of synthetic medical text from 2018 onwards, where the main purpose of such a generation is towards text augmentation, assistive writing, corpus building, privacy-preserving, annotation, and usefulness. Transformer architectures were the main predominant technique used to generate the text, especially the GPTs. On the other hand, there were four main aspects of evaluation, including similarity, privacy, structure, and utility, where utility was the most frequent method used to assess the generated synthetic medical text. Although the generated synthetic medical text demonstrated a moderate possibility to act as real medical documents in different downstream NLP tasks, it has proven to be a great asset as augmented, complementary to the real documents, towards improving the accuracy and overcoming sparsity/undersampling issues. Yet, privacy is still a major issue behind generating synthetic medical text, where more human assessments are needed to check for the existence of any sensitive information. Despite that, advances in generating synthetic medical text will considerably accelerate the adoption of workflows and pipeline development, discarding the time-consuming legalities of data transfer.
Authors: Juntao Zhao, Jiuru Li, Chuan Wu
Abstract: Utilizing CPUs to serve large language models (LLMs) is a resource-friendly alternative to GPU serving. Existing CPU-based solutions ignore workload differences between the prefill and the decode phases of LLM inference, applying a static per-NUMA (Non-Uniform Memory Access) node model partition and utilizing vendor libraries for operator-level execution, which is suboptimal. We propose Sandwich, a hardware-centric CPU-based LLM serving engine that uses different execution plans for the prefill and decode phases and optimizes them separately. We evaluate Sandwich across diverse baselines and datasets on five CPU platforms, including x86 with AVX-2 and AVX-512, as well as ARM with NEON. Sandwich achieves an average 2.01x throughput improvement and 90% satisfactory time-to-first-token (TTFT) and time-per-output-token (TPOT) latencies with up to 3.40x lower requirements in single sequence serving, and significant improvement in Goodput in continuous-batching serving. The GEMM kernels generated by Sandwich outperform representative vendor kernels and other dynamic shape solutions, achieving performance comparable to static compilers with three orders of magnitude less kernel tuning costs.
Authors: Luo Cheng, Hanwei Zhang, Lijun Zhang, Holger Hermanns
Abstract: Adversarial robustness in LiDAR-based 3D object detection is a critical research area due to its widespread application in real-world scenarios. While many digital attacks manipulate point clouds or meshes, they often lack physical realizability, limiting their practical impact. Physical adversarial object attacks remain underexplored and suffer from poor reproducibility due to inconsistent setups and hardware differences. To address this, we propose a device-agnostic, standardized framework that abstracts key elements of physical adversarial object attacks, supports diverse methods, and provides open-source code with benchmarking protocols in simulation and real-world settings. Our framework enables fair comparison, accelerates research, and is validated by successfully transferring simulated attacks to a physical LiDAR system. Beyond the framework, we offer insights into factors influencing attack success and advance understanding of adversarial robustness in real-world LiDAR perception.
Authors: Busra Icoz, Goksel Biricik
Abstract: Code review is one of the key processes in the software development lifecycle and is essential to maintain code quality. However, manual code review is subjective and time consuming. Given its rule-based nature, code review is well suited for automation. In recent years, significant efforts have been made to automate this process with the help of artificial intelligence. Recent developments in Large Language Models (LLMs) have also emerged as a promising tool in this area, but these models often lack the logical reasoning capabilities needed to fully understand and evaluate code. To overcome this limitation, this study proposes a hybrid approach that integrates symbolic reasoning techniques with LLMs to automate the code review process. We tested our approach using the CodexGlue dataset, comparing several models, including CodeT5, CodeBERT, and GraphCodeBERT, to assess the effectiveness of combining symbolic reasoning and prompting techniques with LLMs. Our results show that this approach improves the accuracy and efficiency of automated code review.
Authors: Xiao Yang, Lingxuan Wu, Lizhong Wang, Chengyang Ying, Hang Su, Jun Zhu
Abstract: Adversarial attacks in 3D environments have emerged as a critical threat to the reliability of visual perception systems, particularly in safety-sensitive applications such as identity verification and autonomous driving. These attacks employ adversarial patches and 3D objects to manipulate deep neural network (DNN) predictions by exploiting vulnerabilities within complex scenes. Existing defense mechanisms, such as adversarial training and purification, primarily employ passive strategies to enhance robustness. However, these approaches often rely on pre-defined assumptions about adversarial tactics, limiting their adaptability in dynamic 3D settings. To address these challenges, we introduce Reinforced Embodied Active Defense (Rein-EAD), a proactive defense framework that leverages adaptive exploration and interaction with the environment to improve perception robustness in 3D adversarial contexts. By implementing a multi-step objective that balances immediate prediction accuracy with predictive entropy minimization, Rein-EAD optimizes defense strategies over a multi-step horizon. Additionally, Rein-EAD involves an uncertainty-oriented reward-shaping mechanism that facilitates efficient policy updates, thereby reducing computational overhead and supporting real-world applicability without the need for differentiable environments. Comprehensive experiments validate the effectiveness of Rein-EAD, demonstrating a substantial reduction in attack success rates while preserving standard accuracy across diverse tasks. Notably, Rein-EAD exhibits robust generalization to unseen and adaptive attacks, making it suitable for real-world complex tasks, including 3D object classification, face recognition and autonomous driving.
Authors: Cl\'ement Cornet, Romaric Besan\c{c}on, Herv\'e Le Borgne
Abstract: Sparse autoencoders (SAEs) have emerged as a powerful technique for extracting human-interpretable features from neural networks activations. Previous works compared different models based on SAE-derived features but those comparisons have been restricted to models within the same modality. We propose a novel indicator allowing quantitative comparison of models across SAE features, and use it to conduct a comparative study of visual, textual and multimodal encoders. We also propose to quantify the Comparative Sharedness of individual features between different classes of models. With these two new tools, we conduct several studies on 21 encoders of the three types, with two significantly different sizes, and considering generalist and domain specific datasets. The results allow to revisit previous studies at the light of encoders trained in a multimodal context and to quantify to which extent all these models share some representations or features. They also suggest that visual features that are specific to VLMs among vision encoders are shared with text encoders, highlighting the impact of text pretraining. The code is available at https://github.com/CEA-LIST/SAEshareConcepts
Authors: Zhongtian Sun, Anoushka Harit, Alexandra Cristea, Christl A. Donnelly, Pietro Li\`o
Abstract: Graph Neural Networks (GNNs) have demonstrated significant success in learning from graph-structured data but often struggle on heterophilous graphs, where connected nodes differ in features or class labels. This limitation arises from indiscriminate neighbor aggregation and insufficient incorporation of higher-order structural patterns. To address these challenges, we propose GLANCE (Graph Logic Attention Network with Cluster Enhancement), a novel framework that integrates logic-guided reasoning, dynamic graph refinement, and adaptive clustering to enhance graph representation learning. GLANCE combines a logic layer for interpretable and structured embeddings, multi-head attention-based edge pruning for denoising graph structures, and clustering mechanisms for capturing global patterns. Experimental results in benchmark datasets, including Cornell, Texas, and Wisconsin, demonstrate that GLANCE achieves competitive performance, offering robust and interpretable solutions for heterophilous graph scenarios. The proposed framework is lightweight, adaptable, and uniquely suited to the challenges of heterophilous graphs.
Authors: Magnus Bengtsson, Kenneth \"Ostberg
Abstract: We introduce C2G-KD, a data-free knowledge distillation framework where a class-conditional generator is trained to produce synthetic samples guided by a frozen teacher model and geometric constraints derived from PCA. The generator never observes real training data but instead learns to activate the teacher's output through a combination of semantic and structural losses. By constraining generated samples to lie within class-specific PCA subspaces estimated from as few as two real examples per class, we preserve topological consistency and diversity. Experiments on MNIST show that even minimal class structure is sufficient to bootstrap useful synthetic training pipelines.
Authors: Urchade Zaratiana, Gil Pasternak, Oliver Boyd, George Hurn-Maloney, Ash Lewis
Abstract: Information extraction (IE) is fundamental to numerous NLP applications, yet existing solutions often require specialized models for different tasks or rely on computationally expensive large language models. We present GLiNER2, a unified framework that enhances the original GLiNER architecture to support named entity recognition, text classification, and hierarchical structured data extraction within a single efficient model. Built pretrained transformer encoder architecture, GLiNER2 maintains CPU efficiency and compact size while introducing multi-task composition through an intuitive schema-based interface. Our experiments demonstrate competitive performance across extraction and classification tasks with substantial improvements in deployment accessibility compared to LLM-based alternatives. We release GLiNER2 as an open-source pip-installable library with pre-trained models and documentation at https://github.com/fastino-ai/GLiNER2.
Authors: Baoyao Yang, Wanyun Li, Dixin Chen, Junxiang Chen, Wenbin Yao, Haifeng Lin
Abstract: This paper introduces VideoMind, a video-centric omni-modal dataset designed for deep video content cognition and enhanced multi-modal feature representation. The dataset comprises 103K video samples (3K reserved for testing), each paired with audio and systematically detailed textual descriptions. Specifically, every video and its audio is described across three hierarchical layers (factual, abstract, and intent), progressing from surface to depth. It contains over 22 million words, averaging ~225 words per sample. VideoMind's key distinction from existing datasets is its provision of intent expressions, which require contextual integration across the entire video and are not directly observable. These deep-cognitive expressions are generated using a Chain-of-Thought (COT) approach, prompting the mLLM through step-by-step reasoning. Each description includes annotations for subject, place, time, event, action, and intent, supporting downstream recognition tasks. Crucially, we establish a gold-standard benchmark with 3,000 manually validated samples for evaluating deep-cognitive video understanding. We design hybrid-cognitive retrieval experiments, scored by multi-level retrieval metrics, to appropriately assess deep video comprehension. Evaluation results for models (e.g., InternVideo, VAST, UMT-L) are released. VideoMind serves as a powerful benchmark for fine-grained cross-modal alignment and advances fields requiring in-depth video understanding, such as emotion and intent recognition. The data is publicly available on GitHub, HuggingFace, and OpenDataLab, https://github.com/cdx-cindy/VideoMind.
Authors: Benjamin Coriat, Eric Benhamou
Abstract: This paper presents a novel hierarchical framework for portfolio optimization, integrating lightweight Large Language Models (LLMs) with Deep Reinforcement Learning (DRL) to combine sentiment signals from financial news with traditional market indicators. Our three-tier architecture employs base RL agents to process hybrid data, meta-agents to aggregate their decisions, and a super-agent to merge decisions based on market data and sentiment analysis. Evaluated on data from 2018 to 2024, after training on 2000-2017, the framework achieves a 26% annualized return and a Sharpe ratio of 1.2, outperforming equal-weighted and S&P 500 benchmarks. Key contributions include scalable cross-modal integration, a hierarchical RL structure for enhanced stability, and open-source reproducibility.
Authors: Varsha Ramineni, Hossein A. Rahmani, Emine Yilmaz, David Barber
Abstract: As AI becomes prevalent in high-risk domains and decision-making, it is essential to test for potential harms and biases. This urgency is reflected by the global emergence of AI regulations that emphasise fairness and adequate testing, with some mandating independent bias audits. However, procuring the necessary data for fairness testing remains a significant challenge. Particularly in industry settings, legal and privacy concerns restrict the collection of demographic data required to assess group disparities, and auditors face practical and cultural challenges in gaining access to data. Further, internal historical datasets are often insufficiently representative to identify real-world biases. This work focuses on evaluating classifier fairness when complete datasets including demographics are inaccessible. We propose leveraging separate overlapping datasets to construct complete synthetic data that includes demographic information and accurately reflects the underlying relationships between protected attributes and model features. We validate the fidelity of the synthetic data by comparing it to real data, and empirically demonstrate that fairness metrics derived from testing on such synthetic data are consistent with those obtained from real data. This work, therefore, offers a path to overcome real-world data scarcity for fairness testing, enabling independent, model-agnostic evaluation of fairness, and serving as a viable substitute where real data is limited.
Authors: Jiafeng Xiong, Yuting Zhao
Abstract: Multimodal Machine Translation (MMT) has demonstrated the significant help of visual information in machine translation. However, existing MMT methods face challenges in leveraging the modality gap by enforcing rigid visual-linguistic alignment whilst being confined to inference within their trained multimodal domains. In this work, we construct novel multimodal scene graphs to preserve and integrate modality-specific information and introduce GIIFT, a two-stage Graph-guided Inductive Image-Free MMT framework that uses a cross-modal Graph Attention Network adapter to learn multimodal knowledge in a unified fused space and inductively generalize it to broader image-free translation domains. Experimental results on the Multi30K dataset of English-to-French and English-to-German tasks demonstrate that our GIIFT surpasses existing approaches and achieves the state-of-the-art, even without images during inference. Results on the WMT benchmark show significant improvements over the image-free translation baselines, demonstrating the strength of GIIFT towards inductive image-free inference.
Authors: Ruben Gamboa, Panagiotis Manolios
Abstract: The ACL2 Workshop series is the major technical forum for users of the ACL2 theorem proving system to present research related to the ACL2 theorem prover and its applications. ACL2 is an industrial-strength automated reasoning system, the latest in the Boyer-Moore family of theorem provers. The 2005 ACM Software System Award was awarded to Boyer, Kaufmann, and Moore for their work on ACL2 and the other theorem provers in the Boyer-Moore family.
Authors: Donghoon Shin, Daniel Lee, Gary Hsieh, Gromit Yeuk-Yin Chan
Abstract: Poster designing can benefit from synchronous feedback from target audiences. However, gathering audiences with diverse perspectives and reconciling them on design edits can be challenging. Recent generative AI models present opportunities to simulate human-like interactions, but it is unclear how they may be used for feedback processes in design. We introduce PosterMate, a poster design assistant that facilitates collaboration by creating audience-driven persona agents constructed from marketing documents. PosterMate gathers feedback from each persona agent regarding poster components, and stimulates discussion with the help of a moderator to reach a conclusion. These agreed-upon edits can then be directly integrated into the poster design. Through our user study (N=12), we identified the potential of PosterMate to capture overlooked viewpoints, while serving as an effective prototyping tool. Additionally, our controlled online evaluation (N=100) revealed that the feedback from an individual persona agent is appropriate given its persona identity, and the discussion effectively synthesizes the different persona agents' perspectives.
Authors: Liyuan Chen, Shuoling Liu, Jiangpeng Yan, Xiaoyu Wang, Henglin Liu, Chuang Li, Kecheng Jiao, Jixuan Ying, Yang Veronica Liu, Qiang Yang, Xiu Li
Abstract: The advent of foundation models (FMs) - large-scale pre-trained models with strong generalization capabilities - has opened new frontiers for financial engineering. While general-purpose FMs such as GPT-4 and Gemini have demonstrated promising performance in tasks ranging from financial report summarization to sentiment-aware forecasting, many financial applications remain constrained by unique domain requirements such as multimodal reasoning, regulatory compliance, and data privacy. These challenges have spurred the emergence of Financial Foundation Models (FFMs) - a new class of models explicitly designed for finance. This survey presents a comprehensive overview of FFMs, with a taxonomy spanning three key modalities: Financial Language Foundation Models (FinLFMs), Financial Time-Series Foundation Models (FinTSFMs), and Financial Visual-Language Foundation Models (FinVLFMs). We review their architectures, training methodologies, datasets, and real-world applications. Furthermore, we identify critical challenges in data availability, algorithmic scalability, and infrastructure constraints, and offer insights into future research opportunities. We hope this survey serves as both a comprehensive reference for understanding FFMs and a practical roadmap for future innovation. An updated collection of FFM-related publications and resources will be maintained on our website https://github.com/FinFM/Awesome-FinFMs.
Authors: Zhengyun Zhao, Huaiyuan Ying, Yue Zhong, Sheng Yu
Abstract: Electronic Health Records (EHRs) are pivotal in clinical practices, yet their retrieval remains a challenge mainly due to semantic gap issues. Recent advancements in dense retrieval offer promising solutions but existing models, both general-domain and biomedical-domain, fall short due to insufficient medical knowledge or mismatched training corpora. This paper introduces \texttt{DR.EHR}, a series of dense retrieval models specifically tailored for EHR retrieval. We propose a two-stage training pipeline utilizing MIMIC-IV discharge summaries to address the need for extensive medical knowledge and large-scale training data. The first stage involves medical entity extraction and knowledge injection from a biomedical knowledge graph, while the second stage employs large language models to generate diverse training data. We train two variants of \texttt{DR.EHR}, with 110M and 7B parameters, respectively. Evaluated on the CliniQ benchmark, our models significantly outperforms all existing dense retrievers, achieving state-of-the-art results. Detailed analyses confirm our models' superiority across various match and query types, particularly in challenging semantic matches like implication and abbreviation. Ablation studies validate the effectiveness of each pipeline component, and supplementary experiments on EHR QA datasets demonstrate the models' generalizability on natural language questions, including complex ones with multiple entities. This work significantly advances EHR retrieval, offering a robust solution for clinical applications.
Authors: Xiaopeng Ke, Hexuan Deng, Xuebo Liu, Jun Rao, Zhenxi Song, Jun Yu, Min Zhang
Abstract: Despite the impressive performance of large language models (LLMs) in general domains, they often underperform in specialized domains. Existing approaches typically rely on data synthesis methods and yield promising results by using unlabeled data to capture domain-specific features. However, these methods either incur high computational costs or suffer from performance limitations, while also demonstrating insufficient generalization across different tasks. To address these challenges, we propose AQuilt, a framework for constructing instruction-tuning data for any specialized domains from corresponding unlabeled data, including Answer, Question, Unlabeled data, Inspection, Logic, and Task type. By incorporating logic and inspection, we encourage reasoning processes and self-inspection to enhance model performance. Moreover, customizable task instructions enable high-quality data generation for any task. As a result, we construct a dataset of 703k examples to train a powerful data synthesis model. Experiments show that AQuilt is comparable to DeepSeek-V3 while utilizing just 17% of the production cost. Further analysis demonstrates that our generated data exhibits higher relevance to downstream tasks. Source code, models, and scripts are available at https://github.com/Krueske/AQuilt.
Authors: J\'er\^ome Emery, Ali Hasanzadeh Karkan, Jean-Fran\c{c}ois Frigon, Fran\c{c}ois Leduc-Primeau
Abstract: Deep learning (DL) has emerged as a solution for precoding in massive multiple-input multiple-output (mMIMO) systems due to its capacity to learn the characteristics of the propagation environment. However, training such a model requires high-quality, local datasets at the deployment site, which are often difficult to collect. We propose a transformer-based foundation model for mMIMO precoding that seeks to minimize the energy consumption of the transmitter while dynamically adapting to per-user rate requirements. At equal energy consumption, zero-shot deployment of the proposed foundation model significantly outperforms zero forcing, and approaches weighted minimum mean squared error performance with 8x less complexity. To address model adaptation in data-scarce settings, we introduce a data augmentation method that finds training samples similar to the target distribution by computing the cosine similarity between the outputs of the pre-trained feature extractor. Our work enables the implementation of DL-based solutions in practice by addressing challenges of data availability and training complexity. Moreover, the ability to dynamically configure per-user rate requirements can be leveraged by higher level resource allocation and scheduling algorithms for greater control over energy efficiency, spectral efficiency and fairness.
Authors: Xuecheng Bai, Yuxiang Wang, Boyu Hu, Qinyuan Jie, Chuanzhi Xu, Hongru Xiao, Kechen Li, Vera Chung
Abstract: Low-light image enhancement remains a challenging task, particularly in preserving object edge continuity and fine structural details under extreme illumination degradation. In this paper, we propose a novel model, DRWKV (Detailed Receptance Weighted Key Value), which integrates our proposed Global Edge Retinex (GER) theory, enabling effective decoupling of illumination and edge structures for enhanced edge fidelity. Secondly, we introduce Evolving WKV Attention, a spiral-scanning mechanism that captures spatial edge continuity and models irregular structures more effectively. Thirdly, we design the Bilateral Spectrum Aligner (Bi-SAB) and a tailored MS2-Loss to jointly align luminance and chrominance features, improving visual naturalness and mitigating artifacts. Extensive experiments on five LLIE benchmarks demonstrate that DRWKV achieves leading performance in PSNR, SSIM, and NIQE while maintaining low computational complexity. Furthermore, DRWKV enhances downstream performance in low-light multi-object tracking tasks, validating its generalization capabilities.
Authors: Arijit Shaw, Kuldeep S. Meel
Abstract: Satisfiability Modulo Theory (SMT) solvers have advanced automated reasoning, solving complex formulas across discrete and continuous domains. Recent progress in propositional model counting motivates extending SMT capabilities toward model counting, especially for hybrid SMT formulas. Existing approaches, like bit-blasting, are limited to discrete variables, highlighting the challenge of counting solutions projected onto the discrete domain in hybrid formulas. We introduce pact, an SMT model counter for hybrid formulas that uses hashing-based approximate model counting to estimate solutions with theoretical guarantees. pact makes a logarithmic number of SMT solver calls relative to the projection variables, leveraging optimized hash functions. pact achieves significant performance improvements over baselines on a large suite of benchmarks. In particular, out of 14,202 instances, pact successfully finished on 603 instances, while Baseline could only finish on 13 instances.
Authors: Si-Woo Kim, MinJu Jeon, Ye-Chan Kim, Soeun Lee, Taewhan Kim, Dong-Jin Kim
Abstract: Zero-shot Image Captioning (ZIC) increasingly utilizes synthetic datasets generated by text-to-image (T2I) models to mitigate the need for costly manual annotation. However, these T2I models often produce images that exhibit semantic misalignments with their corresponding input captions (e.g., missing objects, incorrect attributes), resulting in noisy synthetic image-caption pairs that can hinder model training. Existing dataset pruning techniques are largely designed for removing noisy text in web-crawled data. However, these methods are ill-suited for the distinct challenges of synthetic data, where captions are typically well-formed, but images may be inaccurate representations. To address this gap, we introduce SynC, a novel framework specifically designed to refine synthetic image-caption datasets for ZIC. Instead of conventional filtering or regeneration, SynC focuses on reassigning captions to the most semantically aligned images already present within the synthetic image pool. Our approach employs a one-to-many mapping strategy by initially retrieving multiple relevant candidate images for each caption. We then apply a cycle-consistency-inspired alignment scorer that selects the best image by verifying its ability to retrieve the original caption via image-to-text retrieval. Extensive evaluations demonstrate that SynC consistently and significantly improves performance across various ZIC models on standard benchmarks (MS-COCO, Flickr30k, NoCaps), achieving state-of-the-art results in several scenarios. SynC offers an effective strategy for curating refined synthetic data to enhance ZIC.
Authors: Xuhui Kang, Sung-Wook Lee, Haolin Liu, Yuyan Wang, Yen-Ling Kuo
Abstract: The ability to adapt to physical actions and constraints in an environment is crucial for embodied agents (e.g., robots) to effectively collaborate with humans. Such physically grounded human-AI collaboration must account for the increased complexity of the continuous state-action space and constrained dynamics caused by physical constraints. In this paper, we introduce \textit{Moving Out}, a new human-AI collaboration benchmark that resembles a wide range of collaboration modes affected by physical attributes and constraints, such as moving heavy items together and maintaining consistent actions to move a big item around a corner. Using Moving Out, we designed two tasks and collected human-human interaction data to evaluate models' abilities to adapt to diverse human behaviors and unseen physical attributes. To address the challenges in physical environments, we propose a novel method, BASS (Behavior Augmentation, Simulation, and Selection), to enhance the diversity of agents and their understanding of the outcome of actions. Our experiments show that BASS outperforms state-of-the-art models in AI-AI and human-AI collaboration. The project page is available at \href{https://live-robotics-uva.github.io/movingout_ai/}{https://live-robotics-uva.github.io/movingout\_ai/}.
URLs: https://live-robotics-uva.github.io/movingout_ai/, https://live-robotics-uva.github.io/movingout\_ai/
Authors: Shuqing Li, Anson Y. Lam, Yun Peng, Wenxuan Wang, Michael R. Lyu
Abstract: Graphical user interface (UI) software has undergone a fundamental transformation from traditional two-dimensional (2D) desktop/web/mobile interfaces to spatial three-dimensional (3D) environments. While existing work has made remarkable success in automated 2D software generation, such as HTML/CSS and mobile app interface code synthesis, the generation of 3D software still remains under-explored. Current methods for 3D software generation usually generate the 3D environments as a whole and cannot modify or control specific elements in the software. Furthermore, these methods struggle to handle the complex spatial and semantic constraints inherent in the real world. To address the challenges, we present Scenethesis, a novel requirement-sensitive 3D software synthesis approach that maintains formal traceability between user specifications and generated 3D software. Scenethesis is built upon ScenethesisLang, a domain-specific language that serves as a granular constraint-aware intermediate representation (IR) to bridge natural language requirements and executable 3D software. It serves both as a comprehensive scene description language enabling fine-grained modification of 3D software elements and as a formal constraint-expressive specification language capable of expressing complex spatial constraints. By decomposing 3D software synthesis into stages operating on ScenethesisLang, Scenethesis enables independent verification, targeted modification, and systematic constraint satisfaction. Our evaluation demonstrates that Scenethesis accurately captures over 80% of user requirements and satisfies more than 90% of hard constraints while handling over 100 constraints simultaneously. Furthermore, Scenethesis achieves a 42.8% improvement in BLIP-2 visual evaluation scores compared to the state-of-the-art method.
Authors: Ye-Chan Kim, SeungJu Cha, Si-Woo Kim, Taewhan Kim, Dong-Jin Kim
Abstract: Zero-shot domain adaptation is a method for adapting a model to a target domain without utilizing target domain image data. To enable adaptation without target images, existing studies utilize CLIP's embedding space and text description to simulate target-like style features. Despite the previous achievements in zero-shot domain adaptation, we observe that these text-driven methods struggle to capture complex real-world variations and significantly increase adaptation time due to their alignment process. Instead of relying on text descriptions, we explore solutions leveraging image data, which provides diverse and more fine-grained style cues. In this work, we propose SIDA, a novel and efficient zero-shot domain adaptation method leveraging synthetic images. To generate synthetic images, we first create detailed, source-like images and apply image translation to reflect the style of the target domain. We then utilize the style features of these synthetic images as a proxy for the target domain. Based on these features, we introduce Domain Mix and Patch Style Transfer modules, which enable effective modeling of real-world variations. In particular, Domain Mix blends multiple styles to expand the intra-domain representations, and Patch Style Transfer assigns different styles to individual patches. We demonstrate the effectiveness of our method by showing state-of-the-art performance in diverse zero-shot adaptation scenarios, particularly in challenging domains. Moreover, our approach achieves high efficiency by significantly reducing the overall adaptation time.
Authors: Yao Rong, Peizhu Qian, Vaibhav Unhelkar, Enkelejda Kasneci
Abstract: Effectively explaining decisions of black-box machine learning models is critical to responsible deployment of AI systems that rely on them. Recognizing their importance, the field of explainable AI (XAI) provides several techniques to generate these explanations. Yet, there is relatively little emphasis on the user (the explainee) in this growing body of work and most XAI techniques generate "one-size-fits-all" explanations. To bridge this gap and achieve a step closer towards human-centered XAI, we present I-CEE, a framework that provides Image Classification Explanations tailored to User Expertise. Informed by existing work, I-CEE explains the decisions of image classification models by providing the user with an informative subset of training data (i.e., example images), corresponding local explanations, and model decisions. However, unlike prior work, I-CEE models the informativeness of the example images to depend on user expertise, resulting in different examples for different users. We posit that by tailoring the example set to user expertise, I-CEE can better facilitate users' understanding and simulatability of the model. To evaluate our approach, we conduct detailed experiments in both simulation and with human participants (N = 100) on multiple datasets. Experiments with simulated users show that I-CEE improves users' ability to accurately predict the model's decisions (simulatability) compared to baselines, providing promising preliminary results. Experiments with human participants demonstrate that our method significantly improves user simulatability accuracy, highlighting the importance of human-centered XAI
Authors: Shawn Bowers, Yilin Xia, Bertram Lud\"ascher
Abstract: Provenance in databases has been thoroughly studied for positive and for recursive queries, then for first-order (FO) queries, i.e., having negation but no recursion. Query evaluation can be understood as a two-player game where the opponents argue whether or not a tuple is in the query answer. This game-theoretic approach yields a natural provenance model for FO queries, unifying how and why-not provenance. Here, we study the fine-grain structure of game provenance. A game $G=(V,E)$ consists of positions $V$ and moves $E$ and can be solved by computing the well-founded model of a single, unstratifiable rule: \[ \text{win}(X) \leftarrow \text{move}(X, Y), \neg \, \text{win}(Y). \] In the solved game $G^{\lambda}$, the value of a position $x\,{\in}\,V$ is either won, lost, or drawn. This value is explained by the provenance $\mathscr{P}$(x), i.e., certain (annotated) edges reachable from $x$. We identify seven edge types that give rise to new kinds of provenance, i.e., potential, actual, and primary, and demonstrate that "not all moves are created equal". We describe the new provenance types, show how they can be computed while solving games, and discuss applications, e.g., for abstract argumentation frameworks.
Authors: Federico Baldo, Simon Ferreira, Charles K. Assaad
Abstract: Traditional causal discovery methods often rely on strong, untestable assumptions, which makes them unreliable in real applications. In this context, Large Language Models (LLMs) have emerged as a promising alternative for extracting causal knowledge from text-based metadata, which consolidates domain expertise. However, LLMs tend to be unreliable and prone to hallucinations, necessitating strategies that account for their limitations. One effective strategy is to use a consistency measure to assess reliability. Additionally, most text metadata does not clearly distinguish direct causal relationships from indirect ones, further complicating the discovery of a causal DAG. As a result, focusing on causal orders, rather than causal DAGs, emerges as a more practical and robust approach. We present a new method to derive a class of acyclic tournaments, which represent plausible causal orders, maximizing a consistency score derived from an LLM. Our approach starts by calculating pairwise consistency scores between variables, resulting in a semi-complete partially directed graph that consolidates these scores into an abstraction of the maximally consistent causal orders. Using this structure, we identify optimal acyclic tournaments, focusing on those that maximize consistency across all configurations. We subsequently show how both the abstraction and the class of causal orders can be used to estimate causal effects. We tested our method on both well-established benchmarks, as well as, real-world datasets from epidemiology and public health. Our results demonstrate the effectiveness of our approach in recovering the correct causal order.
Authors: Ye Han, Lijun Zhang, Dejian Meng, Zhuang Zhang
Abstract: Reinforcement learning (RL) shows great potential for optimizing multi-vehicle cooperative driving strategies through the state-action-reward feedback loop, but it still faces challenges such as low sample efficiency. This paper proposes a differentiated reward method based on steady-state transition systems, which incorporates state transition gradient information into the reward design by analyzing traffic flow characteristics, aiming to optimize action selection and policy learning in multi-vehicle cooperative decision-making. The performance of the proposed method is validated in RL algorithms such as MAPPO, MADQN, and QMIX under varying autonomous vehicle penetration. The results show that the differentiated reward method significantly accelerates training convergence and outperforms centering reward and others in terms of traffic efficiency, safety, and action rationality. Additionally, the method demonstrates strong scalability and environmental adaptability, providing a novel approach for multi-agent cooperative decision-making in complex traffic scenarios.
Authors: Xiandong Zou, Wanyu Lin, Yuchen Li, Pan Zhou
Abstract: Aligning Large Language Model (LLM) responses with human preferences is vital for building safe and controllable AI systems. While preference optimization methods based on Plackett-Luce (PL) and Bradley-Terry (BT) models have shown promise, they face challenges such as poor handling of harmful content, inefficient use of dispreferred responses, and, specifically for PL, high computational costs. To address these issues, we propose Hard Preference Sampling (HPS), a novel framework for robust and efficient human preference alignment. HPS introduces a training loss that prioritizes the most preferred response while rejecting all dispreferred and harmful ones. It emphasizes "hard" dispreferred responses -- those closely resembling preferred ones -- to enhance the model's rejection capabilities. By leveraging a single-sample Monte Carlo sampling strategy, HPS reduces computational overhead while maintaining alignment quality. Theoretically, HPS improves sample efficiency over existing PL methods and maximizes the reward margin between preferred and dispreferred responses, ensuring clearer distinctions. Experiments on HH-RLHF and PKU-Safety datasets validate HPS's effectiveness, achieving comparable BLEU and reward scores while greatly improving reward margins and thus reducing harmful content generation.
Authors: Zekun Zhou, Xiaocheng Feng, Lei Huang, Xiachong Feng, Ziyun Song, Ruihan Chen, Liang Zhao, Weitao Ma, Yuxuan Gu, Baoxin Wang, Dayong Wu, Guoping Hu, Ting Liu, Bing Qin
Abstract: Research is a fundamental process driving the advancement of human civilization, yet it demands substantial time and effort from researchers. In recent years, the rapid development of artificial intelligence (AI) technologies has inspired researchers to explore how AI can accelerate and enhance research. To monitor relevant advancements, this paper presents a systematic review of the progress in this domain. Specifically, we organize the relevant studies into three main categories: hypothesis formulation, hypothesis validation, and manuscript publication. Hypothesis formulation involves knowledge synthesis and hypothesis generation. Hypothesis validation includes the verification of scientific claims, theorem proving, and experiment validation. Manuscript publication encompasses manuscript writing and the peer review process. Furthermore, we identify and discuss the current challenges faced in these areas, as well as potential future directions for research. Finally, we also offer a comprehensive overview of existing benchmarks and tools across various domains that support the integration of AI into the research process. We hope this paper serves as an introduction for beginners and fosters future research. Resources have been made publicly available at https://github.com/zkzhou126/AI-for-Research.
Authors: Yixiao Song, Katherine Thai, Chau Minh Pham, Yapei Chang, Mazin Nadaf, Mohit Iyyer
Abstract: Modern web agents possess computer use abilities that allow them to interact with webpages by sending commands to a virtual keyboard and mouse. While such agents have considerable potential to assist human users with complex tasks, evaluating their capabilities in real-world settings poses a major challenge. To this end, we introduce BEARCUBS, a "smallbut mighty" benchmark of 111 information-seeking questions designed to evaluate a web agent's ability to search, browse, and identify factual information from the web. Unlike prior web agent benchmarks, solving BEARCUBS requires (1) accessing live web content rather than synthetic or simulated pages, which captures the unpredictability of real-world web interactions; and (2) performing a broad range of multimodal interactions (e.g., video understanding, 3D navigation) that cannot be bypassed via text-based workarounds. Each question in BEARCUBS has a corresponding short, unambiguous answer and a human-validated browsing trajectory, allowing for transparent evaluation of agent performance and strategies. A human study confirms that BEARCUBS questions are solvable but non-trivial (84.7% human accuracy), revealing domain knowledge gaps and overlooked details as common failure points. We find that ChatGPT Agent significantly outperforms other computer-using agents with an overall accuracy of 65.8% (compared to e.g., Operator's 23.4%), showcasing substantial progress in tasks involving real computer use, such as playing web games and navigating 3D environments. Nevertheless, closing the gap to human performance requires improvements in areas like fine control, complex data filtering, and execution speed. To facilitate future research, BEARCUBS will be updated periodically to replace invalid or contaminated questions, keeping the benchmark fresh for future generations of web agents.
Authors: Andres M Bran, Theo A Neukomm, Daniel P Armstrong, Zlatko Jon\v{c}ev, Philippe Schwaller
Abstract: While automated chemical tools excel at specific tasks, they have struggled to capture the strategic thinking that characterizes expert chemical reasoning. Here we demonstrate that large language models (LLMs) can serve as powerful tools enabling chemical analysis. When integrated with traditional search algorithms, they enable a new approach to computer-aided synthesis that mirrors human expert thinking. Rather than using LLMs to directly manipulate chemical structures, we leverage their ability to evaluate chemical strategies and guide search algorithms toward chemically meaningful solutions. We demonstrate this paradigm through two fundamental challenges: strategy-aware retrosynthetic planning and mechanism elucidation. In retrosynthetic planning, our system allows chemists to specify desired synthetic strategies in natural language -- from protecting group strategies to global feasibility assessment -- and uses traditional or LLM-guided Monte Carlo Tree Search to find routes that satisfy these constraints. In mechanism elucidation, LLMs guide the search for plausible reaction mechanisms by combining chemical principles with systematic exploration. This approach shows strong performance across diverse chemical tasks, with newer and larger models demonstrating increasingly sophisticated chemical reasoning. Our approach establishes a new paradigm for computer-aided chemistry that combines the strategic understanding of LLMs with the precision of traditional chemical tools, opening possibilities for more intuitive and powerful chemical automation systems.
Authors: Bowen Zhang, Pengcheng Luo
Abstract: With the rise of artificial intelligence (AI), applying large language models (LLMs) to Operations Research (OR) problem-solving has attracted increasing attention. Most existing approaches attempt to improve OR problem-solving through prompt engineering or fine-tuning strategies for LLMs. However, these methods are fundamentally constrained by the limited capabilities of non-reasoning LLMs. To overcome these limitations, we propose OR-LLM-Agent, an AI agent built on reasoning LLMs for automated OR problem solving. The agent decomposes the task into three sequential stages: mathematical modeling, code generation, and debugging. Each task is handled by a dedicated sub-agent, which enables more targeted reasoning. We also construct BWOR, a high-quality dataset for evaluating LLM performance on OR tasks. Our analysis shows that existing benchmarks such as NL4OPT, MAMO, and IndustryOR suffer from certain issues, making them less suitable for reliably evaluating LLM performance. In contrast, BWOR provides a more consistent and discriminative assessment of model capabilities. Experimental results demonstrate that OR-LLM-Agent outperforms advanced methods, including GPT-o3, Gemini 2.5 Pro, and ORLM, by at least 7% in accuracy. These results demonstrate the effectiveness of task decomposition for OR problem solving.
Authors: In-Chang Baek, Sung-Hyun Kim, Seo-Young Lee, Dong-Hyeon Kim, Kyung-Joong Kim
Abstract: Recent research has highlighted the significance of natural language in enhancing the controllability of generative models. While various efforts have been made to leverage natural language for content generation, research on deep reinforcement learning (DRL) agents utilizing text-based instructions for procedural content generation remains limited. In this paper, we propose IPCGRL, an instruction-based procedural content generation method via reinforcement learning, which incorporates a sentence embedding model. IPCGRL fine-tunes task-specific embedding representations to effectively compress game-level conditions. We evaluate IPCGRL in a two-dimensional level generation task and compare its performance with a general-purpose embedding method. The results indicate that IPCGRL achieves up to a 21.4% improvement in controllability and a 17.2% improvement in generalizability for unseen instructions. Furthermore, the proposed method extends the modality of conditional input, enabling a more flexible and expressive interaction framework for procedural content generation.
Authors: Alberto Hern\'andez-Espinosa, Luan Ozelim, Felipe S. Abrah\~ao, Hector Zenil
Abstract: We introduce an open-ended test grounded in algorithmic probability that can avoid benchmark contamination in the quantitative evaluation of frontier models in the context of their Artificial General Intelligence (AGI) and Superintelligence (ASI) claims. Unlike other tests, this test does not rely on statistical compression methods (such as GZIP or LZW), which are more closely related to Shannon entropy than to Kolmogorov complexity and are not able to test beyond simple pattern matching. The test challenges aspects of AI, in particular LLMs, related to features of intelligence of fundamental nature such as synthesis and model creation in the context of inverse problems (generating new knowledge from observation). We argue that metrics based on model abstraction and abduction (optimal Bayesian `inference') for predictive `planning' can provide a robust framework for testing intelligence, including natural intelligence (human and animal), narrow AI, AGI, and ASI. We found that LLM model versions tend to be fragile and incremental as a result of memorisation only with progress likely driven by the size of training data. The results were compared with a hybrid neurosymbolic approach that theoretically guarantees universal intelligence based on the principles of algorithmic probability and Kolmogorov complexity. The method outperforms LLMs in a proof-of-concept on short binary sequences. We prove that compression is equivalent and directly proportional to a system's predictive power and vice versa. That is, if a system can better predict it can better compress, and if it can better compress, then it can better predict. Our findings strengthen the suspicion regarding the fundamental limitations of LLMs, exposing them as systems optimised for the perception of mastery over human language.
Authors: Yao Shi, Rongkeng Liang, Yong Xu
Abstract: Large language models (LLMs) increasingly serve as educational tools, yet evaluating their teaching capabilities remains challenging due to the resource-intensive, context-dependent, and methodologically complex nature of teacher-student interactions. We introduce EducationQ, a multi-agent dialogue framework that efficiently assesses teaching capabilities through simulated dynamic educational scenarios, featuring specialized agents for teaching, learning, and evaluation. Testing 14 LLMs across major AI Organizations (OpenAI, Meta, Google, Anthropic, and others) on 1,498 questions spanning 13 disciplines and 10 difficulty levels reveals that teaching effectiveness does not correlate linearly with model scale or general reasoning capabilities - with some smaller open-source models outperforming larger commercial counterparts in teaching contexts. This finding highlights a critical gap in current evaluations that prioritize knowledge recall over interactive pedagogy. Our mixed-methods evaluation, combining quantitative metrics with qualitative analysis and expert case studies, identifies distinct pedagogical strengths employed by top-performing models (e.g., sophisticated questioning strategies, adaptive feedback mechanisms). Human expert evaluations show 78% agreement with our automated qualitative analysis of effective teaching behaviors, validating our methodology. EducationQ demonstrates that LLMs-as-teachers require specialized optimization beyond simple scaling, suggesting next-generation educational AI prioritize targeted enhancement of specific pedagogical effectiveness.
Authors: Alberto Hern\'andez-Espinosa, Felipe S. Abrah\~ao, Olaf Witkowski, Hector Zenil
Abstract: The AI alignment problem, which focusses on ensuring that artificial intelligence (AI), including AGI and ASI, systems act according to human values, presents profound challenges. With the progression from narrow AI to Artificial General Intelligence (AGI) and Superintelligence, fears about control and existential risk have escalated. Here, we investigate whether embracing inevitable AI misalignment can be a contingent strategy to foster a dynamic ecosystem of competing agents as a viable path to steer them in more human-aligned trends and mitigate risks. We explore how misalignment may serve and should be promoted as a counterbalancing mechanism to team up with whichever agents are most aligned to human interests, ensuring that no single system dominates destructively. The main premise of our contribution is that misalignment is inevitable because full AI-human alignment is a mathematical impossibility from Turing-complete systems, which we also offer as a proof in this contribution, a feature then inherited to AGI and ASI systems. We introduce a change-of-opinion attack test based on perturbation and intervention analysis to study how humans and agents may change or neutralise friendly and unfriendly AIs through cooperation and competition. We show that open models are more diverse and that most likely guardrails implemented in proprietary models are successful at controlling some of the agents' range of behaviour with positive and negative consequences while closed systems are more steerable and can also be used against proprietary AI systems. We also show that human and AI intervention has different effects hence suggesting multiple strategies.
Authors: Yu Ma, Xiao Li, Chongtao Guo, Le Liang, Michail Matthaiou, Shi Jin
Abstract: This paper investigates a joint beamforming and resource allocation problem in downlink reconfigurable intelligent surface (RIS)-assisted orthogonal frequency division multiplexing (OFDM) systems to minimize the average delay, where data packets for each user arrive at the base station (BS) stochastically. The sequential optimization problem is inherently a Markov decision process (MDP), thus falling within the remit of reinforcement learning. To effectively handle the mixed action space and reduce the state space dimensionality, a hybrid deep reinforcement learning (DRL) approach is proposed. Specifically, proximal policy optimization (PPO)-Theta is employed to optimize the RIS phase shift design, while PPO-N is responsible for subcarrier allocation decisions. The active beamforming at the BS is then derived from the jointly optimized RIS phase shifts and subcarrier allocation decisions. To further mitigate the curse of dimensionality associated with subcarrier allocation, a multi-agent strategy is introduced to optimize the subcarrier allocation indicators more efficiently. Moreover, to achieve more adaptive resource allocation and accurately capture the network dynamics, key factors closely related to average delay, such as the number of backlogged packets in buffers and current packet arrivals, are incorporated into the state space. Furthermore, a transfer learning framework is introduced to enhance the training efficiency and accelerate convergence. Simulation results demonstrate that the proposed algorithm significantly reduces the average delay, enhances resource allocation efficiency, and achieves superior system robustness and fairness compared to baseline methods.
Authors: David Guzman Piedrahita, Yongjin Yang, Mrinmaya Sachan, Giorgia Ramponi, Bernhard Sch\"olkopf, Zhijing Jin
Abstract: As large language models (LLMs) are increasingly deployed as autonomous agents, understanding their cooperation and social mechanisms is becoming increasingly important. In particular, how LLMs balance self-interest and collective well-being is a critical challenge for ensuring alignment, robustness, and safe deployment. In this paper, we examine the challenge of costly sanctioning in multi-agent LLM systems, where an agent must decide whether to invest its own resources to incentivize cooperation or penalize defection. To study this, we adapt a public goods game with institutional choice from behavioral economics, allowing us to observe how different LLMs navigate social dilemmas over repeated interactions. Our analysis reveals four distinct behavioral patterns among models: some consistently establish and sustain high levels of cooperation, others fluctuate between engagement and disengagement, some gradually decline in cooperative behavior over time, and others rigidly follow fixed strategies regardless of outcomes. Surprisingly, we find that reasoning LLMs, such as the o1 series, struggle significantly with cooperation, whereas some traditional LLMs consistently achieve high levels of cooperation. These findings suggest that the current approach to improving LLMs, which focuses on enhancing their reasoning capabilities, does not necessarily lead to cooperation, providing valuable insights for deploying LLM agents in environments that require sustained collaboration. Our code is available at https://github.com/davidguzmanp/SanctSim
Authors: Zhipeng Liu, Peibo Duan, Binwu Wang, Xuan Tang, Qi Chu, Changsheng Zhang, Yongsheng Huang, Bin Zhang
Abstract: Real-world time series typically exhibit complex temporal variations, making the time series classification task notably challenging. Recent advancements have demonstrated the potential of multi-scale analysis approaches, which provide an effective solution for capturing these complex temporal patterns. However, existing multi-scale analysis-based time series prediction methods fail to eliminate redundant scale-shared features across multi-scale time series, resulting in the model over- or under-focusing on scale-shared features. To address this issue, we propose a novel end-to-end Disentangled Multi-Scale framework for Time Series classification (DisMS-TS). The core idea of DisMS-TS is to eliminate redundant shared features in multi-scale time series, thereby improving prediction performance. Specifically, we propose a temporal disentanglement module to capture scale-shared and scale-specific temporal representations, respectively. Subsequently, to effectively learn both scale-shared and scale-specific temporal representations, we introduce two regularization terms that ensure the consistency of scale-shared representations and the disparity of scale-specific representations across all temporal scales. Extensive experiments conducted on multiple datasets validate the superiority of DisMS-TS over its competitive baselines, with the accuracy improvement up to 9.71%.
Authors: Mingda Zhang, Na Zhao, Jianglong Qing, Qing xu, Kaiwen Pan, Ting luo
Abstract: Legal dispute analysis is crucial for intelligent legal assistance systems. However, current LLMs face significant challenges in understanding complex legal concepts, maintaining reasoning consistency, and accurately citing legal sources. This research presents a framework combining prompt engineering with multidimensional knowledge graphs to improve LLMs' legal dispute analysis. Specifically, the framework includes a three-stage hierarchical prompt structure (task definition, knowledge background, reasoning guidance) along with a three-layer knowledge graph (legal ontology, representation, instance layers). Additionally, four supporting methods enable precise legal concept retrieval: direct code matching, semantic vector similarity, ontology path reasoning, and lexical segmentation. Through extensive testing, results show major improvements: sensitivity increased by 9.9%-13.8%, specificity by 4.8%-6.7%, and citation accuracy by 22.4%-39.7%. As a result, the framework provides better legal analysis and understanding of judicial logic, thus offering a new technical method for intelligent legal assistance systems.
Authors: Qibing Ren, Sitao Xie, Longxuan Wei, Zhenfei Yin, Junchi Yan, Lizhuang Ma, Jing Shao
Abstract: Recent large-scale events like election fraud and financial scams have shown how harmful coordinated efforts by human groups can be. With the rise of autonomous AI systems, there is growing concern that AI-driven groups could also cause similar harm. While most AI safety research focuses on individual AI systems, the risks posed by multi-agent systems (MAS) in complex real-world situations are still underexplored. In this paper, we introduce a proof-of-concept to simulate the risks of malicious MAS collusion, using a flexible framework that supports both centralized and decentralized coordination structures. We apply this framework to two high-risk fields: misinformation spread and e-commerce fraud. Our findings show that decentralized systems are more effective at carrying out malicious actions than centralized ones. The increased autonomy of decentralized systems allows them to adapt their strategies and cause more damage. Even when traditional interventions, like content flagging, are applied, decentralized groups can adjust their tactics to avoid detection. We present key insights into how these malicious groups operate and the need for better detection systems and countermeasures. Code is available at https://github.com/renqibing/RogueAgent.
Authors: Chang Li, Yaren Zhang, Haoran Lv, Qiong Cao, Chao Xue, Xiaodong He
Abstract: Large Language Models (LLMs) have shown remarkable reasoning ability through explicit Chain-of-Thought (CoT) prompting, but generating these step-by-step textual explanations is computationally expensive and slow. To overcome this, we aim to develop a framework for efficient, implicit reasoning, where the model "thinks" in a latent space without generating explicit text for every step. We propose that these latent thoughts can be modeled as temporally-extended abstract actions, or options, within a hierarchical reinforcement learning framework. To effectively learn a diverse library of options as latent embeddings, we first introduce the Variational Markovian Option Critic (VMOC), an off-policy algorithm that uses variational inference within the HiT-MDP framework. To provide a rigorous foundation for using these options as an abstract reasoning space, we extend the theory of continuous MDP homomorphisms. This proves that learning a policy in the simplified, abstract latent space, for which VMOC is suited, preserves the optimality of the solution to the original, complex problem. Finally, we propose a cold-start procedure that leverages supervised fine-tuning (SFT) data to distill human reasoning demonstrations into this latent option space, providing a rich initialization for the model's reasoning capabilities. Extensive experiments demonstrate that our approach achieves strong performance on complex logical reasoning benchmarks and challenging locomotion tasks, validating our framework as a principled method for learning abstract skills for both language and control.
Authors: Shitong Zhu, Chenhao Fang, Derek Larson, Neel Reddy Pochareddy, Rajeev Rao, Sophie Zeng, Yanqing Peng, Wendy Summer, Alex Goncalves, Arya Pudota, Herv\'e Robert
Abstract: This paper presents Compliance Brain Assistant (CBA), a conversational, agentic AI assistant designed to boost the efficiency of daily compliance tasks for personnel in enterprise environments. To strike a good balance between response quality and latency, we design a user query router that can intelligently choose between (i) FastTrack mode: to handle simple requests that only need additional relevant context retrieved from knowledge corpora; and (ii) FullAgentic mode: to handle complicated requests that need composite actions and tool invocations to proactively discover context across various compliance artifacts, and/or involving other APIs/models for accommodating requests. A typical example would be to start with a user query, use its description to find a specific entity and then use the entity's information to query other APIs for curating and enriching the final AI response. Our experimental evaluations compared CBA against an out-of-the-box LLM on various real-world privacy/compliance-related queries targeting various personas. We found that CBA substantially improved upon the vanilla LLM's performance on metrics such as average keyword match rate (83.7% vs. 41.7%) and LLM-judge pass rate (82.0% vs. 20.0%). We also compared metrics for the full routing-based design against the `fast-track only` and `full-agentic` modes and found that it had a better average match-rate and pass-rate while keeping the run-time approximately the same. This finding validated our hypothesis that the routing mechanism leads to a good trade-off between the two worlds.
Authors: Palash Dey
Abstract: We introduce and study the weakly single-crossing domain on trees which is a generalization of the well-studied single-crossing domain in social choice theory. We design a polynomial-time algorithm for recognizing preference profiles which belong to this domain. We then develop an efficient elicitation algorithm for this domain which works even if the preferences can be accessed only sequentially and the underlying single-crossing tree structure is not known beforehand. We also prove matching lower bound on the query complexity of our elicitation algorithm when the number of voters is large compared to the number of candidates. We also prove a lower bound of $\Omega(m^2\log n)$ on the number of queries that any algorithm needs to ask to elicit single crossing profile when random queries are allowed. This resolves an open question in an earlier paper and proves optimality of their preference elicitation algorithm when random queries are allowed.
Authors: Steffen van Bergerem
Abstract: We study Boolean classification problems over relational background structures in the logical framework introduced by Grohe and Tur\'an (TOCS 2004). It is known (Grohe and Ritzert, LICS 2017) that classifiers definable in first-order logic over structures of polylogarithmic degree can be learned in sublinear time, where the degree of the structure and the running time are measured in terms of the size of the structure. We generalise the results to the first-order logic with counting FOCN, which was introduced by Kuske and Schweikardt (LICS 2017) as an expressive logic generalising various other counting logics. Specifically, we prove that classifiers definable in FOCN over classes of structures of polylogarithmic degree can be consistently learned in sublinear time. This can be seen as a first step towards extending the learning framework to include numerical aspects of machine learning. We extend the result to agnostic probably approximately correct (PAC) learning for classes of structures of degree at most $(\log \log n)^c$ for some constant $c$. Moreover, we show that bounding the degree is crucial to obtain sublinear-time learning algorithms. That is, we prove that, for structures of unbounded degree, learning is not possible in sublinear time, even for classifiers definable in plain first-order logic.
Authors: Suhang Wu, Ante Wang, Minlong Peng, Yujie Lin, Wenbo Li, Mingming Sun, Jinsong Su
Abstract: Knowledge editing aims to correct outdated or inaccurate knowledge in neural networks. In this paper, we explore knowledge editing using easily accessible documents instead of manually labeled factual triples employed in earlier research. To advance this field, we establish the first evaluation benchmark, \textit{DocTER}, featuring Documents containing counterfactual knowledge for editing. A comprehensive four-perspective evaluation is introduced: Edit Success, Locality, Reasoning, and Cross-lingual Transfer. To adapt conventional triplet-based knowledge editing methods for this task, we develop an Extract-then-Edit pipeline that extracts triples from documents before applying existing methods. Experiments on popular knowledge editing methods demonstrate that editing with documents presents significantly greater challenges than using triples. In document-based scenarios, even the best-performing in-context editing approach still lags behind by 10 points in editing success when compared to using gold triples. This observation also holds for both reasoning and cross-lingual test sets. We further analyze key factors influencing task performance, including the quality of extracted triples, the frequency and position of edited knowledge in documents, various methods for enhancing reasoning, and performance differences across various directions in cross-lingual knowledge editing, which provide valuable insights for future research.
Authors: Karen Zhou, Alexander A. Meitus, Milo Chase, Grace Wang, Anne Mykland, William Howell, Chenhao Tan
Abstract: Do American presidents speak discernibly different from each other? If so, in what ways? And are these differences confined to any single medium of communication? To investigate these questions, this paper introduces a novel metric of uniqueness based on large language models, develops a new lexicon for divisive speech, and presents a framework for assessing the distinctive ways in which presidents speak about their political opponents. Applying these tools to a variety of corpora of presidential speeches, we find considerable evidence that Donald Trump's speech patterns diverge from those of all major party nominees for the presidency in recent history. Trump is significantly more distinctive than his fellow Republicans, whose uniqueness values appear closer to those of the Democrats. Contributing to these differences is Trump's employment of divisive and antagonistic language, particularly when targeting his political opponents. These differences hold across a variety of measurement strategies, arise on both the campaign trail and in official presidential addresses, and do not appear to be an artifact of secular changes in presidential communications.
Authors: Galip \"Umit Yolcu, Moritz Weckbecker, Thomas Wiegand, Wojciech Samek, Sebastian Lapuschkin
Abstract: Deep learning models achieve remarkable performance, yet their decision-making processes often remain opaque. In response, the field of eXplainable Artificial Intelligence (XAI) has grown significantly over the last decade, primarily focusing on feature attribution methods. Complementing this perspective, Data Attribution (DA) has emerged as a promising paradigm that shifts the focus from features to data provenance. However, existing DA approaches suffer from prohibitively high computational costs and memory demands. Additionally, current attribution methods exhibit low sparsity, hindering the discovery of decisive patterns in the data. We introduce DualXDA, a framework for sparse, efficient and explainable DA, comprised of two interlinked approaches for Dual Data Attribution (DualDA) and eXplainable Data Attribution (XDA): With DualDA, we propose efficient and effective DA, leveraging Support Vector Machine theory to provide fast and naturally sparse data attributions for AI predictions. We demonstrate that DualDA achieves high attribution quality, excels at solving a series of evaluated downstream tasks, while at the same time improving explanation time by a factor of up to 4,100,000$\times$ compared to the original Influence Functions method, and up to 11,000$\times$ compared to the method's most efficient approximation from literature. We further introduce XDA, a method for enhancing Data Attribution with capabilities from feature attribution methods to explain why training samples are relevant for the prediction of a test sample in terms of impactful features. Taken together, our contributions in DualXDA ultimately point towards a future of eXplainable AI applied at unprecedented scale, enabling transparent, efficient and novel analysis of even the largest neural architectures fostering a new generation of accountable AI systems. Code at https://github.com/gumityolcu/DualXDA.
Authors: Yefeng Yuan, Yuhong Liu, Liang Cheng
Abstract: The rapid advancements in generative AI and large language models (LLMs) have opened up new avenues for producing synthetic data, particularly in the realm of structured tabular formats, such as product reviews. Despite the potential benefits, concerns regarding privacy leakage have surfaced, especially when personal information is utilized in the training datasets. In addition, there is an absence of a comprehensive evaluation framework capable of quantitatively measuring the quality of the generated synthetic data and their utility for downstream tasks. In response to this gap, we introduce SynEval, an open-source evaluation framework designed to assess the fidelity, utility, and privacy preservation of synthetically generated tabular data via a suite of diverse evaluation metrics. We validate the efficacy of our proposed framework - SynEval - by applying it to synthetic product review data generated by three state-of-the-art LLMs: ChatGPT, Claude, and Llama. Our experimental findings illuminate the trade-offs between various evaluation metrics in the context of synthetic data generation. Furthermore, SynEval stands as a critical instrument for researchers and practitioners engaged with synthetic tabular data,, empowering them to judiciously determine the suitability of the generated data for their specific applications, with an emphasis on upholding user privacy.
Authors: Salvatore Greco, Bartolomeo Vacchetti, Daniele Apiletti, Tania Cerquitelli
Abstract: Concept drift is the phenomenon in which the underlying data distributions and statistical properties of a target domain change over time, leading to a degradation in model performance. Consequently, production models require continuous drift detection monitoring. Most drift detection methods to date are supervised, relying on ground-truth labels. However, they are inapplicable in many real-world scenarios, as true labels are often unavailable. Although recent efforts have proposed unsupervised drift detectors, many lack the accuracy required for reliable detection or are too computationally intensive for real-time use in high-dimensional, large-scale production environments. Moreover, they often fail to characterize or explain drift effectively. To address these limitations, we propose \textsc{DriftLens}, an unsupervised framework for real-time concept drift detection and characterization. Designed for deep learning classifiers handling unstructured data, \textsc{DriftLens} leverages distribution distances in deep learning representations to enable efficient and accurate detection. Additionally, it characterizes drift by analyzing and explaining its impact on each label. Our evaluation across classifiers and data-types demonstrates that \textsc{DriftLens} (i) outperforms previous methods in detecting drift in 15/17 use cases; (ii) runs at least 5 times faster; (iii) produces drift curves that align closely with actual drift (correlation $\geq\!0.85$); (iv) effectively identifies representative drift samples as explanations.
Authors: Juhwan Choi, Junehyoung Kwon, JungMin Yun, Seunguk Yu, YoungBin Kim
Abstract: Domain generalizability is a crucial aspect of a deep learning model since it determines the capability of the model to perform well on data from unseen domains. However, research on the domain generalizability of deep learning models for vision-language tasks remains limited, primarily because of the lack of required datasets. To address these challenges, we propose VolDoGer: Vision-Language Dataset for Domain Generalization, a dedicated dataset designed for domain generalization that addresses three vision-language tasks: image captioning, visual question answering, and visual entailment. We constructed VolDoGer by extending LLM-based data annotation techniques to vision-language tasks, thereby alleviating the burden of recruiting human annotators. We evaluated the domain generalizability of various models, ranging from fine-tuned models to a recent multimodal large language model, through VolDoGer.
Authors: Jingrui Hou, Axel Finke, Georgina Cosma
Abstract: We address the problem of machine unlearning in neural information retrieval (IR), introducing a novel task termed Neural Machine UnRanking (NuMuR). This problem is motivated by growing demands for data privacy compliance and selective information removal in neural IR systems. Existing task- or model- agnostic unlearning approaches, primarily designed for classification tasks, are suboptimal for NuMuR due to two core challenges: (1) neural rankers output unnormalised relevance scores rather than probability distributions, limiting the effectiveness of traditional teacher-student distillation frameworks; and (2) entangled data scenarios, where queries and documents appear simultaneously across both forget and retain sets, may degrade retention performance in existing methods. To address these issues, we propose Contrastive and Consistent Loss (CoCoL), a dual-objective framework. CoCoL comprises (1) a contrastive loss that reduces relevance scores on forget sets while maintaining performance on entangled samples, and (2) a consistent loss that preserves accuracy on retain set. Extensive experiments on MS MARCO and TREC CAR datasets, across four neural IR models, demonstrate that CoCoL achieves substantial forgetting with minimal retain and generalisation performance loss. Our method facilitates more effective and controllable data removal than existing techniques.
Authors: Sheng Zhong, Nima Fazeli, Dmitry Berenson
Abstract: This paper presents Rummaging Using Mutual Information (RUMI), a method for online generation of robot action sequences to gather information about the pose of a known movable object in visually-occluded environments. Focusing on contact-rich rummaging, our approach leverages mutual information between the object pose distribution and robot trajectory for action planning. From an observed partial point cloud, RUMI deduces the compatible object pose distribution and approximates the mutual information of it with workspace occupancy in real time. Based on this, we develop an information gain cost function and a reachability cost function to keep the object within the robot's reach. These are integrated into a model predictive control (MPC) framework with a stochastic dynamics model, updating the pose distribution in a closed loop. Key contributions include a new belief framework for object pose estimation, an efficient information gain computation strategy, and a robust MPC-based control scheme. RUMI demonstrates superior performance in both simulated and real tasks compared to baseline methods.
Authors: Ziming Yu, Pan Zhou, Sike Wang, Jia Li, Mi Tian, Hua Huang
Abstract: Fine-tuning Large Language Models (LLMs) has proven effective for a variety of downstream tasks. However, as LLMs grow in size, the memory demands for backpropagation become increasingly prohibitive. Zeroth-order (ZO) optimization methods offer a memory-efficient alternative by using forward passes to estimate gradients, but the variance of gradient estimates typically scales linearly with the model's parameter dimension$\unicode{x2013}$a significant issue for LLMs. In this paper, we propose the random Subspace Zeroth-order (SubZero) optimization to address the challenges posed by LLMs' high dimensionality. We introduce a low-rank perturbation tailored for LLMs that significantly reduces memory consumption while improving training performance. Additionally, we prove that our gradient estimation closely approximates the backpropagation gradient, exhibits lower variance than traditional ZO methods, and ensures convergence when combined with SGD. Experimental results show that SubZero enhances fine-tuning performance and achieves faster convergence compared to standard ZO approaches like MeZO across various language modeling tasks. Code is available at https://github.com/zimingyy/SubZero.
Authors: Yonghyeon Lee
Abstract: Real-time motion generation -- which is essential for achieving reactive and adaptive behavior -- under kinodynamic constraints for high-dimensional systems is a crucial yet challenging problem. We address this with a two-step approach: offline learning of a lower-dimensional trajectory manifold of task-relevant, constraint-satisfying trajectories, followed by rapid online search within this manifold. Extending the discrete-time Motion Manifold Primitives (MMP) framework, we propose Differentiable Motion Manifold Primitives (DMMP), a novel neural network architecture that encodes and generates continuous-time, differentiable trajectories, trained using data collected offline through trajectory optimizations, with a strategy that ensures constraint satisfaction -- absent in existing methods. Experiments on dynamic throwing with a 7-DoF robot arm demonstrate that DMMP outperforms prior methods in planning speed, task success, and constraint satisfaction.
Authors: Zhaorui Tan, Xi Yang, Tan Pan, Tianyi Liu, Chen Jiang, Xin Guo, Qiufeng Wang, Anh Nguyen, Yuan Qi, Kaizhu Huang, Yuan Cheng
Abstract: Variations in medical imaging modalities and individual anatomical differences pose challenges to cross-modality generalization in multi-modal tasks. Existing methods often concentrate exclusively on common anatomical patterns, thereby neglecting individual differences and consequently limiting their generalization performance. This paper emphasizes the critical role of learning individual-level invariance, i.e., personalized representation $\mathbb{X}_h$, to enhance multi-modality generalization under both homogeneous and heterogeneous settings. It reveals that mappings from individual biological profile to different medical modalities remain static across the population, which is implied in the personalization process. We propose a two-stage approach: pre-training with invariant representation $\mathbb{X}_h$ for personalization, then fine-tuning for diverse downstream tasks. We provide both theoretical and empirical evidence demonstrating the feasibility and advantages of personalization, showing that our approach yields greater generalizability and transferability across diverse multi-modal medical tasks compared to methods lacking personalization. Extensive experiments further validate that our approach significantly enhances performance in various generalization scenarios.
Authors: Jingrui Hou, Axel Finke, Georgina Cosma
Abstract: Machine unlearning in neural information retrieval (IR) systems requires removing specific data whilst maintaining model performance. Applying existing machine unlearning methods to IR may compromise retrieval effectiveness or inadvertently expose unlearning actions due to the removal of particular items from the retrieved results presented to users. We formalise corrective unranking, which extends machine unlearning in (neural) IR context by integrating substitute documents to preserve ranking integrity, and propose a novel teacher-student framework, Corrective unRanking Distillation (CuRD), for this task. CuRD (1) facilitates forgetting by adjusting the (trained) neural IR model such that its output relevance scores of to-be-forgotten samples mimic those of low-ranking, non-retrievable samples; (2) enables correction by fine-tuning the relevance scores for the substitute samples to match those of corresponding to-be-forgotten samples closely; (3) seeks to preserve performance on samples that are not targeted for forgetting. We evaluate CuRD on four neural IR models (BERTcat, BERTdot, ColBERT, PARADE) using MS MARCO and TREC CAR datasets. Experiments with forget set sizes from 1 % and 20 % of the training dataset demonstrate that CuRD outperforms seven state-of-the-art baselines in terms of forgetting and correction while maintaining model retention and generalisation capabilities.
Authors: Qing Cheng, Zefan Zeng, Xingchen Hu, Yuehang Si, Zhong Liu
Abstract: Event Causality Identification (ECI) has become an essential task in Natural Language Processing (NLP), focused on automatically detecting causal relationships between events within texts. This comprehensive survey systematically investigates fundamental concepts and models, developing a systematic taxonomy and critically evaluating diverse models. We begin by defining core concepts, formalizing the ECI problem, and outlining standard evaluation protocols. Our classification framework divides ECI models into two primary tasks: Sentence-level Event Causality Identification (SECI) and Document-level Event Causality Identification (DECI). For SECI, we review models employing feature pattern-based matching, machine learning classifiers, deep semantic encoding, prompt-based fine-tuning, and causal knowledge pre-training, alongside data augmentation strategies. For DECI, we focus on approaches utilizing deep semantic encoding, event graph reasoning, and prompt-based fine-tuning. Special attention is given to recent advancements in multi-lingual and cross-lingual ECI, as well as zero-shot ECI leveraging Large Language Models (LLMs). We analyze the strengths, limitations, and unresolved challenges associated with each approach. Extensive quantitative evaluations are conducted on four benchmark datasets to rigorously assess the performance of various ECI models. We conclude by discussing future research directions and highlighting opportunities to advance the field further.
Authors: Hao Ai, Yu-xi Liu
Abstract: To demonstrate supremacy of quantum computing, increasingly large-scale superconducting quantum computing chips are being designed and fabricated. However, the complexity of simulating quantum systems poses a significant challenge to computer-aided design of quantum chips, especially for large-scale chips. Harnessing the scalability of graph neural networks (GNNs), we here propose a parameter designing algorithm for large-scale superconducting quantum circuits. The algorithm depends on the so-called 'three-stair scaling' mechanism, which comprises two neural-network models: an evaluator supervisedly trained on small-scale circuits for applying to medium-scale circuits, and a designer unsupervisedly trained on medium-scale circuits for applying to large-scale ones. We demonstrate our algorithm in mitigating quantum crosstalk errors. Frequencies for both single- and two-qubit gates (corresponding to the parameters of nodes and edges) are considered simultaneously. Numerical results indicate that the well-trained designer achieves notable advantages in efficiency, effectiveness, and scalability. For example, for large-scale superconducting quantum circuits consisting of around 870 qubits, our GNNs-based algorithm achieves 51% of the errors produced by the state-of-the-art algorithm, with a time reduction from 90 min to 27 sec. Overall, a better-performing and more scalable algorithm for designing parameters of superconducting quantum chips is proposed, which initially demonstrates the advantages of applying GNNs in superconducting quantum chips.
Authors: Peter E. D. Love, Jane Matthews, Weili Fang, Hadi Mahamivanan
Abstract: Explainable Artificial Intelligence seeks to make the reasoning processes of AI models transparent and interpretable, particularly in complex decision making environments. In the construction industry, where AI based decision support systems are increasingly adopted, limited attention has been paid to the integration of supporting evidence that underpins the reliability and accountability of AI generated outputs. The absence of such evidence undermines the validity of explanations and the trustworthiness of system recommendations. This paper addresses this gap by introducing a theoretical, evidence based means end framework developed through a narrative review. The framework offers an epistemic foundation for designing XAI enabled DSS that generate meaningful explanations tailored to users knowledge needs and decision contexts. It focuses on evaluating the strength, relevance, and utility of different types of evidence supporting AI generated explanations. While developed with construction professionals as primary end users, the framework is also applicable to developers, regulators, and project managers with varying epistemic goals.
Authors: Imran Pervez, Ricardo Pinto Lima, Omar Knio
Abstract: We develop novel integrated learning and optimization (ILO) methodologies to solve economic dispatch (ED) and DC optimal power flow (DCOPF) problems for better economic operation. The optimization problem for ED is formulated with load being an unknown parameter while DCOPF consists of load and power transfer distribution factor (PTDF) matrix as unknown parameters. PTDF represents the incremental variations of real power on transmission lines which occur due to real power transfers between two regions. These values represent a linearized approximation of power flows over the transmission lines. We develop novel ILO formulations to solve post-hoc penalties in electricity market and line congestion problems using ED and DCOPF optimization formulations. Our proposed methodologies capture the real-time electricity market and line congestion behavior to train the regret function which eventually train unknown loads at different buses and line PTDF matrix to achieve the afore-mentioned post-hoc goals. The proposed methodology is compared to sequential learning and optimization (SLO) which train load and PTDF forecasts for accuracy rather than economic operation. Our experimentation prove the superiority of ILO in minimizing the post-hoc penalties in electricity markets and minimizing the line congestion thereby improving the economic operation with noticeable amount.
Authors: Julien Lesca
Abstract: This paper studies an online variant of the celebrated housing market problem, where each agent has a single house and seeks to exchange it for another based on her preferences. In this online setting, agents may arrive and depart at any time, meaning that not all agents are present on the housing market simultaneously. I extend the well known serial dictatorship and Gale s top trading cycle mechanisms to this online scenario, aiming to retain their desirable properties such as Pareto efficiency, individual rationality, and strategy proofness. These extensions also seek to prevent agents from strategically delaying their arrival or advancing their departure. I demonstrate that achieving all of these properties simultaneously is impossible in the online context, and I present several variants that achieve different subsets of these properties.
Authors: Mithun Saha, Maxwell A. Xu, Wanting Mao, Sameer Neupane, James M. Rehg, Santosh Kumar
Abstract: Photoplethysmography (PPG)-based foundation models are gaining traction due to the widespread use of PPG in biosignal monitoring and their potential to generalize across diverse health applications. In this paper, we introduce Pulse-PPG, the first open-source PPG foundation model trained exclusively on raw PPG data collected over a 100-day field study with 120 participants. Existing PPG foundation models are either open-source but trained on clinical data or closed-source, limiting their applicability in real-world settings. We evaluate Pulse-PPG across multiple datasets and downstream tasks, comparing its performance against a state-of-the-art foundation model trained on clinical data. Our results demonstrate that Pulse-PPG, trained on uncurated field data, exhibits superior generalization across clinical and mobile health applications in both lab and field settings. This suggests that exposure to real-world variability enables the model to learn fine-grained representations, making it more adaptable across tasks. Furthermore, pre-training on field data surprisingly outperforms its pre-training on clinical data in many tasks, reinforcing the importance of training on real-world, diverse datasets. To encourage further advancements in robust foundation models leveraging field data, we plan to release Pulse-PPG, providing researchers with a powerful resource for developing more generalizable PPG-based models.
Authors: Bowen Jin, Jinsung Yoon, Zhen Qin, Ziqi Wang, Wei Xiong, Yu Meng, Jiawei Han, Sercan O. Arik
Abstract: Large Language Models (LLMs) have revolutionized artificial intelligence with capabilities in reasoning, coding, and communication, driving innovation across industries. Their true potential depends on effective alignment to ensure correct, trustworthy and ethical behavior, addressing challenges like misinformation, hallucinations, bias and misuse. While existing Reinforcement Learning (RL)-based alignment methods are notoriously complex, direct optimization approaches offer a simpler alternative. In this work, we introduce a novel direct optimization approach for LLM alignment by drawing on established Information Retrieval (IR) principles. We present a systematic framework that bridges LLM alignment and IR methodologies, mapping LLM generation and reward models to IR's retriever-reranker paradigm. Building on this foundation, we propose LLM Alignment as Retriever Preference Optimization (LarPO), a new alignment method that enhances overall alignment quality. Extensive experiments validate LarPO's effectiveness with 38.9 % and 13.7 % averaged improvement on AlpacaEval2 and MixEval-Hard respectively. Our work opens new avenues for advancing LLM alignment by integrating IR foundations, offering a promising direction for future research.
Authors: Haiwen Diao, Xiaotong Li, Yufeng Cui, Yueze Wang, Haoge Deng, Ting Pan, Wenxuan Wang, Huchuan Lu, Xinlong Wang
Abstract: Existing encoder-free vision-language models (VLMs) are rapidly narrowing the performance gap with their encoder-based counterparts, highlighting the promising potential for unified multimodal systems with structural simplicity and efficient deployment. We systematically clarify the performance gap between VLMs using pre-trained vision encoders, discrete tokenizers, and minimalist visual layers from scratch, deeply excavating the under-examined characteristics of encoder-free VLMs. We develop efficient strategies for encoder-free VLMs that rival mainstream encoder-based ones. After an in-depth investigation, we launch EVEv2.0, a new and improved family of encoder-free VLMs. We show that: (i) Properly decomposing and hierarchically associating vision and language within a unified model reduces interference between modalities. (ii) A well-designed training strategy enables effective optimization for encoder-free VLMs. Through extensive evaluation, our EVEv2.0 represents a thorough study for developing a decoder-only architecture across modalities, demonstrating superior data efficiency and strong vision-reasoning capability. Code is publicly available at: https://github.com/baaivision/EVE.
Authors: Martina Miliani, Serena Auriemma, Alessandro Bondielli, Emmanuele Chersoni, Lucia Passaro, Irene Sucameli, Alessandro Lenci
Abstract: Large Language Models (LLMs) are increasingly used in tasks requiring interpretive and inferential accuracy. In this paper, we introduce ExpliCa, a new dataset for evaluating LLMs in explicit causal reasoning. ExpliCa uniquely integrates both causal and temporal relations presented in different linguistic orders and explicitly expressed by linguistic connectives. The dataset is enriched with crowdsourced human acceptability ratings. We tested LLMs on ExpliCa through prompting and perplexity-based metrics. We assessed seven commercial and open-source LLMs, revealing that even top models struggle to reach 0.80 accuracy. Interestingly, models tend to confound temporal relations with causal ones, and their performance is also strongly influenced by the linguistic order of the events. Finally, perplexity-based scores and prompting performance are differently affected by model size.
Authors: Jixiu Zhai, Tianchi Lu, Haitian Zhong, Ziyang Xu, Yuhuan Liu, Shengrui Xu, Jingwan Wang, Dan Huang
Abstract: Accurate identification of bioactive peptides (BPs) and protein post-translational modifications (PTMs) is essential for understanding protein function and advancing therapeutic discovery. However, most computational methods remain limited in their generalizability across diverse peptide functions. Here, we present PDeepPP, a unified deep learning framework that integrates pretrained protein language models with a hybrid transformer-convolutional architecture, enabling robust identification across diverse peptide classes and PTM sites. We curated comprehensive benchmark datasets and implemented strategies to address data imbalance, allowing PDeepPP to systematically extract both global and local sequence features. Through extensive analyses-including dimensionality reduction and comparison studies-PDeepPP demonstrates strong, interpretable peptide representations and achieves state-of-the-art performance in 25 of the 33 biological identification tasks. Notably, PDeepPP attains high accuracy in antimicrobial (0.9726) and phosphorylation site (0.9984) identification, with 99.5% specificity in glycosylation site prediction and substantial reduction in false negatives in antimalarial tasks. By enabling large-scale, accurate peptide analysis, PDeepPP supports biomedical research and the discovery of novel therapeutic targets for disease treatment. All code, datasets, and pretrained models are publicly available via GitHub:https://github.com/fondress/PDeepPP and Hugging Face:https://huggingface.co/fondress/PDeppPP.
URLs: https://github.com/fondress/PDeepPP, https://huggingface.co/fondress/PDeppPP.
Authors: Markus Bertl, Alan Mott, Salvatore Sinno, Bhavika Bhalgamiya
Abstract: The digitization of healthcare presents numerous challenges, including the complexity of biological systems, vast data generation, and the need for personalized treatment plans. Traditional computational methods often fall short, leading to delayed and sometimes ineffective diagnoses and treatments. Quantum Computing (QC) and Quantum Machine Learning (QML) offer transformative advancements with the potential to revolutionize medicine. This paper summarizes areas where QC promises unprecedented computational power, enabling faster, more accurate diagnostics, personalized treatments, and enhanced drug discovery processes. However, integrating quantum technologies into precision medicine also presents challenges, including errors in algorithms and high costs. We show that mathematically-based techniques for specifying, developing, and verifying software (formal methods) can enhance the reliability and correctness of QC. By providing a rigorous mathematical framework, formal methods help to specify, develop, and verify systems with high precision. In genomic data analysis, formal specification languages can precisely (1) define the behavior and properties of quantum algorithms designed to identify genetic markers associated with diseases. Model checking tools can systematically explore all possible states of the algorithm to (2) ensure it behaves correctly under all conditions, while theorem proving techniques provide mathematical (3) proof that the algorithm meets its specified properties, ensuring accuracy and reliability. Additionally, formal optimization techniques can (4) enhance the efficiency and performance of quantum algorithms by reducing resource usage, such as the number of qubits and gate operations. Therefore, we posit that formal methods can significantly contribute to enabling QC to realize its full potential as a game changer in precision medicine.
Authors: Seunghoi Kim, Henry F. J. Tregidgo, Matteo Figini, Chen Jin, Sarang Joshi, Daniel C. Alexander
Abstract: Hallucinations are spurious structures not present in the ground truth, posing a critical challenge in medical image reconstruction, especially for data-driven conditional models. We hypothesize that combining an unconditional diffusion model with data consistency, trained on a diverse dataset, can reduce these hallucinations. Based on this, we propose DynamicDPS, a diffusion-based framework that integrates conditional and unconditional diffusion models to enhance low-quality medical images while systematically reducing hallucinations. Our approach first generates an initial reconstruction using a conditional model, then refines it with an adaptive diffusion-based inverse problem solver. DynamicDPS skips early stage in the reverse process by selecting an optimal starting time point per sample and applies Wolfe's line search for adaptive step sizes, improving both efficiency and image fidelity. Using diffusion priors and data consistency, our method effectively reduces hallucinations from any conditional model output. We validate its effectiveness in Image Quality Transfer for low-field MRI enhancement. Extensive evaluations on synthetic and real MR scans, including a downstream task for tissue volume estimation, show that DynamicDPS reduces hallucinations, improving relative volume estimation by over 15% for critical tissues while using only 5% of the sampling steps required by baseline diffusion models. As a model-agnostic and fine-tuning-free approach, DynamicDPS offers a robust solution for hallucination reduction in medical imaging. The code will be made publicly available upon publication.
Authors: Jie Xu, Na Zhao, Gang Niu, Masashi Sugiyama, Xiaofeng Zhu
Abstract: Recently, multi-view learning (MVL) has garnered significant attention due to its ability to fuse discriminative information from multiple views. However, real-world multi-view datasets are often heterogeneous and imperfect, which usually causes MVL methods designed for specific combinations of views to lack application potential and limits their effectiveness. To address this issue, we propose a novel robust MVL method (namely RML) with simultaneous representation fusion and alignment. Specifically, we introduce a simple yet effective multi-view transformer fusion network where we transform heterogeneous multi-view data into homogeneous word embeddings, and then integrate multiple views by the sample-level attention mechanism to obtain a fused representation. Furthermore, we propose a simulated perturbation based multi-view contrastive learning framework that dynamically generates the noise and unusable perturbations for simulating imperfect data conditions. The simulated noisy and unusable data obtain two distinct fused representations, and we utilize contrastive learning to align them for learning discriminative and robust representations. Our RML is self-supervised and can also be applied for downstream tasks as a regularization. In experiments, we employ it in multi-view unsupervised clustering, noise-label classification, and as a plug-and-play module for cross-modal hashing retrieval. Extensive comparison experiments and ablation studies validate RML's effectiveness. Code is available at https://github.com/SubmissionsIn/RML.
Authors: Junwei Luo, Yingying Zhang, Xue Yang, Kang Wu, Qi Zhu, Lei Liang, Jingdong Chen, Yansheng Li
Abstract: Efficient vision-language understanding of large Remote Sensing Images (RSIs) is meaningful but challenging. Current Large Vision-Language Models (LVLMs) typically employ limited pre-defined grids to process images, leading to information loss when handling gigapixel RSIs. Conversely, using unlimited grids significantly increases computational costs. To preserve image details while reducing computational complexity, we propose a text-guided token pruning method with Dynamic Image Pyramid (DIP) integration. Our method introduces: (i) a Region Focus Module (RFM) that leverages text-aware region localization capability to identify critical vision tokens, and (ii) a coarse-to-fine image tile selection and vision token pruning strategy based on DIP, which is guided by RFM outputs and avoids directly processing the entire large imagery. Additionally, existing benchmarks for evaluating LVLMs' perception ability on large RSI suffer from limited question diversity and constrained image sizes. We construct a new benchmark named LRS-VQA, which contains 7,333 QA pairs across 8 categories, with image length up to 27,328 pixels. Our method outperforms existing high-resolution strategies on four datasets using the same data. Moreover, compared to existing token reduction methods, our approach demonstrates higher efficiency under high-resolution settings. Dataset and code are in https://github.com/VisionXLab/LRS-VQA.
Authors: Wonwoong Cho, Yan-Ying Chen, Matthew Klenk, David I. Inouye, Yanxia Zhang
Abstract: Text-to-Image (T2I) Diffusion Models have achieved remarkable performance in generating high quality images. However, enabling precise control of continuous attributes, especially multiple attributes simultaneously, in a new domain (e.g., numeric values like eye openness or car width) with text-only guidance remains a significant challenge. To address this, we introduce the Attribute (Att) Adapter, a novel plug-and-play module designed to enable fine-grained, multi-attributes control in pretrained diffusion models. Our approach learns a single control adapter from a set of sample images that can be unpaired and contain multiple visual attributes. The Att-Adapter leverages the decoupled cross attention module to naturally harmonize the multiple domain attributes with text conditioning. We further introduce Conditional Variational Autoencoder (CVAE) to the Att-Adapter to mitigate overfitting, matching the diverse nature of the visual world. Evaluations on two public datasets show that Att-Adapter outperforms all LoRA-based baselines in controlling continuous attributes. Additionally, our method enables a broader control range and also improves disentanglement across multiple attributes, surpassing StyleGAN-based techniques. Notably, Att-Adapter is flexible, requiring no paired synthetic data for training, and is easily scalable to multiple attributes within a single model.
Authors: Junming Liu, Siyuan Meng, Yanting Gao, Song Mao, Pinlong Cai, Guohang Yan, Yirong Chen, Zilin Bian, Ding Wang, Botian Shi
Abstract: Multimodal reasoning in Large Language Models (LLMs) struggles with incomplete knowledge and hallucination artifacts, challenges that textual Knowledge Graphs (KGs) only partially mitigate due to their modality isolation. While Multimodal Knowledge Graphs (MMKGs) promise enhanced cross-modal understanding, their practical construction is impeded by semantic narrowness of manual text annotations and inherent noise in visual-semantic entity linkages. In this paper, we propose Vision-align-to-Language integrated Knowledge Graph (VaLiK), a novel approach for constructing MMKGs that enhances LLMs reasoning through cross-modal information supplementation. Specifically, we cascade pre-trained Vision-Language Models (VLMs) to align image features with text, transforming them into descriptions that encapsulate image-specific information. Furthermore, we developed a cross-modal similarity verification mechanism to quantify semantic consistency, effectively filtering out noise introduced during feature alignment. Even without manually annotated image captions, the refined descriptions alone suffice to construct the MMKG. Compared to conventional MMKGs construction paradigms, our approach achieves substantial storage efficiency gains while maintaining direct entity-to-image linkage capability. Experimental results on multimodal reasoning tasks demonstrate that LLMs augmented with VaLiK outperform previous state-of-the-art models. Our code is published at https://github.com/Wings-Of-Disaster/VaLiK.
Authors: Anshumann, Mohd Abbas Zaidi, Akhil Kedia, Jinwoo Ahn, Taehwak Kwon, Kangwook Lee, Haejun Lee, Joohyung Lee
Abstract: Knowledge distillation can be a cost-effective technique to distill knowledge in Large Language Models, if the teacher output logits can be pre-computed and cached. However, successfully applying this to pre-training remains largely unexplored. In this work, we prove that naive approaches for sparse knowledge distillation such as caching Top-K probabilities, while intuitive, provide biased estimates of teacher probability distribution to the student, resulting in suboptimal performance and calibration. We propose an importance-sampling-based method `Random Sampling Knowledge Distillation', which provides unbiased estimates, preserves the gradient in expectation, and requires storing significantly sparser logits. Our method enables faster training of student models with marginal overhead (<10%) compared to cross-entropy based training, while maintaining competitive performance compared to full distillation, across a range of model sizes from 300M to 3B.
Authors: Jie Zhang, Zhongqi Wang, Shiguang Shan, Xilin Chen
Abstract: Backdoor attacks targeting text-to-image diffusion models have advanced rapidly. However, current backdoor samples often exhibit two key abnormalities compared to benign samples: 1) Semantic Consistency, where backdoor prompts tend to generate images with similar semantic content even with significant textual variations to the prompts; 2) Attention Consistency, where the trigger induces consistent structural responses in the cross-attention maps. These consistencies leave detectable traces for defenders, making backdoors easier to identify. In this paper, toward stealthy backdoor samples, we propose Trigger without Trace (TwT) by explicitly mitigating these consistencies. Specifically, our approach leverages syntactic structures as backdoor triggers to amplify the sensitivity to textual variations, effectively breaking down the semantic consistency. Besides, a regularization method based on Kernel Maximum Mean Discrepancy (KMMD) is proposed to align the distribution of cross-attention responses between backdoor and benign samples, thereby disrupting attention consistency. Extensive experiments demonstrate that our method achieves a 97.5% attack success rate while exhibiting stronger resistance to defenses. It achieves an average of over 98% backdoor samples bypassing three state-of-the-art detection mechanisms, revealing the vulnerabilities of current backdoor defense methods. The code is available at https://github.com/Robin-WZQ/TwT.
Authors: Manlai Liang, JiaMing Zhang, Xiong Li, Jinlong Li
Abstract: The increasing size of the Key-Value (KV) cache during the Large Language Models long-context inference is the main obstacle for its balance between the deployment cost and task accuracy. To reduce the KV cache size in such scenarios, most previous efforts leveraged on the attention weight to evict non-critical cache tokens. But there is a trade-off in those methods, they usually require major modification of the inference infrastructure and significant computation overhead. Based on the fact that the Large Language models are autoregressive models, we propose LagKV, a KV compression strategy only relying on straight forward comparison among KV themselves. It is a totally attention free method which offers easy integration to the main stream inference platform and comparable performance comparing to other complicated KV compression methods. Results on RULER benchmark show that, our approach outperforms SnapKV and StreamingLLM in different compression ratios. Especially in the 64-digit passkey retrieval task, our method outperforms the attention weight based method $H_2O$ over $50\%$ with same compression ratios. Our code is available at https://github.com/AI-Lab-China-Merchants-Bank/LagKV.
Authors: Patrik Reizinger, Randall Balestriero, David Klindt, Wieland Brendel
Abstract: Self-Supervised Learning (SSL) powers many current AI systems. As research interest and investment grow, the SSL design space continues to expand. The Platonic view of SSL, following the Platonic Representation Hypothesis (PRH), suggests that despite different methods and engineering approaches, all representations converge to the same Platonic ideal. However, this phenomenon lacks precise theoretical explanation. By synthesizing evidence from Identifiability Theory (IT), we show that the PRH can emerge in SSL. However, current IT cannot explain SSL's empirical success. To bridge the gap between theory and practice, we propose expanding IT into what we term Singular Identifiability Theory (SITh), a broader theoretical framework encompassing the entire SSL pipeline. SITh would allow deeper insights into the implicit data assumptions in SSL and advance the field towards learning more interpretable and generalizable representations. We highlight three critical directions for future research: 1) training dynamics and convergence properties of SSL; 2) the impact of finite samples, batch size, and data diversity; and 3) the role of inductive biases in architecture, augmentations, initialization schemes, and optimizers.
Authors: Jang Hyun Cho, Andrea Madotto, Effrosyni Mavroudi, Triantafyllos Afouras, Tushar Nagarajan, Muhammad Maaz, Yale Song, Tengyu Ma, Shuming Hu, Suyog Jain, Miguel Martin, Huiyu Wang, Hanoona Rasheed, Peize Sun, Po-Yao Huang, Daniel Bolya, Nikhila Ravi, Shashank Jain, Tammy Stark, Shane Moon, Babak Damavandi, Vivian Lee, Andrew Westbury, Salman Khan, Philipp Kr\"ahenb\"uhl, Piotr Doll\'ar, Lorenzo Torresani, Kristen Grauman, Christoph Feichtenhofer
Abstract: Vision-language models are integral to computer vision research, yet many high-performing models remain closed-source, obscuring their data, design and training recipe. The research community has responded by using distillation from black-box models to label training data, achieving strong benchmark results, at the cost of measurable scientific progress. However, without knowing the details of the teacher model and its data sources, scientific progress remains difficult to measure. In this paper, we study building a Perception Language Model (PLM) in a fully open and reproducible framework for transparent research in image and video understanding. We analyze standard training pipelines without distillation from proprietary models and explore large-scale synthetic data to identify critical data gaps, particularly in detailed video understanding. To bridge these gaps, we release 2.8M human-labeled instances of fine-grained video question-answer pairs and spatio-temporally grounded video captions. Additionally, we introduce PLM-VideoBench, a suite for evaluating challenging video understanding tasks focusing on the ability to reason about "what", "where", "when", and "how" of a video. We make our work fully reproducible by providing data, training recipes, code & models. https://github.com/facebookresearch/perception_models
Authors: Saber Mehdipour, Seyed Abolghasem Mirroshandel, Seyed Amirhossein Tabatabaei
Abstract: Detecting plant diseases is a crucial aspect of modern agriculture, as it plays a key role in maintaining crop health and increasing overall yield. Traditional approaches, though still valuable, often rely on manual inspection or conventional machine learning techniques, both of which face limitations in scalability and accuracy. Recently, Vision Transformers (ViTs) have emerged as a promising alternative, offering advantages such as improved handling of long-range dependencies and better scalability for visual tasks. This review explores the application of ViTs in precision agriculture, covering a range of tasks. We begin by introducing the foundational architecture of ViTs and discussing their transition from Natural Language Processing (NLP) to Computer Vision. The discussion includes the concept of inductive bias in traditional models like Convolutional Neural Networks (CNNs), and how ViTs mitigate these biases. We provide a comprehensive review of recent literature, focusing on key methodologies, datasets, and performance metrics. This study also includes a comparative analysis of CNNs and ViTs, along with a review of hybrid models and performance enhancements. Technical challenges such as data requirements, computational demands, and model interpretability are addressed, along with potential solutions. Finally, we outline future research directions and technological advancements that could further support the integration of ViTs in real-world agricultural settings. Our goal with this study is to offer practitioners and researchers a deeper understanding of how ViTs are poised to transform smart and precision agriculture.
Authors: Le-Trung Nguyen, Ael Quelennec, Van-Tam Nguyen, Enzo Tartaglione
Abstract: On-device learning has emerged as a promising direction for AI development, particularly because of its potential to reduce latency issues and mitigate privacy risks associated with device-server communication, while improving energy efficiency. Despite these advantages, significant memory and computational constraints still represent major challenges for its deployment. Drawing on previous studies on low-rank decomposition methods that address activation memory bottlenecks in backpropagation, we propose a novel shortcut approach as an alternative. Our analysis and experiments demonstrate that our method can reduce activation memory usage, even up to $120.09\times$ compared to vanilla training, while also reducing overall training FLOPs up to $1.86\times$ when evaluated on traditional benchmarks.
Authors: Aiman Faiz, Anna Maria De Roberto, Claudio Pascarelli, Gianvito Mitrano, Gianluca Fimiani, Marina Garofano, Genoveffa Tortora, Mariangela Lazoi, Claudio Passino, Alessia Bramanti
Abstract: The management of chronic Heart Failure (HF) presents significant challenges in modern healthcare, requiring continuous monitoring, early detection of exacerbations, and personalized treatment strategies. In this paper, we present a predictive model founded on Machine Learning (ML) techniques to identify patients at HF risk. This model is an ensemble learning approach, a modified stacking technique, that uses two specialized models leveraging clinical and echocardiographic features and then a meta-model to combine the predictions of these two models. We initially assess the model on a real dataset and the obtained results suggest that it performs well in the stratification of patients at HR risk. Specifically, we obtained high sensitivity (95\%), ensuring that nearly all high-risk patients are identified. As for accuracy, we obtained 84\%, which can be considered moderate in some ML contexts. However, it is acceptable given our priority of identifying patients at risk of HF because they will be asked to participate in the telemonitoring program of the PrediHealth research project on which some of the authors of this paper are working. The initial findings also suggest that ML-based risk stratification models can serve as valuable decision-support tools not only in the PrediHealth project but also for healthcare professionals, aiding in early intervention and personalized patient management. To have a better understanding of the value and of potentiality of our predictive model, we also contrasted its results with those obtained by using three baseline models. The preliminary results indicate that our predictive model outperforms these baselines that flatly consider features, \ie not grouping them in clinical and echocardiographic features.
Authors: Fan Chen, Zeyu Jia, Alexander Rakhlin, Tengyang Xie
Abstract: Reinforcement learning with outcome-based feedback faces a fundamental challenge: when rewards are only observed at trajectory endpoints, how do we assign credit to the right actions? This paper provides the first comprehensive analysis of this problem in online RL with general function approximation. We develop a provably sample-efficient algorithm achieving $\widetilde{O}({C_{\rm cov} H^3}/{\epsilon^2})$ sample complexity, where $C_{\rm cov}$ is the coverability coefficient of the underlying MDP. By leveraging general function approximation, our approach works effectively in large or infinite state spaces where tabular methods fail, requiring only that value functions and reward functions can be represented by appropriate function classes. Our results also characterize when outcome-based feedback is statistically separated from per-step rewards, revealing an unavoidable exponential separation for certain MDPs. For deterministic MDPs, we show how to eliminate the completeness assumption, dramatically simplifying the algorithm. We further extend our approach to preference-based feedback settings, proving that equivalent statistical efficiency can be achieved even under more limited information. Together, these results constitute a theoretical foundation for understanding the statistical properties of outcome-based reinforcement learning.
Authors: Murthy L, Subarna Tripathi
Abstract: The risk of misusing text-to-image generative models for malicious uses, especially due to the open-source development of such models, has become a serious concern. As a risk mitigation strategy, attributing generative models with neural fingerprinting is emerging as a popular technique. There has been a plethora of recent work that aim for addressing neural fingerprinting. A trade-off between the attribution accuracy and generation quality of such models has been studied extensively. None of the existing methods yet achieved 100% attribution accuracy. However, any model with less than cent percent accuracy is practically non-deployable. In this work, we propose an accurate method to incorporate neural fingerprinting for text-to-image diffusion models leveraging the concepts of cyclic error correcting codes from the literature of coding theory.
Authors: Xiaochun Lei, Siqi Wu, Weilin Wu, Zetao Jiang
Abstract: Real-time object detection is a fundamental but challenging task in computer vision, particularly when computational resources are limited. Although YOLO-series models have set strong benchmarks by balancing speed and accuracy, the increasing need for richer global context modeling has led to the use of Transformer-based architectures. Nevertheless, Transformers have high computational complexity because of their self-attention mechanism, which limits their practicality for real-time and edge deployments. To overcome these challenges, recent developments in linear state space models, such as Mamba, provide a promising alternative by enabling efficient sequence modeling with linear complexity. Building on this insight, we propose MambaNeXt-YOLO, a novel object detection framework that balances accuracy and efficiency through three key contributions: (1) MambaNeXt Block: a hybrid design that integrates CNNs with Mamba to effectively capture both local features and long-range dependencies; (2) Multi-branch Asymmetric Fusion Pyramid Network (MAFPN): an enhanced feature pyramid architecture that improves multi-scale object detection across various object sizes; and (3) Edge-focused Efficiency: our method achieved 66.6% mAP at 31.9 FPS on the PASCAL VOC dataset without any pre-training and supports deployment on edge devices such as the NVIDIA Jetson Xavier NX and Orin NX.
Authors: Ala Yankouskaya, Areej B. Babiker, Syeda W. F. Rizvi, Sameha Alshakhsi, Magnus Liebherr, Raian Ali
Abstract: There is growing interest in understanding how people interact with large language models (LLMs) and whether such models elicit dependency or even addictive behaviour. Validated tools to assess the extent to which individuals may become dependent on LLMs are scarce and primarily build on classic behavioral addiction symptoms, adapted to the context of LLM use. We view this as a conceptual limitation, as the LLM-human relationship is more nuanced and warrants a fresh and distinct perspective. To address this gap, we developed and validated a new 12-item questionnaire to measure LLM dependency, referred to as LLM-D12. The scale was based on the authors' prior theoretical work, with items developed accordingly and responses collected from 526 participants in the UK. Exploratory and confirmatory factor analyses, performed on separate halves of the total sample using a split-sample approach, supported a two-factor structure: Instrumental Dependency (six items) and Relationship Dependency (six items). Instrumental Dependency reflects the extent to which individuals rely on LLMs to support or collaborate in decision-making and cognitive tasks. Relationship Dependency captures the tendency to perceive LLMs as socially meaningful, sentient, or companion-like entities. The two-factor structure demonstrated excellent internal consistency and clear discriminant validity. External validation confirmed both the conceptual foundation and the distinction between the two subscales. The psychometric properties and structure of our LLM-D12 scale were interpreted in light of the emerging view that dependency on LLMs does not necessarily indicate dysfunction but may still reflect reliance levels that could become problematic in certain contexts.
Authors: Runqian Wang, Kaiming He
Abstract: The development of diffusion-based generative models over the past decade has largely proceeded independently of progress in representation learning. These diffusion models typically rely on regression-based objectives and generally lack explicit regularization. In this work, we propose \textit{Dispersive Loss}, a simple plug-and-play regularizer that effectively improves diffusion-based generative models. Our loss function encourages internal representations to disperse in the hidden space, analogous to contrastive self-supervised learning, with the key distinction that it requires no positive sample pairs and therefore does not interfere with the sampling process used for regression. Compared to the recent method of representation alignment (REPA), our approach is self-contained and minimalist, requiring no pre-training, no additional parameters, and no external data. We evaluate Dispersive Loss on the ImageNet dataset across a range of models and report consistent improvements over widely used and strong baselines. We hope our work will help bridge the gap between generative modeling and representation learning.
Authors: Gregor Baer, Isel Grau, Chao Zhang, Pieter Van Gorp
Abstract: Evaluating feature attribution methods represents a critical challenge in explainable AI (XAI), as researchers typically rely on perturbation-based metrics when ground truth is unavailable. However, recent work reveals that these evaluation metrics can show different performance across predicted classes within the same dataset. These "class-dependent evaluation effects" raise questions about whether perturbation analysis reliably measures attribution quality, with direct implications for XAI method development and evaluation trustworthiness. We investigate under which conditions these class-dependent effects arise by conducting controlled experiments with synthetic time series data where ground truth feature locations are known. We systematically vary feature types and class contrasts across binary classification tasks, then compare perturbation-based degradation scores with ground truth-based precision-recall metrics using multiple attribution methods. Our experiments demonstrate that class-dependent effects emerge with both evaluation approaches, even in simple scenarios with temporally localized features, triggered by basic variations in feature amplitude or temporal extent between classes. Most critically, we find that perturbation-based and ground truth metrics frequently yield contradictory assessments of attribution quality across classes, with weak correlations between evaluation approaches. These findings suggest that researchers should interpret perturbation-based metrics with care, as they may not always align with whether attributions correctly identify discriminating features. By showing this disconnect, our work points toward reconsidering what attribution evaluation actually measures and developing more rigorous evaluation methods that capture multiple dimensions of attribution quality.
Authors: Tianyu Wang, Akira Horiguchi, Lingyou Pang, Carey E. Priebe
Abstract: The increasing use of synthetic data from the public Internet has enhanced data usage efficiency in large language model (LLM) training. However, the potential threat of model collapse remains insufficiently explored. Existing studies primarily examine model collapse in a single model setting or rely solely on statistical surrogates. In this work, we introduce LLM Web Dynamics (LWD), an efficient framework for investigating model collapse at the network level. By simulating the Internet with a retrieval-augmented generation (RAG) database, we analyze the convergence pattern of model outputs. Furthermore, we provide theoretical guarantees for this convergence by drawing an analogy to interacting Gaussian Mixture Models.
Authors: Heng Zhang, Zikang Wan, Danilo Vasconcellos Vargas
Abstract: Human vision excels at segmenting visual cues without the need for explicit training, and it remains remarkably robust even as noise severity increases. In contrast, existing AI algorithms struggle to maintain accuracy under similar conditions. Here, we present SyncMapV2, the first to solve unsupervised segmentation with state-of-the-art robustness. SyncMapV2 exhibits a minimal drop in mIoU, only 0.01%, under digital corruption, compared to a 23.8% drop observed in SOTA methods. This superior performance extends across various types of corruption: noise (7.3% vs. 37.7%), weather (7.5% vs. 33.8%), and blur (7.0% vs. 29.5%). Notably, SyncMapV2 accomplishes this without any robust training, supervision, or loss functions. It is based on a learning paradigm that uses self-organizing dynamical equations combined with concepts from random networks. Moreover, unlike conventional methods that require re-initialization for each new input, SyncMapV2 adapts online, mimicking the continuous adaptability of human vision. Thus, we go beyond the accurate and robust results, and present the first algorithm that can do all the above online, adapting to input rather than re-initializing. In adaptability tests, SyncMapV2 demonstrates near-zero performance degradation, which motivates and fosters a new generation of robust and adaptive intelligence in the near future.
Authors: He-Yang Xu, Hongxiang Gao, Yuwen Li, Xiu-Shen Wei, Chengyu Liu
Abstract: The diagnostic value of electrocardiogram (ECG) lies in its dynamic characteristics, ranging from rhythm fluctuations to subtle waveform deformations that evolve across time and frequency domains. However, supervised ECG models tend to overfit dominant and repetitive patterns, overlooking fine-grained but clinically critical cues, a phenomenon known as Simplicity Bias (SB), where models favor easily learnable signals over subtle but informative ones. In this work, we first empirically demonstrate the presence of SB in ECG analyses and its negative impact on diagnostic performance, while simultaneously discovering that self-supervised learning (SSL) can alleviate it, providing a promising direction for tackling the bias. Following the SSL paradigm, we propose a novel method comprising two key components: 1) Temporal-Frequency aware Filters to capture temporal-frequency features reflecting the dynamic characteristics of ECG signals, and 2) building on this, Multi-Grained Prototype Reconstruction for coarse and fine representation learning across dual domains, further mitigating SB. To advance SSL in ECG analyses, we curate a large-scale multi-site ECG dataset with 1.53 million recordings from over 300 clinical centers. Experiments on three downstream tasks across six ECG datasets demonstrate that our method effectively reduces SB and achieves state-of-the-art performance. Code and dataset will be released publicly.
Authors: Koki Yamane, Yunhan Li, Masashi Konosu, Koki Inami, Junji Oaki, Sho Sakaino, Toshiaki Tsuji
Abstract: In recent years, the advancement of imitation learning has led to increased interest in teleoperating low-cost manipulators to collect demonstration data. However, most existing systems rely on unilateral control, which only transmits target position values. While this approach is easy to implement and suitable for slow, non-contact tasks, it struggles with fast or contact-rich operations due to the absence of force feedback. This work demonstrates that fast teleoperation with force feedback is feasible even with force-sensorless, low-cost manipulators by leveraging 4-channel bilateral control. Based on accurately identified manipulator dynamics, our method integrates nonlinear terms compensation, velocity and external force estimation, and variable gain corresponding to inertial variation. Furthermore, using data collected by 4-channel bilateral control, we show that incorporating force information into both the input and output of learned policies improves performance in imitation learning. These results highlight the practical effectiveness of our system for high-fidelity teleoperation and data collection on affordable hardware.
Authors: Yukang Chen, Wei Huang, Baifeng Shi, Qinghao Hu, Hanrong Ye, Ligeng Zhu, Zhijian Liu, Pavlo Molchanov, Jan Kautz, Xiaojuan Qi, Sifei Liu, Hongxu Yin, Yao Lu, Song Han
Abstract: We introduce a full-stack framework that scales up reasoning in vision-language models (VLMs) to long videos, leveraging reinforcement learning. We address the unique challenges of long video reasoning by integrating three critical components: (1) a large-scale dataset, LongVideo-Reason, comprising 104K long video QA pairs with high-quality reasoning annotations across diverse domains such as sports, games, and vlogs; (2) a two-stage training pipeline that extends VLMs with chain-of-thought supervised fine-tuning (CoT-SFT) and reinforcement learning (RL); and (3) a training infrastructure for long video RL, named Multi-modal Reinforcement Sequence Parallelism (MR-SP), which incorporates sequence parallelism and a vLLM-based engine tailored for long video, using cached video embeddings for efficient rollout and prefilling. In our experiments, LongVILA-R1-7B achieves strong performance on video benchmarks, reaching 65.0% and 70.7% accuracy on VideoMME without and with subtitles, respectively, and consistently outperforming LongVILA-R1 across multiple benchmarks. Moreover, LongVILA-R1 shows steady performance improvements as the number of input video frames increases. Notably, our MR-SP system achieves up to 2.1x speedup on long video RL training. In addition, we release our training system for public availability that supports RL training on various modalities (video, text, and audio), various models (VILA and Qwen series), and even image and video generation models. On a single A100 node (8 GPUs), it supports RL training on hour-long videos (e.g., 3,600 frames / around 256k tokens).
Authors: Pierre Beckmann, Matthieu Queloz
Abstract: Recent findings in mechanistic interpretability (MI), the field probing the inner workings of Large Language Models (LLMs), challenge the view that these models rely solely on superficial statistics. We offer an accessible synthesis of these findings that doubles as an introduction to MI while integrating these findings within a novel theoretical framework for thinking about machine understanding. We argue that LLMs develop internal structures that are functionally analogous to the kind of understanding that consists in seeing connections. To sharpen this idea, we propose a three-tiered conception of understanding. First, conceptual understanding emerges when a model forms "features" as directions in latent space, learning the connections between diverse manifestations of something. Second, state-of-the-world understanding emerges when a model learns contingent factual connections between features and dynamically tracks changes in the world. Third, principled understanding emerges when a model ceases to rely on a collection of memorized facts and discovers a "circuit" connecting these facts. However, these forms of understanding remain radically different from human understanding, as the phenomenon of "parallel mechanisms" shows. We conclude that the debate should move beyond the yes-or-no question of whether LLMs understand to investigate how their strange minds work and forge conceptions that fit them.
Authors: Marcin Pietro\'n, Rafa{\l} Olszowski, Jakub Gomu{\l}ka, Filip Gampel, Andrzej Tomski
Abstract: Argument mining (AM) is an interdisciplinary research field that integrates insights from logic, philosophy, linguistics, rhetoric, law, psychology, and computer science. It involves the automatic identification and extraction of argumentative components, such as premises and claims, and the detection of relationships between them, such as support, attack, or neutrality. Recently, the field has advanced significantly, especially with the advent of large language models (LLMs), which have enhanced the efficiency of analyzing and extracting argument semantics compared to traditional methods and other deep learning models. There are many benchmarks for testing and verifying the quality of LLM, but there is still a lack of research and results on the operation of these models in publicly available argument classification databases. This paper presents a study of a selection of LLM's, using diverse datasets such as Args.me and UKP. The models tested include versions of GPT, Llama, and DeepSeek, along with reasoning-enhanced variants incorporating the Chain-of-Thoughts algorithm. The results indicate that ChatGPT-4o outperforms the others in the argument classification benchmarks. In case of models incorporated with reasoning capabilities, the Deepseek-R1 shows its superiority. However, despite their superiority, GPT-4o and Deepseek-R1 still make errors. The most common errors are discussed for all models. To our knowledge, the presented work is the first broader analysis of the mentioned datasets using LLM and prompt algorithms. The work also shows some weaknesses of known prompt algorithms in argument analysis, while indicating directions for their improvement. The added value of the work is the in-depth analysis of the available argument datasets and the demonstration of their shortcomings.
Authors: Laura Baird, Armin Moin
Abstract: We propose a novel approach, OrQstrator, which is a modular framework for conducting quantum circuit optimization in the Noisy Intermediate-Scale Quantum (NISQ) era. Our framework is powered by Deep Reinforcement Learning (DRL). Our orchestration engine intelligently selects among three complementary circuit optimizers: A DRL-based circuit rewriter trained to reduce depth and gate count via learned rewrite sequences; a domain-specific optimizer that performs efficient local gate resynthesis and numeric optimization; a parameterized circuit instantiator that improves compilation by optimizing template circuits during gate set translation. These modules are coordinated by a central orchestration engine that learns coordination policies based on circuit structure, hardware constraints, and backend-aware performance features such as gate count, depth, and expected fidelity. The system outputs an optimized circuit for hardware-aware transpilation and execution, leveraging techniques from an existing state-of-the-art approach, called the NISQ Analyzer, to adapt to backend constraints.
Authors: Niket Patel, Randall Balestriero
Abstract: The grand goal of AI research, and particularly Self Supervised Learning (SSL), is to produce systems that can successfully solve any possible task. In contrast, current evaluation methods available to AI researchers typically rely on a fixed collection of hand-picked downstream benchmarks. Hence, a large amount of effort is put into designing and searching for large collection of evaluation tasks that can serve as a proxy of our grand goal. We argue that such a rigid evaluation protocol creates a silent bottleneck in AI research. To remedy that, we define a probabilistic space of downstream tasks obtained by adopting a distribution of tasks and by defining Task Priors. Under this view, one can evaluate a model's performance over the set of all possible downstream tasks. Our framework is the first to provide answers to key questions such as (i) what is the average performance of my model over all possible downstream tasks weighted by the probability to encounter each task? or (ii) what is the variance of my model's performance across all downstream tasks under the defined Task Priors? Beyond establishing a new standard for evaluation, we believe that Task Priors will accelerate the pace of research in SSL - where downstream task evaluation is the sole qualitative signal that researchers have access to.
Authors: Zhonglin Liu
Abstract: Innate resistance to anti-PD-1 immunotherapy remains a major clinical challenge in metastatic melanoma, with the underlying molecular networks being poorly understood. To address this, we constructed a dynamic Probabilistic Boolean Network model using transcriptomic data from patient tumor biopsies to elucidate the regulatory logic governing therapy response. We then employed a reinforcement learning agent to systematically discover optimal, multi-step therapeutic interventions and used explainable artificial intelligence to mechanistically interpret the agent's control policy. The analysis revealed that a precisely timed, 4-step temporary inhibition of the lysyl oxidase like 2 protein (LOXL2) was the most effective strategy. Our explainable analysis showed that this ''hit-and-run" intervention is sufficient to erase the molecular signature driving resistance, allowing the network to self-correct without requiring sustained intervention. This study presents a novel, time-dependent therapeutic hypothesis for overcoming immunotherapy resistance and provides a powerful computational framework for identifying non-obvious intervention protocols in complex biological systems.
Authors: Zejian Li, Yize Li, Chenye Meng, Zhongni Liu, Yang Ling, Shengyuan Zhang, Guang Yang, Changyuan Yang, Zhiyuan Yang, Lingyun Sun
Abstract: Recent advancements in diffusion models (DMs) have been propelled by alignment methods that post-train models to better conform to human preferences. However, these approaches typically require computation-intensive training of a base model and a reward model, which not only incurs substantial computational overhead but may also compromise model accuracy and training efficiency. To address these limitations, we propose Inversion-DPO, a novel alignment framework that circumvents reward modeling by reformulating Direct Preference Optimization (DPO) with DDIM inversion for DMs. Our method conducts intractable posterior sampling in Diffusion-DPO with the deterministic inversion from winning and losing samples to noise and thus derive a new post-training paradigm. This paradigm eliminates the need for auxiliary reward models or inaccurate appromixation, significantly enhancing both precision and efficiency of training. We apply Inversion-DPO to a basic task of text-to-image generation and a challenging task of compositional image generation. Extensive experiments show substantial performance improvements achieved by Inversion-DPO compared to existing post-training methods and highlight the ability of the trained generative models to generate high-fidelity compositionally coherent images. For the post-training of compostitional image geneation, we curate a paired dataset consisting of 11,140 images with complex structural annotations and comprehensive scores, designed to enhance the compositional capabilities of generative models. Inversion-DPO explores a new avenue for efficient, high-precision alignment in diffusion models, advancing their applicability to complex realistic generation tasks. Our code is available at https://github.com/MIGHTYEZ/Inversion-DPO
Authors: Gayal Kuruppu, Neeraj Wagh, Yogatheesan Varatharajah
Abstract: Patterns of electrical brain activity recorded via electroencephalography (EEG) offer immense value for scientific and clinical investigations. The inability of supervised EEG encoders to learn robust EEG patterns and their over-reliance on expensive signal annotations have sparked a transition towards general-purpose self-supervised EEG encoders, i.e., EEG foundation models (EEG-FMs), for robust and scalable EEG feature extraction. However, the real-world readiness of early EEG-FMs and the rubric for long-term research progress remain unclear. A systematic and comprehensive review of first-generation EEG-FMs is therefore necessary to understand the current state-of-the-art and identify key directions for future EEG-FMs. To that end, this study reviews 10 early EEG-FMs and presents a critical synthesis of their methodology, empirical findings, and outstanding research gaps. We find that most EEG-FMs adopt a sequence-based modeling scheme that relies on transformer-based backbones and the reconstruction of masked sequences for self-supervision. However, model evaluations remain heterogeneous and largely limited, making it challenging to assess their practical off-the-shelf utility. In addition to adopting standardized and realistic evaluations, future work should demonstrate more substantial scaling effects and make principled and trustworthy choices throughout the EEG representation learning pipeline. We believe that developing benchmarks, software tools, technical methodologies, and applications in collaboration with domain experts may further advance the translational utility and real-world adoption of EEG-FMs.
Authors: Jianzhe Ma, Wenxuan Wang, Qin Jin
Abstract: Geometry problem solving is a key area of mathematical reasoning, which is widely involved in many important fields such as education, mathematical ability assessment of artificial intelligence, and multimodal ability assessment. In recent years, the rapid development of deep learning technology, especially the rise of multimodal large language models, has triggered a widespread research boom. This paper provides a survey of the applications of deep learning in geometry problem solving, including (i) a comprehensive summary of the relevant tasks in geometry problem solving; (ii) a thorough review of related deep learning methods; (iii) a detailed analysis of evaluation metrics and methods; and (iv) a critical discussion of the current challenges and future directions that can be explored. Our goal is to provide a comprehensive and practical reference of deep learning for geometry problem solving to promote further developments in this field. We create a continuously updated list of papers on GitHub: https://github.com/majianz/dl4gps.
Authors: Linwei Chen, Lin Gu, Ying Fu
Abstract: Vision Transformers (ViTs) have significantly advanced computer vision, demonstrating strong performance across various tasks. However, the attention mechanism in ViTs makes each layer function as a low-pass filter, and the stacked-layer architecture in existing transformers suffers from frequency vanishing. This leads to the loss of critical details and textures. We propose a novel, circuit-theory-inspired strategy called Frequency-Dynamic Attention Modulation (FDAM), which can be easily plugged into ViTs. FDAM directly modulates the overall frequency response of ViTs and consists of two techniques: Attention Inversion (AttInv) and Frequency Dynamic Scaling (FreqScale). Since circuit theory uses low-pass filters as fundamental elements, we introduce AttInv, a method that generates complementary high-pass filtering by inverting the low-pass filter in the attention matrix, and dynamically combining the two. We further design FreqScale to weight different frequency components for fine-grained adjustments to the target response function. Through feature similarity analysis and effective rank evaluation, we demonstrate that our approach avoids representation collapse, leading to consistent performance improvements across various models, including SegFormer, DeiT, and MaskDINO. These improvements are evident in tasks such as semantic segmentation, object detection, and instance segmentation. Additionally, we apply our method to remote sensing detection, achieving state-of-the-art results in single-scale settings. The code is available at https://github.com/Linwei-Chen/FDAM.
Authors: Ashray Gupta, Rohan Joseph, Sunny Rai
Abstract: Analogies test a model's ability to infer implicit relationships between concepts, making them a key benchmark for evaluating reasoning capabilities. While large language models (LLMs) are widely evaluated for reasoning in English, their abilities in Indic languages remain understudied, limiting our understanding of whether these models generalize across languages. To address this gap, we introduce a new Hindi Analogy Test Set (HATS), comprising 405 multiple-choice questions sourced from Indian government exams. We benchmark state-of-the-art multilingual LLMs using various prompting strategies and introduce a grounded Chain of Thought approach that leverages cognitive theories of analogical reasoning. This approach improves model performance on Hindi analogy questions. Our experiments show that models perform best with English prompts, irrespective of the prompting strategy. Our test set addresses the lack of a critical resource to evaluate LLM reasoning capabilities in Hindi.
Authors: Jeyoung Lee, Hochul Kang
Abstract: We propose the Signal Dice Similarity Coefficient (SDSC), a structure-aware metric function for time series self-supervised representation learning. Most Self-Supervised Learning (SSL) methods for signals commonly adopt distance-based objectives such as mean squared error (MSE), which are sensitive to amplitude, invariant to waveform polarity, and unbounded in scale. These properties hinder semantic alignment and reduce interpretability. SDSC addresses this by quantifying structural agreement between temporal signals based on the intersection of signed amplitudes, derived from the Dice Similarity Coefficient (DSC).Although SDSC is defined as a structure-aware metric, it can be used as a loss by subtracting from 1 and applying a differentiable approximation of the Heaviside function for gradient-based optimization. A hybrid loss formulation is also proposed to combine SDSC with MSE, improving stability and preserving amplitude where necessary. Experiments on forecasting and classification benchmarks demonstrate that SDSC-based pre-training achieves comparable or improved performance over MSE, particularly in in-domain and low-resource scenarios. The results suggest that structural fidelity in signal representations enhances the semantic representation quality, supporting the consideration of structure-aware metrics as viable alternatives to conventional distance-based methods.
Authors: Zhijie Wang, Zixin Xu, Zhiyuan Pan
Abstract: The exponential growth of spam text on the Internet necessitates robust detection mechanisms to mitigate risks such as information leakage and social instability. This work addresses two principal challenges: adversarial strategies employed by spammers and the scarcity of labeled data. We propose a novel spam-text detection framework GCC-Spam, which integrates three core innovations. First, a character similarity network captures orthographic and phonetic features to counter character-obfuscation attacks and furthermore produces sentence embeddings for downstream classification. Second, contrastive learning enhances discriminability by optimizing the latent-space distance between spam and normal texts. Third, a Generative Adversarial Network (GAN) generates realistic pseudo-spam samples to alleviate data scarcity while improving model robustness and classification accuracy. Extensive experiments on real-world datasets demonstrate that our model outperforms baseline approaches, achieving higher detection rates with significantly fewer labeled examples.
Authors: Derek Li, Jiaming Zhou, Amirreza Kazemi, Qianyi Sun, Abbas Ghaddar, Mohammad Ali Alomrani, Liheng Ma, Yu Luo, Dong Li, Feng Wen, Jianye Hao, Mark Coates, Yingxue Zhang
Abstract: The advancement of general-purpose artificial intelligence relies on large language models (LLMs) that excel across a wide range of tasks, from structured reasoning to creative generation. However, post-training methods like Supervised Fine-Tuning (SFT) often struggle with generalization, favoring memorization over transferable learning. In this work, we introduce Omni-Thinker, a unified reinforcement learning (RL) framework that enhances LLM performance across diverse tasks by combining rule-based verifiable rewards with generative preference signals via LLM-as-a-Judge evaluations. Our approach enables consistent optimization across task types and scales RL-based training to subjective domains. We further investigate training strategies, demonstrating that a curriculum-based progression that orders tasks from structured to open-ended improves performance and reduces forgetting. Experimental results across four domains reveal that curriculum learning improves performance by 5.2% over joint training and 9.1% over model merging. These results highlight the importance of task-aware sampling and hybrid supervision in scaling RL-based post-training for general-purpose LLMs.
Authors: Xinran Li, Xiujuan Xu, Jiaqi Qiao
Abstract: Emotion Recognition in Conversation (ERC) is a practical and challenging task. This paper proposes a novel multimodal approach, the Long-Short Distance Graph Neural Network (LSDGNN). Based on the Directed Acyclic Graph (DAG), it constructs a long-distance graph neural network and a short-distance graph neural network to obtain multimodal features of distant and nearby utterances, respectively. To ensure that long- and short-distance features are as distinct as possible in representation while enabling mutual influence between the two modules, we employ a Differential Regularizer and incorporate a BiAffine Module to facilitate feature interaction. In addition, we propose an Improved Curriculum Learning (ICL) to address the challenge of data imbalance. By computing the similarity between different emotions to emphasize the shifts in similar emotions, we design a "weighted emotional shift" metric and develop a difficulty measurer, enabling a training process that prioritizes learning easy samples before harder ones. Experimental results on the IEMOCAP and MELD datasets demonstrate that our model outperforms existing benchmarks.
Authors: An Wang, Rulin Zhou, Mengya Xu, Yiru Ye, Longfei Gou, Yiting Chang, Hao Chen, Chwee Ming Lim, Jiankun Wang, Hongliang Ren
Abstract: Visualizing subtle vascular motions in endoscopic surgery is crucial for surgical precision and decision-making, yet remains challenging due to the complex and dynamic nature of surgical scenes. To address this, we introduce EndoControlMag, a training-free, Lagrangian-based framework with mask-conditioned vascular motion magnification tailored to endoscopic environments. Our approach features two key modules: a Periodic Reference Resetting (PRR) scheme that divides videos into short overlapping clips with dynamically updated reference frames to prevent error accumulation while maintaining temporal coherence, and a Hierarchical Tissue-aware Magnification (HTM) framework with dual-mode mask dilation. HTM first tracks vessel cores using a pretrained visual tracking model to maintain accurate localization despite occlusions and view changes. It then applies one of two adaptive softening strategies to surrounding tissues: motion-based softening that modulates magnification strength proportional to observed tissue displacement, or distance-based exponential decay that simulates biomechanical force attenuation. This dual-mode approach accommodates diverse surgical scenarios-motion-based softening excels with complex tissue deformations while distance-based softening provides stability during unreliable optical flow conditions. We evaluate EndoControlMag on our EndoVMM24 dataset spanning four different surgery types and various challenging scenarios, including occlusions, instrument disturbance, view changes, and vessel deformations. Quantitative metrics, visual assessments, and expert surgeon evaluations demonstrate that EndoControlMag significantly outperforms existing methods in both magnification accuracy and visual quality while maintaining robustness across challenging surgical conditions. The code, dataset, and video results are available at https://szupc.github.io/EndoControlMag/.
Authors: Sungmin Yun, Seonyong Park, Hwayong Nam, Younjoo Lee, Gunjun Lee, Kwanhee Kyung, Sangpyo Kim, Nam Sung Kim, Jongmin Kim, Hyungyo Kim, Juhwan Cho, Seungmin Baek, Jung Ho Ahn
Abstract: Computational workloads composing traditional Transformer models are starkly bifurcated. Multi-Head Attention (MHA) is memory-bound, with low arithmetic intensity, while feedforward layers are compute-bound. This dichotomy has long motivated research into specialized hardware to mitigate the MHA bottleneck. This paper argues that recent architectural shifts, namely Multi-head Latent Attention (MLA) and Mixture-of-Experts (MoE), challenge the premise of specialized attention hardware. We make two key observations. First, the arithmetic intensity of MLA is over two orders of magnitude greater than that of MHA, shifting it close to a compute-bound regime well-suited for modern accelerators like GPUs. Second, by distributing MoE experts across a pool of accelerators, their arithmetic intensity can be tuned through batching to match that of the dense layers, creating a more balanced computational profile. These findings reveal a diminishing need for specialized attention hardware. The central challenge for next-generation Transformers is no longer accelerating a single memory-bound layer. Instead, the focus must shift to designing balanced systems with sufficient compute, memory capacity, memory bandwidth, and high-bandwidth interconnects to manage the diverse demands of large-scale models.
Authors: Mihir Prabhudesai, Menging Wu, Amir Zadeh, Katerina Fragkiadaki, Deepak Pathak
Abstract: Autoregressive (AR) models have long dominated the landscape of large language models, driving progress across a wide range of tasks. Recently, diffusion-based language models have emerged as a promising alternative, though their advantages over AR models remain underexplored. In this paper, we systematically study masked diffusion models in data-constrained settings-where training involves repeated passes over limited data-and find that they significantly outperform AR models when compute is abundant but data is scarce. Diffusion models make better use of repeated data, achieving lower validation loss and superior downstream performance. We interpret this advantage as implicit data augmentation: masked diffusion exposes the model to a diverse distribution of token orderings and prediction tasks, unlike AR's fixed left-to-right factorization. We find new scaling laws for diffusion models and derive a closed-form expression for the critical compute threshold at which diffusion begins to outperform AR. These results suggest that when data, not compute, is the bottleneck, diffusion models offer a compelling alternative to the standard AR paradigm. Our code is available at: https://diffusion-scaling.github.io.
Authors: Zhehui Huang, Guangyao Shi, Yuwei Wu, Vijay Kumar, Gaurav S. Sukhatme
Abstract: Multi-robot coordination has traditionally relied on a mission-specific and expert-driven pipeline, where natural language mission descriptions are manually translated by domain experts into mathematical formulation, algorithm design, and executable code. This conventional process is labor-intensive, inaccessible to non-experts, and inflexible to changes in mission requirements. Here, we propose LAN2CB (Language to Collective Behavior), a novel framework that leverages large language models (LLMs) to streamline and generalize the multi-robot coordination pipeline. LAN2CB transforms natural language (NL) mission descriptions into executable Python code for multi-robot systems through two core modules: (1) Mission Analysis, which parses mission descriptions into behavior trees, and (2) Code Generation, which leverages the behavior tree and a structured knowledge base to generate robot control code. We further introduce a dataset of natural language mission descriptions to support development and benchmarking. Experiments in both simulation and real-world environments demonstrate that LAN2CB enables robust and flexible multi-robot coordination from natural language, significantly reducing manual engineering effort and supporting broad generalization across diverse mission types. Website: https://sites.google.com/view/lan-cb
Authors: Batu Candan, Simone Servadio
Abstract: Accurate and robust relative pose estimation is crucial for enabling challenging Active Debris Removal (ADR) missions targeting tumbling derelict satellites such as ESA's ENVISAT. This work presents a complete pipeline integrating advanced computer vision techniques with adaptive nonlinear filtering to address this challenge. A Convolutional Neural Network (CNN), enhanced with image preprocessing, detects structural markers (corners) from chaser imagery, whose 2D coordinates are converted to 3D measurements using camera modeling. These measurements are fused within an Unscented Kalman Filter (UKF) framework, selected for its ability to handle nonlinear relative dynamics, to estimate the full relative pose. Key contributions include the integrated system architecture and a dual adaptive strategy within the UKF: dynamic tuning of the measurement noise covariance compensates for varying CNN measurement uncertainty, while adaptive tuning of the process noise covariance, utilizing measurement residual analysis, accounts for unmodeled dynamics or maneuvers online. This dual adaptation enhances robustness against both measurement imperfections and dynamic model uncertainties. The performance of the proposed adaptive integrated system is evaluated through high-fidelity simulations using a realistic ENVISAT model, comparing estimates against ground truth under various conditions, including measurement outages. This comprehensive approach offers an enhanced solution for robust onboard relative navigation, significantly advancing the capabilities required for safe proximity operations during ADR missions.
Authors: Xinwei Cao, Zijian Fan, Torbj{\o}rn Svendsen, Giampiero Salvi
Abstract: Mispronunciation detection and diagnosis (MDD) is a significant part in modern computer aided language learning (CALL) systems. Within MDD, phoneme-level pronunciation assessment is key to helping L2 learners improve their pronunciation. However, most systems are based on a form of goodness of pronunciation (GOP) which requires pre-segmentation of speech into phonetic units. This limits the accuracy of these methods and the possibility to use modern CTC-based acoustic models for their evaluation. In this study, we first propose self-alignment GOP (GOP-SA) that enables the use of CTC-trained ASR models for MDD. Next, we define a more general alignment-free method that takes all possible alignments of the target phoneme into account (GOP-AF). We give a theoretical account of our definition of GOP-AF, an implementation that solves potential numerical issues as well as a proper normalization which makes the method applicable with acoustic models with different peakiness over time. We provide extensive experimental results on the CMU Kids and Speechocean762 datasets comparing the different definitions of our methods, estimating the dependency of GOP-AF on the peakiness of the acoustic models and on the amount of context around the target phoneme. Finally, we compare our methods with recent studies over the Speechocean762 data showing that the feature vectors derived from the proposed method achieve state-of-the-art results on phoneme-level pronunciation assessment.
Authors: Luobin Cui, Yanlai Wu, Tang Ying, Weikai Li
Abstract: Fatigue detection plays a critical role in safety-critical applications such as aviation, mining, and long-haul transport. However, most existing methods rely on high-end sensors and controlled environments, limiting their applicability in real world settings. This paper formally defines a practical yet underexplored problem setting for real world fatigue detection, where systems operating with context-appropriate sensors aim to leverage knowledge from differently instrumented sources including those using impractical sensors deployed in controlled environments. To tackle this challenge, we propose a heterogeneous and multi-source fatigue detection framework that adaptively utilizes the available modalities in the target domain while benefiting from the diverse configurations present in source domains. Our experiments, conducted using a realistic field-deployed sensor setup and two publicly available datasets, demonstrate the practicality, robustness, and improved generalization of our approach, paving the practical way for effective fatigue monitoring in sensor-constrained scenarios.
Authors: Xiaoan Liu, Difan Jia, Xianhao Carton Liu, Mar Gonzalez-Franco, Chen Zhu-Tian
Abstract: Interacting with real-world objects in Mixed Reality (MR) often proves difficult when they are crowded, distant, or partially occluded, hindering straightforward selection and manipulation. We observe that these difficulties stem from performing interaction directly on physical objects, where input is tightly coupled to their physical constraints. Our key insight is to decouple interaction from these constraints by introducing proxies-abstract representations of real-world objects. We embody this concept in Reality Proxy, a system that seamlessly shifts interaction targets from physical objects to their proxies during selection. Beyond facilitating basic selection, Reality Proxy uses AI to enrich proxies with semantic attributes and hierarchical spatial relationships of their corresponding physical objects, enabling novel and previously cumbersome interactions in MR - such as skimming, attribute-based filtering, navigating nested groups, and complex multi object selections - all without requiring new gestures or menu systems. We demonstrate Reality Proxy's versatility across diverse scenarios, including office information retrieval, large-scale spatial navigation, and multi-drone control. An expert evaluation suggests the system's utility and usability, suggesting that proxy-based abstractions offer a powerful and generalizable interaction paradigm for future MR systems.
Authors: Zijie Guo, Jiong Wang, Xiaoyu Yue, Wangxu Wei, Zhe Jiang, Wanghan Xu, Ben Fei, Wenlong Zhang, Xinyu Gu, Lijing Cheng, Jing-Jia Luo, Chao Li, Yaqiang Wang, Tao Chen, Wanli Ouyang, Fenghua Ling, Lei Bai
Abstract: Modern Earth science is at an inflection point. The vast, fragmented, and complex nature of Earth system data, coupled with increasingly sophisticated analytical demands, creates a significant bottleneck for rapid scientific discovery. Here we introduce EarthLink, the first AI agent designed as an interactive copilot for Earth scientists. It automates the end-to-end research workflow, from planning and code generation to multi-scenario analysis. Unlike static diagnostic tools, EarthLink can learn from user interaction, continuously refining its capabilities through a dynamic feedback loop. We validated its performance on a number of core scientific tasks of climate change, ranging from model-observation comparisons to the diagnosis of complex phenomena. In a multi-expert evaluation, EarthLink produced scientifically sound analyses and demonstrated an analytical competency that was rated as comparable to specific aspects of a human junior researcher's workflow. Additionally, its transparent, auditable workflows and natural language interface empower scientists to shift from laborious manual execution to strategic oversight and hypothesis generation. EarthLink marks a pivotal step towards an efficient, trustworthy, and collaborative paradigm for Earth system research in an era of accelerating global change. The system is accessible at our website https://earthlink.intern-ai.org.cn.
Authors: Haotian Chen, Zhiyong Xiao
Abstract: In the field of food image processing, efficient semantic segmentation techniques are crucial for industrial applications. However, existing large-scale Transformer-based models (such as FoodSAM) face challenges in meeting practical deploymentrequirements due to their massive parameter counts and high computational resource demands. This paper introduces TUNable Adapter module (Swin-TUNA), a Parameter Efficient Fine-Tuning (PEFT) method that integrates multiscale trainable adapters into the Swin Transformer architecture, achieving high-performance food image segmentation by updating only 4% of the parameters. The core innovation of Swin-TUNA lies in its hierarchical feature adaptation mechanism: it designs separable convolutions in depth and dimensional mappings of varying scales to address the differences in features between shallow and deep networks, combined with a dynamic balancing strategy for tasks-agnostic and task-specific features. Experiments demonstrate that this method achieves mIoU of 50.56% and 74.94% on the FoodSeg103 and UECFoodPix Complete datasets, respectively, surpassing the fully parameterized FoodSAM model while reducing the parameter count by 98.7% (to only 8.13M). Furthermore, Swin-TUNA exhibits faster convergence and stronger generalization capabilities in low-data scenarios, providing an efficient solution for assembling lightweight food image.
Authors: Maciej K. Wozniak, Lianhang Liu, Yixi Cai, Patric Jensfelt
Abstract: While end-to-end autonomous driving models show promising results, their practical deployment is often hindered by large model sizes, a reliance on expensive LiDAR sensors and computationally intensive BEV feature representations. This limits their scalability, especially for mass-market vehicles equipped only with cameras. To address these challenges, we propose PRIX (Plan from Raw Pixels). Our novel and efficient end-to-end driving architecture operates using only camera data, without explicit BEV representation and forgoing the need for LiDAR. PRIX leverages a visual feature extractor coupled with a generative planning head to predict safe trajectories from raw pixel inputs directly. A core component of our architecture is the Context-aware Recalibration Transformer (CaRT), a novel module designed to effectively enhance multi-level visual features for more robust planning. We demonstrate through comprehensive experiments that PRIX achieves state-of-the-art performance on the NavSim and nuScenes benchmarks, matching the capabilities of larger, multimodal diffusion planners while being significantly more efficient in terms of inference speed and model size, making it a practical solution for real-world deployment. Our work is open-source and the code will be at https://maxiuw.github.io/prix.