Hamid Alinejad-Rokny

Tariq M. Khan, Syed Saud Naqvi, Thantrira Porntaveetus et al.

Medical image segmentation is often framed as a search for stronger architectures, but this can obscure a more fundamental question: what does the dataset require from the model? In medical imaging, this requirement is shaped by foreground occupancy, morphology, boundary ambiguity, topology sensitivity, annotation quality, acquisition variation, and operating point. This paper introduces the Medical Segmentation Dataset Knowledge Card (MS-DKC), a framework for making these factors explicit. MS-DKC records dataset evidence through image/acquisition, morphology, supervision, context-dependence, and deployment-risk descriptors. These descriptors are mapped to failure modes, design priors, and risk-aligned criteria, making segmentation design more traceable than architecture-first comparison. We evaluate MS-DKC on DRIVE, ISIC2018, and ACDC, representing distinct regimes. DRIVE contains sparse, thin, branching vessels, favoring detail-preserving models, sensitivity-aware optimization, threshold analysis, and topology-aware metrics. DKC-TNet-v2 achieved Dice 0.8044 and IoU 0.6730 with 35103 parameters, while SA-UNetv2-DKC-AmbRef reached Dice 0.8141, IoU 0.6865, sensitivity 0.8265, specificity 0.9804, and AUC 0.9853. ISIC2018 involves compact but appearance-variable lesions; validation-constrained score-function selection on Att-Next-Topo/ATTNext produced MS-DKC-AttNextTopo-VCSF-NoAug with Dice 0.8872, IoU 0.8214, precision 0.9173, Boundary F1 0.4878, and ASSD 4.13, while plausible additions failed to improve the risk-aligned profile. ACDC provides a multi-class cardiac case, where MS-DKC recommends four-class softmax segmentation, class-balanced Dice/CE supervision, and class-wise surface evaluation. Overall, the results support dataset-conditioned design: different datasets require different priors, operating points, and evidence before a model can be judged appropriate.

Guhong Chen, Liyang Fan, Zihan Gong et al.

Current research in LLM-based simulation systems lacks comprehensive solutions for modeling real-world court proceedings, while existing legal language models struggle with dynamic courtroom interactions. We present AgentCourt, a comprehensive legal simulation framework that addresses these challenges through adversarial evolution of LLM-based agents. Our AgentCourt introduces a new adversarial evolutionary approach for agents called AdvEvol, which performs dynamic knowledge learning and evolution through structured adversarial interactions in a simulated courtroom program, breaking the limitations of the traditional reliance on static knowledge bases or manual annotations. By simulating 1,000 civil cases, we construct an evolving knowledge base that enhances the agents' legal reasoning abilities. The evolved lawyer agents demonstrated outstanding performance on our newly introduced CourtBench benchmark, achieving a 12.1% improvement in performance compared to the original lawyer agents. Evaluations by professional lawyers confirm the effectiveness of our approach across three critical dimensions: cognitive agility, professional knowledge, and logical rigor. Beyond outperforming specialized legal models in interactive reasoning tasks, our findings emphasize the importance of adversarial learning in legal AI and suggest promising directions for extending simulation-based legal reasoning to broader judicial and regulatory contexts. The project's code is available at: https://github.com/relic-yuexi/AgentCourt

Mahboobeh Jafari, Delaram Sadeghi, Afshin Shoeibi et al.

Schizophrenia (SZ) is a prevalent mental disorder characterized by cognitive, emotional, and behavioral changes. Symptoms of SZ include hallucinations, illusions, delusions, lack of motivation, and difficulties in concentration. Diagnosing SZ involves employing various tools, including clinical interviews, physical examinations, psychological evaluations, the Diagnostic and Statistical Manual of Mental Disorders (DSM), and neuroimaging techniques. Electroencephalography (EEG) recording is a significant functional neuroimaging modality that provides valuable insights into brain function during SZ. However, EEG signal analysis poses challenges for neurologists and scientists due to the presence of artifacts, long-term recordings, and the utilization of multiple channels. To address these challenges, researchers have introduced artificial intelligence (AI) techniques, encompassing conventional machine learning (ML) and deep learning (DL) methods, to aid in SZ diagnosis. This study reviews papers focused on SZ diagnosis utilizing EEG signals and AI methods. The introduction section provides a comprehensive explanation of SZ diagnosis methods and intervention techniques. Subsequently, review papers in this field are discussed, followed by an introduction to the AI methods employed for SZ diagnosis and a summary of relevant papers presented in tabular form. Additionally, this study reports on the most significant challenges encountered in SZ diagnosis, as identified through a review of papers in this field. Future directions to overcome these challenges are also addressed. The discussion section examines the specific details of each paper, culminating in the presentation of conclusions and findings.

Afshar Shamsi, Rejisa Becirovic, Ahmadreza Argha et al.

Test time adaptation (TTA) equips deep learning models to handle unseen test data that deviates from the training distribution, even when source data is inaccessible. While traditional TTA methods often rely on entropy as a confidence metric, its effectiveness can be limited, particularly in biased scenarios. Extending existing approaches like the Pseudo Label Probability Difference (PLPD), we introduce ETAGE, a refined TTA method that integrates entropy minimization with gradient norms and PLPD, to enhance sample selection and adaptation. Our method prioritizes samples that are less likely to cause instability by combining high entropy with high gradient norms out of adaptation, thus avoiding the overfitting to noise often observed in previous methods. Extensive experiments on CIFAR-10-C and CIFAR-100-C datasets demonstrate that our approach outperforms existing TTA techniques, particularly in challenging and biased scenarios, leading to more robust and consistent model performance across diverse test scenarios. The codebase for ETAGE is available on https://github.com/afsharshamsi/ETAGE.

Jiahao Zhao, Luxin Xu, Minghuan Tan et al.

Numerous medical systems powered by Large Language Models (LLMs) have achieved remarkable progress in diverse healthcare tasks. However, research on their medication safety remains limited due to the lack of real world datasets, constrained by privacy and accessibility issues. Moreover, evaluation of LLMs in realistic clinical consultation settings, particularly regarding medication safety, is still underexplored. To address these gaps, we propose a framework that simulates and evaluates clinical consultations to systematically assess the medication safety capabilities of LLMs. Within this framework, we generate inquiry diagnosis dialogues with embedded medication risks and construct a dedicated medication safety database, RxRisk DB, containing 6,725 contraindications, 28,781 drug interactions, and 14,906 indication-drug pairs. A two-stage filtering strategy ensures clinical realism and professional quality, resulting in the benchmark RxSafeBench with 2,443 high-quality consultation scenarios. We evaluate leading open-source and proprietary LLMs using structured multiple choice questions that test their ability to recommend safe medications under simulated patient contexts. Results show that current LLMs struggle to integrate contraindication and interaction knowledge, especially when risks are implied rather than explicit. Our findings highlight key challenges in ensuring medication safety in LLM-based systems and provide insights into improving reliability through better prompting and task-specific tuning. RxSafeBench offers the first comprehensive benchmark for evaluating medication safety in LLMs, advancing safer and more trustworthy AI-driven clinical decision support.

Mohsen Karami, Khadijeh, Jahanian et al.

In recent years, Reinforcement Learning (RL) has emerged as a powerful tool for solving a wide range of problems, including decision-making and genomics. The exponential growth of raw genomic data over the past two decades has exceeded the capacity of manual analysis, leading to a growing interest in automatic data analysis and processing. RL algorithms are capable of learning from experience with minimal human supervision, making them well-suited for genomic data analysis and interpretation. One of the key benefits of using RL is the reduced cost associated with collecting labeled training data, which is required for supervised learning. While there have been numerous studies examining the applications of Machine Learning (ML) in genomics, this survey focuses exclusively on the use of RL in various genomics research fields, including gene regulatory networks (GRNs), genome assembly, and sequence alignment. We present a comprehensive technical overview of existing studies on the application of RL in genomics, highlighting the strengths and limitations of these approaches. We then discuss potential research directions that are worthy of future exploration, including the development of more sophisticated reward functions as RL heavily depends on the accuracy of the reward function, the integration of RL with other machine learning techniques, and the application of RL to new and emerging areas in genomics research. Finally, we present our findings and conclude by summarizing the current state of the field and the future outlook for RL in genomics.

Jingwei Zhu, Minghuan Tan, Min Yang et al.

The rapid progress in Large Language Models (LLMs) has prompted the creation of numerous benchmarks to evaluate their capabilities.This study focuses on the Comprehensive Medical Benchmark in Chinese (CMB), showcasing how dataset diversity and distribution in supervised fine-tuning (SFT) may enhance LLM performance.Remarkably, We successfully trained a smaller base model to achieve scores comparable to larger models, indicating that a diverse and well-distributed dataset can optimize performance regardless of model size.This study suggests that even smaller models may reach high performance levels with carefully curated and varied datasets. By integrating a wide range of instructional content, our approach addresses potential issues such as data quality inconsistencies. Our results imply that a broader spectrum of training data may enhance a model's ability to generalize and perform effectively across different medical scenarios, highlighting the importance of dataset quality and diversity in fine-tuning processes. We open-source the model for future research at https://github.com/CAS-SIAT-XinHai/CollectiveSFT

Bao Gia Doan, Afshar Shamsi, Xiao-Yu Guo et al.

Computational complexity of Bayesian learning is impeding its adoption in practical, large-scale tasks. Despite demonstrations of significant merits such as improved robustness and resilience to unseen or out-of-distribution inputs over their non- Bayesian counterparts, their practical use has faded to near insignificance. In this study, we introduce an innovative framework to mitigate the computational burden of Bayesian neural networks (BNNs). Our approach follows the principle of Bayesian techniques based on deep ensembles, but significantly reduces their cost via multiple low-rank perturbations of parameters arising from a pre-trained neural network. Both vanilla version of ensembles as well as more sophisticated schemes such as Bayesian learning with Stein Variational Gradient Descent (SVGD), previously deemed impractical for large models, can be seamlessly implemented within the proposed framework, called Bayesian Low-Rank LeArning (Bella). In a nutshell, i) Bella achieves a dramatic reduction in the number of trainable parameters required to approximate a Bayesian posterior; and ii) it not only maintains, but in some instances, surpasses the performance of conventional Bayesian learning methods and non-Bayesian baselines. Our results with large-scale tasks such as ImageNet, CAMELYON17, DomainNet, VQA with CLIP, LLaVA demonstrate the effectiveness and versatility of Bella in building highly scalable and practical Bayesian deep models for real-world applications.

Yuheng Liang, Lucy Chuo, Ahmadreza Argha et al.

Immune checkpoint inhibitors (ICIs) have transformed cancer therapy; yet substantial proportion of patients exhibit intrinsic or acquired resistance, making accurate pre-treatment response prediction a critical unmet need. Transcriptomics-based biomarkers derived from bulk and single-cell RNA sequencing (scRNA-seq) offer a promising avenue for capturing tumour-immune interactions, yet the cross-cohort generalisability of existing prediction models remains unclear.We systematically benchmark nine state-of-the-art transcriptomic ICI response predictors, five bulk RNA-seq-based models (COMPASS, IRNet, NetBio, IKCScore, and TNBC-ICI) and four scRNA-seq-based models (PRECISE, DeepGeneX, Tres and scCURE), using publicly available independent datasets unseen during model development. Overall, predictive performance was modest: bulk RNA-seq models performed at or near chance level across most cohorts, while scRNA-seq models showed only marginal improvements. Pathway-level analyses revealed sparse and inconsistent biomarker signals across models. Although scRNA-seq-based predictors converged on immune-related programs such as allograft rejection, bulk RNA-seq-based models exhibited little reproducible overlap. PRECISE and NetBio identified the most coherent immune-related themes, whereas IRNet predominantly captured metabolic pathways weakly aligned with ICI biology. Together, these findings demonstrate the limited cross-cohort robustness and biological consistency of current transcriptomic ICI prediction models, underscoring the need for improved domain adaptation, standardised preprocessing, and biologically grounded model design.

Mansooreh Montazerin, Elahe Rahimian, Farnoosh Naderkhani et al.

Development of advance surface Electromyogram (sEMG)-based Human-Machine Interface (HMI) systems is of paramount importance to pave the way towards emergence of futuristic Cyber-Physical-Human (CPH) worlds. In this context, the main focus of recent literature was on development of different Deep Neural Network (DNN)-based architectures that perform Hand Gesture Recognition (HGR) at a macroscopic level (i.e., directly from sEMG signals). At the same time, advancements in acquisition of High-Density sEMG signals (HD-sEMG) have resulted in a surge of significant interest on sEMG decomposition techniques to extract microscopic neural drive information. However, due to complexities of sEMG decomposition and added computational overhead, HGR at microscopic level is less explored than its aforementioned DNN-based counterparts. In this regard, we propose the HYDRA-HGR framework, which is a hybrid model that simultaneously extracts a set of temporal and spatial features through its two independent Vision Transformer (ViT)-based parallel architectures (the so called Macro and Micro paths). The Macro Path is trained directly on the pre-processed HD-sEMG signals, while the Micro path is fed with the p-to-p values of the extracted Motor Unit Action Potentials (MUAPs) of each source. Extracted features at macroscopic and microscopic levels are then coupled via a Fully Connected (FC) fusion layer. We evaluate the proposed hybrid HYDRA-HGR framework through a recently released HD-sEMG dataset, and show that it significantly outperforms its stand-alone counterparts. The proposed HYDRA-HGR framework achieves average accuracy of 94.86% for the 250 ms window size, which is 5.52% and 8.22% higher than that of the Macro and Micro paths, respectively.

Qiyao Wang, Xinyi Chen, Longze Chen et al.

Patent examination is a complex, multi-stage process requiring both technical expertise and legal reasoning, increasingly challenged by rising application volumes. Prior benchmarks predominantly view patent examination as discriminative classification or static extraction, failing to capture its inherently interactive and iterative nature, similar to the peer review and rebuttal process in academic publishing. In this paper, we introduce PatRe, the first benchmark that models the full patent examination lifecycle, including Office Action generation and applicant rebuttal. PatRe comprises 480 real-world cases and supports both oracle and retrieval-simulated evaluation settings. Our benchmark reframes patent examination as a dynamic, multi-turn process of justification and response. Extensive experiments across various LLMs reveal critical insights into model performance, including differences between proprietary and open-source models, as well as task asymmetries between examiner analysis and applicant-side rebuttal. These findings highlight both the potential and current limitations of LLMs in modeling complex, real-world legal reasoning and technical novelty judgment in patent examination. We release our code and dataset to facilitate future research on patent examination modeling.

Guhong Chen, Chenghao Sun, Cheng Fu et al.

As code large language models (LLMs) evolve into tool-interactive agents via the Model Context Protocol (MCP), their generalization is increasingly limited by low-quality synthetic data and the diminishing returns of quantity scaling. Moreover, quantity-centric scaling exhibits an early bottleneck that underutilizes trajectory data. We propose TDScaling, a Trajectory Diversity Scaling-based data synthesis framework for code agents that scales performance through diversity rather than raw volume. Under a fixed training budget, increasing trajectory diversity yields larger gains than adding more trajectories, improving the performance-cost trade-off for agent training. TDScaling integrates four innovations: (1) a Business Cluster mechanism that captures real-service logical dependencies; (2) a blueprint-driven multi-agent paradigm that enforces trajectory coherence; (3) an adaptive evolution mechanism that steers synthesis toward long-tail scenarios using Domain Entropy, Reasoning Mode Entropy, and Cumulative Action Complexity to prevent mode collapse; and (4) a sandboxed code tool that mitigates catastrophic forgetting of intrinsic coding capabilities. Experiments on general tool-use benchmarks (BFCL, tau^2-Bench) and code agent tasks (RebenchT, CodeCI, BIRD) demonstrate a win-win outcome: TDScaling improves both tool-use generalization and inherent coding proficiency. We plan to release the full codebase and the synthesized dataset (including 30,000+ tool clusters) upon publication.

Dingwei Chen, Feiteng Fang, Shiwen Ni et al.

Large Language Models (LLMs) have demonstrated exceptional performance across various natural language processing tasks. However, they occasionally generate inaccurate and counterfactual outputs, a phenomenon commonly referred to as "hallucinations''. To tackle this issue, recent studies have explored contrastive decoding between the original model and an amateur model with induced hallucination, showing promising results. Nevertheless, this approach can disrupt the original LLM's output distribution due to coarse contrast and simple subtraction operations, potentially leading to errors. In this paper, we introduce a novel contrastive decoding framework, termed LOL (LOwer Layer Matters). Unlike prior methods that focus solely on the final layer, our approach integrates contrastive information from lower layers to enable multi-layer fusion during contrastive decoding. Additionally, we incorporate a truthfulness refocused module that leverages instruction guidance to further improve truthfulness in contrastive decoding. Extensive experiments on four publicly available datasets demonstrate that the LOL framework significantly mitigates hallucination while outperforming existing baselines in most cases. For reproducibility, we will release our code and data upon acceptance.

Qiyao Wang, Shiwen Ni, Huaren Liu et al.

As the capabilities of Large Language Models (LLMs) continue to advance, the field of patent processing has garnered increased attention within the natural language processing community. However, the majority of research has been concentrated on classification tasks, such as patent categorization and examination, or on short text generation tasks like patent summarization and patent quizzes. In this paper, we introduce a novel and practical task known as Draft2Patent, along with its corresponding D2P benchmark, which challenges LLMs to generate full-length patents averaging 17K tokens based on initial drafts. Patents present a significant challenge to LLMs due to their specialized nature, standardized terminology, and extensive length. We propose a multi-agent framework called AutoPatent which leverages the LLM-based planner agent, writer agents, and examiner agent with PGTree and RRAG to generate lengthy, intricate, and high-quality complete patent documents. The experimental results demonstrate that our AutoPatent framework significantly enhances the ability to generate comprehensive patents across various LLMs. Furthermore, we have discovered that patents generated solely with the AutoPatent framework based on the Qwen2.5-7B model outperform those produced by larger and more powerful LLMs, such as GPT-4o, Qwen2.5-72B, and LLAMA3.1-70B, in both objective metrics and human evaluations. We will make the data and code available upon acceptance at \url{https://github.com/QiYao-Wang/AutoPatent}.

Renhao Li, Jianhong Tu, Yang Su et al.

Reward models (RMs) play a critical role in aligning large language models (LLMs) with human preferences. Yet in the domain of tool learning, the lack of RMs specifically designed for function-calling tasks has limited progress toward more capable agentic AI. We introduce ToolRM, a family of lightweight generative RMs tailored for general tool-use scenarios. To build these models, we propose a novel pipeline that constructs pairwise preference data using rule-based scoring and multidimensional sampling. This yields ToolPref-Pairwise-30K, a diverse, balanced, and challenging dataset of critique tasks that supports reinforcement learning with verifiable feedback. To evaluate tool-use RMs, we also introduce TRBench$_{BFCL}$, a benchmark built on the agentic evaluation suite BFCL. Trained on our constructed data, models from the Qwen3-4B/8B series achieve up to 14.28% higher accuracy, substantially outperforming frontier models such as Claude 4 and OpenAI o3 in pairwise reward judgments. Beyond training objectives, ToolRM generalizes to broader critique tasks, including Best-of-N sampling and self-correction. Experiments on ACEBench highlight its effectiveness and efficiency, enabling inference-time scaling and reducing output token usage by over 66%. We release data and model checkpoints to facilitate future research.

Sunbowen Lee, Shiwen Ni, Chi Wei et al.

Safety alignment mechanism are essential for preventing large language models (LLMs) from generating harmful information or unethical content. However, cleverly crafted prompts can bypass these safety measures without accessing the model's internal parameters, a phenomenon known as black-box jailbreak. Existing heuristic black-box attack methods, such as genetic algorithms, suffer from limited effectiveness due to their inherent randomness, while recent reinforcement learning (RL) based methods often lack robust and informative reward signals. To address these challenges, we propose a novel black-box jailbreak method leveraging RL, which optimizes prompt generation by analyzing the embedding proximity between benign and malicious prompts. This approach ensures that the rewritten prompts closely align with the intent of the original prompts while enhancing the attack's effectiveness. Furthermore, we introduce a comprehensive jailbreak evaluation framework incorporating keywords, intent matching, and answer validation to provide a more rigorous and holistic assessment of jailbreak success. Experimental results show the superiority of our approach, achieving state-of-the-art (SOTA) performance on several prominent open and closed-source LLMs, including Qwen2.5-7B-Instruct, Llama3.1-8B-Instruct, and GPT-4o-0806. Our method sets a new benchmark in jailbreak attack effectiveness, highlighting potential vulnerabilities in LLMs. The codebase for this work is available at https://github.com/Aegis1863/xJailbreak.

Jiahao Zhao, Feng Jiang, Shaowei Qin et al.

Large language models (LLMs) are increasingly applied in scientific research, offering new capabilities for knowledge discovery and reasoning. In single-cell biology, however, evaluation practices for both general and specialized LLMs remain inadequate: existing benchmarks are fragmented across tasks, adopt formats such as multiple-choice classification that diverge from real-world usage, and rely on metrics lacking interpretability and biological grounding. We present SC-ARENA, a natural language evaluation framework tailored to single-cell foundation models. SC-ARENA formalizes a virtual cell abstraction that unifies evaluation targets by representing both intrinsic attributes and gene-level interactions. Within this paradigm, we define five natural language tasks (cell type annotation, captioning, generation, perturbation prediction, and scientific QA) that probe core reasoning capabilities in cellular biology. To overcome the limitations of brittle string-matching metrics, we introduce knowledge-augmented evaluation, which incorporates external ontologies, marker databases, and scientific literature to support biologically faithful and interpretable judgments. Experiments and analysis across both general-purpose and domain-specialized LLMs demonstrate that (i) under the Virtual Cell unified evaluation paradigm, current models achieve uneven performance on biologically complex tasks, particularly those demanding mechanistic or causal understanding; and (ii) our knowledge-augmented evaluation framework ensures biological correctness, provides interpretable, evidence-grounded rationales, and achieves high discriminative capacity, overcoming the brittleness and opacity of conventional metrics. SC-Arena thus provides a unified and interpretable framework for assessing LLMs in single-cell biology, pointing toward the development of biology-aligned, generalizable foundation models.

Sunbowen Lee, Junting Zhou, Chang Ao et al.

Model distillation is a fundamental technique in building large language models (LLMs), transferring knowledge from a teacher model to a student model. However, distillation can lead to model homogenization, reducing diversity among models and impairing their ability to robustly handle complex or novel tasks. These limitations underscore the need to systematically quantify the distillation process and its impact. In this work, we propose a framework to evaluate and quantify model distillation. Our method addresses two key aspects: (1) Identifying identity cognition contradictions to assess discrepancies in how models perceive and represent identity-related information, and (2) Analyzing multi-granularity response similarities across models to measure the extent of homogenization. Experimental results demonstrate two key insights: (1) Well-known closed-source and open-source LLMs usually exhibit high distillation degrees, except for Claude, Doubao, and Gemini. (2) Base LLMs show higher distillation degrees compared to aligned LLMs. By offering a systematic approach to improve the transparency of LLM data distillation, we call for LLMs with more independent development and more transparent technical reports to improve LLMs' robustness and safety. The code and data are available under https://github.com/Aegis1863/LLMs-Distillation-Quantification.

Shuaimin Li, Liyang Fan, Yufang Lin et al.

Existing paper review methods often rely on superficial manuscript features or directly on large language models (LLMs), which are prone to hallucinations, biased scoring, and limited reasoning capabilities. Moreover, these methods often fail to capture the complex argumentative reasoning and negotiation dynamics inherent in reviewer-author interactions. To address these limitations, we propose ReViewGraph (Reviewer-Author Debates Graph Reasoner), a novel framework that performs heterogeneous graph reasoning over LLM-simulated multi-round reviewer-author debates. In our approach, reviewer-author exchanges are simulated through LLM-based multi-agent collaboration. Diverse opinion relations (e.g., acceptance, rejection, clarification, and compromise) are then explicitly extracted and encoded as typed edges within a heterogeneous interaction graph. By applying graph neural networks to reason over these structured debate graphs, ReViewGraph captures fine-grained argumentative dynamics and enables more informed review decisions. Extensive experiments on three datasets demonstrate that ReViewGraph outperforms strong baselines with an average relative improvement of 15.73%, underscoring the value of modeling detailed reviewer-author debate structures.

Yukun Chen, Jiaming Li, Longze Chen et al.

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a prevailing paradigm for enhancing reasoning in Multimodal Large Language Models (MLLMs). However, relying solely on outcome supervision risks reward hacking, where models learn spurious reasoning patterns to satisfy final answer checks. While recent rubric-based approaches offer fine-grained supervision signals, they suffer from high computational costs of instance-level generation and inefficient training dynamics caused by treating all rubrics as equally learnable. In this paper, we propose Stratified Rubric-based Curriculum Learning (RuCL), a novel framework that reformulates curriculum learning by shifting the focus from data selection to reward design. RuCL generates generalized rubrics for broad applicability and stratifies them based on the model's competence. By dynamically adjusting rubric weights during training, RuCL guides the model from mastering foundational perception to tackling advanced logical reasoning. Extensive experiments on various visual reasoning benchmarks show that RuCL yields a remarkable +7.83% average improvement over the Qwen2.5-VL-7B model, achieving a state-of-the-art accuracy of 60.06%.

Run Luo, Ting-En Lin, Haonan Zhang et al.

Recent advancements in omnimodal learning have significantly improved understanding and generation across images, text, and speech, yet these developments remain predominantly confined to proprietary models. The lack of high-quality omnimodal datasets and the challenges of real-time emotional speech synthesis have notably hindered progress in open-source research. To address these limitations, we introduce \name, a two-stage training framework that integrates omnimodal alignment and speech generation to develop a state-of-the-art omnimodal large language model. In the alignment phase, a pre-trained speech model undergoes further training on text-image tasks, enabling (near) zero-shot generalization from vision to speech, outperforming models trained on tri-modal datasets. In the speech generation phase, a lightweight decoder is trained on speech tasks with direct preference optimization, enabling real-time emotional speech synthesis with high fidelity. Experiments show that \name surpasses state-of-the-art models across omnimodal, vision-language, and speech-language benchmarks. It achieves a 4-point absolute improvement on OmniBench over the leading open-source model VITA, despite using 5x fewer training samples and a smaller model size (7B vs. 7x8B). Additionally, \name achieves real-time speech generation with <1s latency at non-autoregressive mode, reducing inference time by 5x compared to autoregressive methods, and improves emotion classification accuracy by 7.7\%

Liang Zhu, Yuelin Bai, Xiankun Ren et al.

Preference learning in Large Language Models (LLMs) has advanced significantly, yet existing methods remain limited by modest performance gains, high computational costs, hyperparameter sensitivity, and insufficient modeling of global token-level relationships. We introduce PLOT, which enhances Preference Learning in fine-tuning-based alignment through a token-level loss derived from Optimal Transport. By formulating preference learning as an Optimal Transport Problem, PLOT aligns model outputs with human preferences while preserving the original distribution of LLMs, ensuring stability and robustness. Furthermore, PLOT leverages token embeddings to capture semantic relationships, enabling globally informed optimization. Experiments across two preference categories - Human Values and Logic & Problem Solving - spanning seven subpreferences demonstrate that PLOT consistently improves alignment performance while maintaining fluency and coherence. These results substantiate optimal transport as a principled methodology for preference learning, establishing a theoretically grounded framework that provides new insights for preference learning of LLMs.

Dingwei Chen, Ziqiang Liu, Feiteng Fang et al.

Large Language Models (LLMs) demonstrate remarkable capabilities in text understanding and generation. However, their tendency to produce factually inconsistent outputs, commonly referred to as ''hallucinations'', remains a critical challenge. Existing approaches, such as retrieval-based and inference-time correction methods, primarily address this issue at the input or output level, often overlooking the intrinsic information refinement process and the role of premature layers. Meanwhile, alignment- and fine-tuning-based methods are resource-intensive. In this paper, we propose PLI (Premature Layers Interpolation), a novel, training-free, and plug-and-play intervention designed to enhance factuality. PLI mitigates hallucinations by inserting premature layers formed through mathematical interpolation with adjacent layers. Inspired by stable diffusion and sampling steps, PLI extends the depth of information processing and transmission in LLMs, improving factual coherence. Experiments on four publicly available datasets demonstrate that PLI effectively reduces hallucinations while outperforming existing baselines in most cases. Further analysis suggests that the success of layer interpolation is closely linked to LLMs' internal mechanisms. To promote reproducibility, we will release our code and data upon acceptance.

Feiteng Fang, Ting-En Lin, Yuchuan Wu et al.

Role-Playing Language Agents (RPLAs) aim to simulate characters for realistic and engaging human-computer interactions. However, traditional reward models often struggle with scalability and adapting to subjective conversational preferences. We propose ChARM, a Character-based Act-adaptive Reward Model, addressing these challenges through two innovations: (1) an act-adaptive margin that significantly enhances learning efficiency and generalizability, and (2) a self-evolution mechanism leveraging large-scale unlabeled data to improve training coverage. Additionally, we introduce RoleplayPref, the first large-scale preference dataset specifically for RPLAs, featuring 1,108 characters, 13 subcategories, and 16,888 bilingual dialogues, alongside RoleplayEval, a dedicated evaluation benchmark. Experimental results show a 13% improvement over the conventional Bradley-Terry model in preference rankings. Furthermore, applying ChARM-generated rewards to preference learning techniques (e.g., direct preference optimization) achieves state-of-the-art results on CharacterEval and RoleplayEval. Code and dataset are available at https://github.com/calubkk/ChARM.

Ancheng Xu, Zhihao Yang, Jingpeng Li et al.

E-commerce platforms increasingly rely on Large Language Models (LLMs) and Vision-Language Models (VLMs) to detect illicit or misleading product content. However, these models remain vulnerable to evasive content: inputs (text or images) that superficially comply with platform policies while covertly conveying prohibited claims. Unlike traditional adversarial attacks that induce overt failures, evasive content exploits ambiguity and context, making it far harder to detect. Existing robustness benchmarks provide little guidance for this demanding, real-world challenge. We introduce EVADE, the first expert-curated, Chinese, multimodal benchmark specifically designed to evaluate foundation models on evasive content detection in e-commerce. The dataset contains 2,833 annotated text samples and 13,961 images spanning six demanding product categories, including body shaping, height growth, and health supplements. Two complementary tasks assess distinct capabilities: Single-Violation, which probes fine-grained reasoning under short prompts, and All-in-One, which tests long-context reasoning by merging overlapping policy rules into unified instructions. Notably, the All-in-One setting significantly narrows the performance gap between partial and full-match accuracy, suggesting that clearer rule definitions improve alignment between human and model judgment. We benchmark 26 mainstream LLMs and VLMs and observe substantial performance gaps: even state-of-the-art models frequently misclassify evasive samples. By releasing EVADE and strong baselines, we provide the first rigorous standard for evaluating evasive-content detection, expose fundamental limitations in current multimodal reasoning, and lay the groundwork for safer and more transparent content moderation systems in e-commerce. The dataset is publicly available at https://huggingface.co/datasets/koenshen/EVADE-Bench.

Jiaming Li, Haoran Ye, Yukun Chen et al.

As large language models (LLMs) grow in scale and capability, understanding their internal mechanisms becomes increasingly critical. Sparse autoencoders (SAEs) have emerged as a key tool in mechanistic interpretability, enabling the extraction of human-interpretable features from LLMs. However, existing SAE training methods are primarily designed for base models, resulting in reduced reconstruction quality and interpretability when applied to instruct models. To bridge this gap, we propose $\underline{\textbf{F}}$inetuning-$\underline{\textbf{a}}$ligned $\underline{\textbf{S}}$equential $\underline{\textbf{T}}$raining ($\textit{FAST}$), a novel training method specifically tailored for instruct models. $\textit{FAST}$ aligns the training process with the data distribution and activation patterns characteristic of instruct models, resulting in substantial improvements in both reconstruction and feature interpretability. On Qwen2.5-7B-Instruct, $\textit{FAST}$ achieves a mean squared error of 0.6468 in token reconstruction, significantly outperforming baseline methods with errors of 5.1985 and 1.5096. In feature interpretability, $\textit{FAST}$ yields a higher proportion of high-quality features, for Llama3.2-3B-Instruct, $21.1\%$ scored in the top range, compared to $7.0\%$ and $10.2\%$ for $\textit{BT(P)}$ and $\textit{BT(F)}$. Surprisingly, we discover that intervening on the activations of special tokens via the SAEs leads to improvements in output quality, suggesting new opportunities for fine-grained control of model behavior. Code, data, and 240 trained SAEs are available at https://github.com/Geaming2002/FAST.

Qiyao Wang, Guhong Chen, Hongbo Wang et al.

Intellectual Property (IP) is a highly specialized domain that integrates technical and legal knowledge, making it inherently complex and knowledge-intensive. Recent advancements in LLMs have demonstrated their potential to handle IP-related tasks, enabling more efficient analysis, understanding, and generation of IP-related content. However, existing datasets and benchmarks focus narrowly on patents or cover limited aspects of the IP field, lacking alignment with real-world scenarios. To bridge this gap, we introduce IPBench, the first comprehensive IP task taxonomy and a large-scale bilingual benchmark encompassing 8 IP mechanisms and 20 distinct tasks, designed to evaluate LLMs in real-world IP scenarios. We benchmark 17 main LLMs, ranging from general purpose to domain-specific, including chat-oriented and reasoning-focused models, under zero-shot, few-shot, and chain-of-thought settings. Our results show that even the top-performing model, DeepSeek-V3, achieves only 75.8% accuracy, indicating significant room for improvement. Notably, open-source IP and law-oriented models lag behind closed-source general-purpose models. To foster future research, we publicly release IPBench, and will expand it with additional tasks to better reflect real-world complexities and support model advancements in the IP domain. We provide the data and code in the supplementary URLs.

Qiyao Wang, Haoran Hu, Longze Chen et al.

With the advancement of multimodal large language models (MLLMs) and coding agents, the website development has shifted from manual programming to agent-based project-level code synthesis. Existing benchmarks rely on idealized assumptions, especially for well-structured, information-rich inputs and static execution settings. In contrast, real-world development is constrained by a critical bottleneck: the semantic misalignment between ambiguous, low-quality instructions from non-expert users and model understanding, which results in a failure mode that we term blind execution. To address this gap, we introduce InteractWeb-Bench, the first multimodal interactive benchmark for website generation under non-expert low-code user conditions. InteractWeb-Bench introduces four types of user agents and persona-driven instruction perturbations to systematically simulate diverse user behaviors, including ambiguity, redundancy, and contradiction, grounded in requirement engineering defect taxonomies. We develop an interactive execution environment for agents, featuring a unified action space comprising Clarify, Implement, Verify, and Submit, enabling iterative intent refinement, code synthesis, and visual feedback-based validation. Extensive experiments and analysis reveal that frontier MLLM-based agents remain trapped in blind execution, exposing limitations in intent recognition and adaptive interaction.

Qiyao Wang, Hongbo Wang, Longze Chen et al.

Scientific idea generation (SIG) is critical to AI-driven autonomous research, yet existing approaches are often constrained by a static retrieval-then-generation paradigm, leading to homogeneous and insufficiently divergent ideas. In this work, we propose FlowPIE, a tightly coupled retrieval-generation framework that treats literature exploration and idea generation as a co-evolving process. FlowPIE expands literature trajectories via a flow-guided Monte Carlo Tree Search (MCTS) inspired by GFlowNets, using the quality of current ideas assessed by an LLM-based generative reward model (GRM) as a supervised signal to guide adaptive retrieval and construct a diverse, high-quality initial population. Based on this population, FlowPIE models idea generation as a test-time idea evolution process, applying selection, crossover, and mutation with the isolation island paradigm and GRM-based fitness computation to incorporate cross-domain knowledge. It effectively mitigates the information cocoons arising from over-reliance on parametric knowledge and static literature. Extensive evaluations demonstrate that FlowPIE consistently produces ideas with higher novelty, feasibility and diversity compared to strong LLM-based and agent-based frameworks, while enabling reward scaling during test time.

Jiaming Li, Yukun Chen, Ziqiang Liu et al.

Stories are central to human culture, serving to share ideas, preserve traditions, and foster connections. Automatic story generation, a key advancement in artificial intelligence (AI), offers new possibilities for creating personalized content, exploring creative ideas, and enhancing interactive experiences. However, existing methods struggle to maintain narrative coherence and logical consistency. This disconnect compromises the overall storytelling experience, underscoring the need for substantial improvements. Inspired by human cognitive processes, we introduce Storyteller, a novel approach that systemically improves the coherence and consistency of automatically generated stories. Storyteller introduces a plot node structure based on linguistically grounded subject verb object (SVO) triplets, which capture essential story events and ensure a consistent logical flow. Unlike previous methods, Storyteller integrates two dynamic modules, the STORYLINE and narrative entity knowledge graph (NEKG),that continuously interact with the story generation process. This integration produces structurally sound, cohesive and immersive narratives. Extensive experiments demonstrate that Storyteller significantly outperforms existing approaches, achieving an 84.33% average win rate through human preference evaluation. At the same time, it is also far ahead in other aspects including creativity, coherence, engagement, and relevance.

Longze Chen, Renke Shan, Huiming Wang et al.

Speculative decoding (SD) is a promising method for accelerating the decoding process of Large Language Models (LLMs). The efficiency of SD primarily hinges on the consistency between the draft model and the verify model. However, existing drafting approaches typically require additional modules to be trained, which can be challenging to implement and ensure compatibility across various LLMs. In this paper, we propose CLaSp, an in-context layer-skipping strategy for self-speculative decoding. Unlike prior methods, CLaSp does not require additional drafting modules or extra training. Instead, it employs a plug-and-play mechanism by skipping intermediate layers of the verify model to construct a compressed draft model. Specifically, we develop a dynamic programming algorithm that optimizes the layer-skipping process by leveraging the complete hidden states from the last verification stage as an objective. This enables CLaSp to dynamically adjust its layer-skipping strategy after each verification stage, without relying on pre-optimized sets of skipped layers. Experimental results across diverse downstream tasks demonstrate that CLaSp achieves a speedup of 1.3x ~ 1.7x on LLaMA3 series models without altering the original distribution of the generated text.

Shiwen Ni, Haihong Wu, Di Yang et al.

The quality of instruction data directly affects the performance of fine-tuned Large Language Models (LLMs). Previously, \cite{li2023one} proposed \texttt{NUGGETS}, which identifies and selects high-quality quality data from a large dataset by identifying those individual instruction examples that can significantly improve the performance of different tasks after being learnt as one-shot instances. In this work, we propose \texttt{SuperNUGGETS}, an improved variant of \texttt{NUGGETS} optimised for efficiency and performance. Our \texttt{SuperNUGGETS} uses a small language model (SLM) instead of a large language model (LLM) to filter the data for outstanding one-shot instances and refines the predefined set of tests. The experimental results show that the performance of \texttt{SuperNUGGETS} only decreases by 1-2% compared to \texttt{NUGGETS}, but the efficiency can be increased by a factor of 58. Compared to the original \texttt{NUGGETS}, our \texttt{SuperNUGGETS} has a higher utility value due to the significantly lower resource consumption.

Siyi Li, Jiajun Shi, Shiwen Ni et al.

Large Reasoning Models (LRMs) have demonstrated strong performance by producing extended Chain-of-Thought (CoT) traces before answering. However, this paradigm often induces over-reasoning: redundant calculations and circular self-verification that increase computational cost without improving outcomes. Existing evaluations largely emphasize final accuracy or coarse token counts, and lack automated tools to separate essential logic from structural redundancy. We introduce CoTJudger, a graph-driven framework that quantifies reasoning efficiency by converting free-form CoTs into directed dependency graphs and extracting the Shortest Effective Path (SEP) needed to reach a correct solution. This yields an interpretable efficiency signal -- how much of a CoT is necessary versus structurally redundant -- that is comparable across models and tasks. Evaluating 21 LRMs, CoTJudger reveals pervasive redundancy and surfaces recurring failure modes, including verification obsession and compensatory redundancy. These results provide a practical metric for disentangling reasoning ability from computational waste, enabling more targeted evaluation and diagnosis of LRM efficiency.

Hamzeh Asgharnezhad, Afshar Shamsi, Roohallah Alizadehsani et al.

Knowing the uncertainty associated with the output of a deep neural network is of paramount importance in making trustworthy decisions, particularly in high-stakes fields like medical diagnosis and autonomous systems. Monte Carlo Dropout (MCD) is a widely used method for uncertainty quantification, as it can be easily integrated into various deep architectures. However, conventional MCD often struggles with providing well-calibrated uncertainty estimates. To address this, we introduce innovative frameworks that enhances MCD by integrating different search solutions namely Grey Wolf Optimizer (GWO), Bayesian Optimization (BO), and Particle Swarm Optimization (PSO) as well as an uncertainty-aware loss function, thereby improving the reliability of uncertainty quantification. We conduct comprehensive experiments using different backbones, namely DenseNet121, ResNet50, and VGG16, on various datasets, including Cats vs. Dogs, Myocarditis, Wisconsin, and a synthetic dataset (Circles). Our proposed algorithm outperforms the MCD baseline by 2-3% on average in terms of both conventional accuracy and uncertainty accuracy while achieving significantly better calibration. These results highlight the potential of our approach to enhance the trustworthiness of deep learning models in safety-critical applications.

Zhihao Yang, Ancheng Xu, Jingpeng Li et al.

Large language models (LLMs) face significant challenges when processing complex rule systems, as they typically treat interdependent rules as unstructured textual data rather than as logically organized frameworks. This limitation results in reasoning divergence, where models often overlook critical rule dependencies essential for accurate interpretation. Although existing approaches such as Chain-of-Thought (CoT) reasoning have shown promise, they lack systematic methodologies for structured rule processing and are particularly susceptible to error propagation through sequential reasoning chains. To address these limitations, we propose the Dynamic Adjudication Template (DAT), a novel framework inspired by expert human reasoning processes. DAT structures the inference mechanism into three methodical stages: qualitative analysis, evidence gathering, and adjudication. During the qualitative analysis phase, the model comprehensively evaluates the contextual landscape. The subsequent evidence gathering phase involves the targeted extraction of pertinent information based on predefined template elements ([placeholder]), followed by systematic verification against applicable rules. Finally, in the adjudication phase, the model synthesizes these validated components to formulate a comprehensive judgment. Empirical results demonstrate that DAT consistently outperforms conventional CoT approaches in complex rule-based tasks. Notably, DAT enables smaller language models to match, and in some cases exceed, the performance of significantly larger LLMs, highlighting its efficiency and effectiveness in managing intricate rule systems.

Roxana Zahedi, Ahmadreza Argha, Nona Farbehi et al.

Spatial transcriptomics (ST) technologies enable gene expression profiling with spatial resolution, offering unprecedented insights into tissue organization and disease heterogeneity. However, current analysis methods often struggle with noisy data, limited scalability, and inadequate modelling of complex cellular relationships. We present SemanticST, a biologically informed, graph-based deep learning framework that models diverse cellular contexts through multi-semantic graph construction. SemanticST builds multiple context-specific graphs capturing spatial proximity, gene expression similarity, and tissue domain structure, and learns disentangled embeddings for each. These are fused using an attention-inspired strategy to yield a unified, biologically meaningful representation. A community-aware min-cut loss improves robustness over contrastive learning, particularly in sparse ST data. SemanticST supports mini-batch training, making it the first graph neural network scalable to large-scale datasets such as Xenium (500,000 cells). Benchmarking across four platforms (Visium, Slide-seq, Stereo-seq, Xenium) and multiple human and mouse tissues shows consistent 20 percentage gains in ARI, NMI, and trajectory fidelity over DeepST, GraphST, and IRIS. In re-analysis of breast cancer Xenium data, SemanticST revealed rare and clinically significant niches, including triple receptor-positive clusters, spatially distinct DCIS-to-IDC transition zones, and FOXC2 tumour-associated myoepithelial cells, suggesting non-canonical EMT programs with stem-like features. SemanticST thus provides a scalable, interpretable, and biologically grounded framework for spatial transcriptomics analysis, enabling robust discovery across tissue types and diseases, and paving the way for spatially resolved tissue atlases and next-generation precision medicine.

Alireza Sadeghi, Farshid Hajati, Ahmadreza Argha et al.

Integrating heterogeneous biomedical data including imaging, omics, and clinical records supports accurate diagnosis and personalised care. Graph-based models fuse such non-Euclidean data by capturing spatial and relational structure, yet clinical uptake requires regulator-ready interpretability. We present the first technical survey of interpretable graph based models for multimodal biomedical data, covering 26 studies published between Jan 2019 and Sep 2024. Most target disease classification, notably cancer and rely on static graphs from simple similarity measures, while graph-native explainers are rare; post-hoc methods adapted from non-graph domains such as gradient saliency, and SHAP predominate. We group existing approaches into four interpretability families, outline trends such as graph-in-graph hierarchies, knowledge-graph edges, and dynamic topology learning, and perform a practical benchmark. Using an Alzheimer disease cohort, we compare Sensitivity Analysis, Gradient Saliency, SHAP and Graph Masking. SHAP and Sensitivity Analysis recover the broadest set of known AD pathways and Gene-Ontology terms, whereas Gradient Saliency and Graph Masking surface complementary metabolic and transport signatures. Permutation tests show all four beat random gene sets, but with distinct trade-offs: SHAP and Graph Masking offer deeper biology at higher compute cost, while Gradient Saliency and Sensitivity Analysis are quicker though coarser. We also provide a step-by-step flowchart covering graph construction, explainer choice and resource budgeting to help researchers balance transparency and performance. This review synthesises the state of interpretable graph learning for multimodal medicine, benchmarks leading techniques, and charts future directions, from advanced XAI tools to under-studied diseases, serving as a concise reference for method developers and translational scientists.

Amir Masoud Rahmani, Parisa Khoshvaght, Hamid Alinejad-Rokny et al.

Acute lymphoblastic leukemia (ALL) severity is determined by the presence and ratios of blast cells (abnormal white blood cells) in both bone marrow and peripheral blood. Manual diagnosis of this disease is a tedious and time-consuming operation, making it difficult for professionals to accurately examine blast cell characteristics. To address this difficulty, researchers use deep learning and machine learning. In this paper, a ResNet-based feature extractor is utilized to detect ALL, along with a variety of feature selectors and classifiers. To get the best results, a variety of transfer learning models, including the Resnet, VGG, EfficientNet, and DensNet families, are used as deep feature extractors. Following extraction, different feature selectors are used, including Genetic algorithm, PCA, ANOVA, Random Forest, Univariate, Mutual information, Lasso, XGB, Variance, and Binary ant colony. After feature qualification, a variety of classifiers are used, with MLP outperforming the others. The recommended technique is used to categorize ALL and HEM in the selected dataset which is C-NMC 2019. This technique got an impressive 90.71% accuracy and 95.76% sensitivity for the relevant classifications, and its metrics on this dataset outperformed others.

Amir Masoud Rahmani, Amir Haider, Mohammad Adeli et al.

This paper explores the efficacy of Mel Frequency Cepstral Coefficients (MFCCs) in detecting abnormal heart sounds using two classification strategies: a single classifier and an ensemble classifier approach. Heart sounds were first pre-processed to remove noise and then segmented into S1, systole, S2, and diastole intervals, with thirteen MFCCs estimated from each segment, yielding 52 MFCCs per beat. Finally, MFCCs were used for heart sound classification. For that purpose, in the single classifier strategy, the MFCCs from nine consecutive beats were averaged to classify heart sounds by a single classifier (either a support vector machine (SVM), the k nearest neighbors (kNN), or a decision tree (DT)). Conversely, the ensemble classifier strategy employed nine classifiers (either nine SVMs, nine kNN classifiers, or nine DTs) to individually assess beats as normal or abnormal, with the overall classification based on the majority vote. Both methods were tested on a publicly available phonocardiogram database. The heart sound classification accuracy was 91.95% for the SVM, 91.9% for the kNN, and 87.33% for the DT in the single classifier strategy. Also, the accuracy was 93.59% for the SVM, 91.84% for the kNN, and 92.22% for the DT in the ensemble classifier strategy. Overall, the results demonstrated that the ensemble classifier strategy improved the accuracies of the DT and the SVM by 4.89% and 1.64%, establishing MFCCs as more effective than other features, including time, time-frequency, and statistical features, evaluated in similar studies.