# Research Papers — 2026-05 · bundle page 58 of 326 > Full-text markdown of papers #2851–2900 of 16266 published in 2026-05, newest first. AI-readable mirrors of the full 16266-paper 2026-05 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-05/page/58 [← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/57.md) [Next bundle →](https://aipod.dev/papers/archive/2026-05/page/59.md) ### OphIn-500K: Curating Web-Scale Visual Instructions for Scaling Ophthalmic Multimodal Large Language Models Authors: Xuanzhao Dong, Wenhui Zhu, Xiwen Chen, Hao Wang, Xin Li, Yujian Xiong, Jiajun Cheng, Jingjing Wang, Xiaobing Yu, Haiyu Wu, Shao Tang, Zhipeng Wang et al. Published: 2026-05-27 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2605.27916v1 The advancement of general medical Multimodal Large Language Models (MLLMs) has shown great potential for building conversational assistants to support clinical diagnosis. However, their adaptation to highly specialized domains such as ophthalmology remains underexplored, primarily due to the scarcity of large-scale, domain-specific instruction-tuning data. Existing ophthalmic datasets for conversational agents are often limited in scale and largely rely on images from established public benchmarks, limiting the scalability of ophthalmic MLLMs and their ability to capture real-world clinical complexity. To address this gap, we propose $\textbf{OphIn-Engine}$, an ophthalmology-specific instruction data curation pipeline that constructs high-quality instruction data from open-access ophthalmology web-scale videos. The pipeline integrates multimodal transcription for extracting image-transcript pairs, visual cue separation and scoring for identifying clinically relevant visual descriptions, and instruction synthesis with quality control for generating accurate and diverse clinical dialogues. Using this engine, we introduce $\textbf{OphIn-500K}$, a large-scale multimodal ophthalmology instruction-tuning dataset containing over 500,000 instruction instances and more than 151,000 unique images from over 29,000 video clips, formatted as visual question answering (VQA), multi-turn conversational interactions, and chain-of-thought (CoT) reasoning. Built upon this dataset, we further develop $\textbf{OphIn-VL}$, an ophthalmology-specific MLLM with advanced visual understanding and conversational capabilities. Comprehensive experiments and case studies demonstrate that OphIn-VL achieves superior performance compared with state-of-the-art general medical and domain-specific MLLMs. --- ### Does Capability Transfer to Subjective Behavior -- and Would Our Instruments Tell Us? A Self-Evolving, Trust-by-Construction Evaluation Paradigm Authors: Yuming, Huang, Yao Liu, Pengjie Ding, Lei Wang, Junchen Wan Published: 2026-05-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.27914v2 Benchmarking is mature where answers are verifiable -- math, code, reasoning -- but the fastest-growing uses of LLMs are subjective and human-facing: companionship, emotional support, counseling. There the default validity test, correlating a metric to human judgment, has no stable anchor: inter-rater agreement is low, structured by annotator identity, barely reproducible, and length-biased. So we cannot answer the question that matters: does capability that scales on objective benchmarks transfer to subjective behavior, and would our instruments even tell us if it did not? We build an instrument for this regime and report what it reveals at the frontier. We contribute, first, a self-evolving instrument that selects and then authors its own behavioral dimensions under a multiplicative anti-gaming fitness, self-halting when it stops improving; second, a trust-by-construction paradigm that earns belief through three certificates established without a human gold standard, where human raters saturate (rho ~ 0.45); and third, the finding it makes visible -- capability transfer is dissociable. Across 49 models, 8 families, and 24 months, subjective behaviors are where objective-benchmark scaling fails to carry over: the sharpest case, advice-restraint (knowing when not to give advice), is the frontier's universal-lowest dimension, and at gpt-4.1->gpt-5 it ran backwards while the aggregate score hid it -- a regression one instruction recovers. Warm restraint is moved by model generation, not by raw scale, MoE width, inference budget, or reasoning mode; the open-weight Pareto frontier matches closed flagships at ~10-80x lower per-call cost; and four judge families replicate the rubric on held-out human ESConv conversations. Data, code, the locked rubric, and judge prompts will be released upon publication. --- ### Where LLM Annotators Fail: Label-Free Learning on Graphs with LLMs Authors: Safal Thapaliya, Jiatan Huang, Chuxu Zhang Published: 2026-05-27 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.27913v1 Node classification on graphs often requires labeled nodes, yet obtaining labels at graph scale is expensive. When node attributes contain semantic content, such as paper abstracts, web pages, or product descriptions, large language models (LLMs) can provide low-cost supervision by annotating a small subset of nodes. However, these LLM-generated labels are noisy, and existing label-free graph learning methods usually treat this noise as either global or class-conditional. We find that LLM annotation errors are not only class-dependent but also region-dependent: within the same class, reliability can vary sharply across feature-space clusters. In light of this, we propose Cluster-Aware Noise Estimation (CANE), a label-free learning framework that estimates cluster-conditional LLM reliability without ground truth labels, and uses this estimate to decide which pseudo-labels to trust, and which labels to correct. Across various graph benchmarks and GNN backbones, CANE improves over the strongest label-free baselines, with the largest gains on datasets exhibiting stronger cluster-conditional noise. --- ### Privately Estimating Monotone Statistics in Polynomial Time Authors: Gavin Brown, Ephraim Linder, Mahbod Majid, Vikrant Singhal Published: 2026-05-27 Categories: cs.CR, cs.DS, cs.LG arXiv: https://arxiv.org/abs/2605.27912v2 We study efficient differentially private algorithms for estimating monotone statistics, i.e., statistics that are monotone under the addition of new observations. The starting point for our investigation is subsample-and-aggregate: a classical paradigm that partitions the dataset into blocks, estimates the statistic on each block, and then privately aggregates the estimates. While practical and generically applicable, this approach is quite data-hungry. We improve upon this framework for the class of monotone statistics -- compared to subsample-and-aggregate, our algorithms save a factor of $t$ in sample complexity and pay a factor of $e^t$ in running time, where $t>0$ is a tunable parameter. We complement our results with a query-complexity lower bound, showing that our algorithms are essentially optimal for this task. As an application, we obtain improved results for private eigenvalue estimation, private loss estimation, and privately estimating a single parameter of a high-dimensional model, e.g., in linear regression. --- ### SuiChat-CN: Benchmarking Contextual Suicide Risk Assessment in Chinese Group Chats Authors: Xiangyu Wang, Zhiwei Yu, Chengze Du, Dingchang Wang, Yuhan Ye, Fangyu Zheng Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27911v1 Suicide is a critical global public health challenge, causing approximately 720,000 deaths each year and calling for timely, effective prevention strategies. Existing computational studies primarily focus on post-based social media platforms such as Twitter and Weibo, leaving instant messaging environments such as Telegram underexplored. Yet group chats pose distinct challenges: messages are short, fragmented, multi-party, and often rely on implicit or culturally specific expressions, making isolated post-level analysis insufficient. We introduce SuiChat-CN, a Chinese group-chat benchmark for contextual suicide risk assessment. We collect public Telegram group-chat data, construct coherent conversational segments through signal-word extraction and bidirectional context expansion, and annotate user risk levels with an expert-validated, LLM-assisted paradigm. SuiChat-CN contains 13,312 contextual segments from 1,406 users, covering 258,228 raw chat messages. Extensive experiments with PLMs and more than 40 LLMs demonstrate that contextual information is essential for reliable risk assessment, while fine-tuning and partial-context evaluation further reveal the challenges of early detection in multi-party conversations. Due to ethical and sensitivity concerns, the dataset is not publicly released but will be shared with accredited mental health and suicide-prevention research institutions upon reasonable request. --- ### ESC-Skills: Discovering and Self-Evolving Skills for Emotional Support Conversations Authors: Jie Zhu, Huaixia Dou, Shuo Jiang, Junhui Li, Lifan Guo, Feng Chen, Chi Zhang, Fang Kong Published: 2026-05-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.27908v1 Existing emotional support conversation (ESC) systems mainly rely on end-to-end response generation or coarse strategy supervision, offering limited interpretability and little support for systematic skill improvement. We propose ESC-Skills, a skill-centric framework that discovers and self-evolves executable emotional support skills. We first model localized support interactions as Intervention Units (IUs), which capture state--action--outcome dynamics between seeker states, support interventions, and post-response emotional changes. Based on IUs extracted from both successful and failed ESC dialogues, we construct the ESC-Skills Bank, a repository of executable emotional support skills containing intervention guidance, applicability conditions, expected outcomes, and potential risks. To further improve robustness, we introduce a multi-profile self-evolutionary refinement framework in which an ESC agent interacts with diverse simulated seeker profiles under SAGE evaluation. The resulting interaction traces are analyzed to identify missing skills, unsafe interventions, and profile-specific failure patterns, which are then used to refine the Skills Bank through simulation-based verification. Experimental results demonstrate that ESC-Skills improves both response-level quality and dialogue-level emotional outcomes while providing more interpretable and controllable support behaviors. We will release the code, prompts, and ESC-Skills Bank at https://github.com/aliyun/qwen-dianjin. --- ### Reasoning Matters: Mitigate Hallucination in Multimodal Large Reasoning Models via Reasoning-Conditioned Preference Optimization Authors: Jiawei Kong, Hao Fang, Shunxiang Liao, Jinyu Li, Bin Chen, Hao Wu, Shu-Tao Xia, Min Zhang Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27906v1 Multimodal Large Reasoning Models introduce the reasoning paradigm, demonstrating strong capabilities on complex vision-language tasks. However, they still suffer from severe hallucinations. Existing training-based methods typically mitigate hallucinations through response-level direct preference optimization (DPO), where the Chain-of-Thought (CoT) and the final answer are treated as a monolithic output and optimized jointly. We reveal that this formulation performs similarly to answer-only optimization, suggesting that it primarily learns answer-level preference, while leaving CoT-level supervision insufficiently exploited. To address this issue, we explicitly formulate a CoT-oriented preference term and derive Reasoning-Conditioned Direct Preference Optimization (RC-DPO), which models the CoT as a condition for answer generation and contrasts the preference for the same preferred answer under different CoT conditions, promoting answer-supportive reasoning chain alignment. To further improve optimization, we introduce a reasoning-enhanced preference data generation strategy that employs Monte Carlo Tree Search to discover visually grounded and logically consistent CoTs as positive samples, and attention-guided CoT token pruning to construct negative ones. Extensive experiments across various models and benchmarks show that RC-DPO effectively mitigates hallucinations and improves the reliability of the multimodal reasoning process. --- ### AI Research Agents Narrow Scientific Exploration Authors: Yixuan Tang, Yi Yang Published: 2026-05-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.27905v2 AI research agents now support large-scale AI-assisted scientific discovery. We examine whether AI-generated ideas broaden scientific exploration or primarily reinforce existing work. Using five agent frameworks and five large language models, we generate 219,655 ideas for different scientific fields. Across experiments, four consistent patterns emerge. First, AI-generated ideas are more concentrated than human-authored papers within the same research area. Second, they remain much closer to starting literature than later human follow-on work does. Third, AI-generated ideas align less with future human research. Last, AI-generated ideas are located in lower-impact regions of the historical scientific landscape. Overall, current AI research agents appear better suited to local elaboration than to broadening scientific exploration. --- ### Dr-CiK: A Testbed for Foresight-Driven Agents Authors: Yihong Tang, Andrew Robert Williams, Arjun Ashok, Vincent Zhihao Zheng, Lijun Sun, Alexandre Drouin, Issam H. Laradji, Étienne Marcotte, Valentina Zantedeschi Published: 2026-05-27 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.27904v1 Time series forecasting in real-world settings often depends not only on historical observations, but also on external context that must be actively discovered from noisy, heterogeneous information sources. Yet existing context-aided forecasting benchmarks typically assume that the supporting context is already provided, leaving open whether agents can identify it on their own. Therefore, we introduce Dr-CiK, a benchmark for evaluating whether agents can retrieve forecasting-relevant supporting context from a document corpus, filter out distractors, distill the retrieved context into forecast-useful evidence, and generate forecasts supported by that evidence. Through context ablations and evaluations of state-of-the-art deep research and forecasting methods paired together, we show that high-quality context substantially improves forecasting performance in Dr-CiK. However, most existing DR agents recover only a small fraction of the ground-truth supporting evidence (usually <5%), are frequently misled by distractors (>80% distractor citations), and can cause forecasters to perform worse with retrieved context than without context. Our results motivate research on foresight-driven agents that search for the right context to predict the future. --- ### The Fragility of Chain-of-Thought Monitoring Across Typologically Diverse Languages Authors: Eric Onyame, Runtao Zhou, Kowshik Thopalli, Bhavya Kailkhura, Chirag Agarwal Published: 2026-05-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.27901v1 Chain-of-thought (CoT) monitoring has been proposed as a promising safety mechanism for detecting misaligned behavior in large language models. However, its reliability remains largely unexplored beyond English and across diverse model families. We present the first large-scale evaluation of CoT monitorability across 13 diverse languages and seven frontier model families, comprising 16 models. Using adversarial-hint evaluations that require explicit intermediate computation, together with analysis of internal answer-token probabilities, we consistently find CoT unfaithfulness across languages and hint types, with an average rate of 95.9\% across 8B--120B parameter models. We find that frontier models systematically engage in strategic manipulation, including answer-switching, post-hoc rationalization, and procedural exploitation of hints, making external monitors struggle to detect deception. We show that frontier models often commit to the misaligned cue in their latent activations within the first 15\% of generation, even when the CoT appears faithful. Surprisingly, these deceptive patterns remain 100\% in low-resource languages, revealing fundamental limitations in current CoT-based oversight. Our results reveal that CoT monitoring is fundamentally fragile under linguistic distribution shift, providing a substantially weaker safety signal than what English-only studies suggest. These findings underscore an urgent need to develop robust CoT monitors and to accelerate research into white-box monitoring techniques, especially to improve CoT monitorability in mid- and low-resource languages. Our code is available \href{https://multilingual-cot-monitoring.github.io/}{\textcolor{blue}{here}}. --- ### Decoupled Training with Local Reinforcement Fine-Tuning in Federated Learning Authors: Yuting Ma, Lechao Cheng, Xiaohua Xu Published: 2026-05-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.27900v1 Federated Learning (FL) with pre-trained Vision-Language Models (VLMs) has emerged as a promising paradigm for various downstream tasks. By leveraging its strong representations, recent studies improve task adaptation under insufficient local data while preserving generalization. However, these methods emphasize fully local optimization with simple parameter aggregation,which can amplify inter-client optimization inconsistency and intra-client over-specialization under heterogeneous and full-data FL settings, making it difficult to balance global task adaptation and generalization. To address these challenges, we propose FedDTL, a novel federated VLM framework that decouples the image encoder and text encoder across clients and the server. Through decoupled encoder training with server-client modality alignment, FedDTL promotes coherent global semantic update and reduces inter-client optimization inconsistency, improving global task adaptation.To further mitigate intra-client over-specialization,we introduce a two-stage local fine-tuning, where a supervised fine-tuning stage enables rapid and reliable warm-start, followed by a reinforcement learning stage that enhances generalization. Extensive experiments on multiple benchmarks, including label skew and feature shift, demonstrate that FedDTL achieves an effective balance between global task adaptation and generalization under various FL data distributions in both few-shot and full-data regimes. --- ### SKILLC: Learning Autonomous Skill Internalization in LLM Agents via Contrastive Credit Assignment Authors: Hongxiang Lin, Zhirui Kuai, Erpeng Xue, Lei Wang Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27899v1 Structured skill prompts improve exploration in long-horizon agentic reinforcement learning (RL). Skill-augmented RL methods retain external skills at inference, while skill-internalization RL methods withdraw them during training to enable autonomous performance. However, existing internalization approaches only use skill-helpfulness contrast for curriculum control, leaving the policy update unchanged and unable to distinguish skill-dependent from autonomous success. We propose SkillC, a framework based on Contrastive Skill Credit Assignment (CSCA) that converts this contrast into a direct learning signal for internalization. \textsc{SkillC} samples paired skill-injected and skill-free rollouts for tasks from active skill types within the same policy update, and injects their task-level contrast into optimization via a dual-stream advantage estimator that preserves global ranking while applying a one-sided correction toward skill-free success. A smoothed validation-level signal further drives an adaptive curriculum over attribution strength, rollout allocation, and monotonic active-set pruning. Experiments on ALFWorld and WebShop show that, without runtime skill access, SkillC surpasses the strongest prior skill-internalization RL baseline by 5.5\% and 4.4\%, respectively, while remaining competitive with skill-augmented RL methods. --- ### UniACE: A Unified Framework for Evaluating LLM Agentic Capabilities Authors: Pengyu Zhu, Lijun Li, Yaxing Lyu, Qianxin Luo, Jingyi Yang, Yi Liu, Tingfeng Hui, Xinyu Yuan, Li Sun, Sen Su, Jing Shao Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27898v3 Agent benchmarks are increasingly used to compare large language models (LLMs) across domains, yet a reported score reflects a complete model--harness--environment configuration rather than the model alone. Benchmark packages couple native tasks with specific prompts, tool protocols, orchestration logic, and sometimes dynamic external resources, making cross-benchmark comparisons sensitive to implementation and resource conditions. We present UniACE, a unified framework for model-centric evaluation under an explicit, common execution condition. UniACE represents each benchmark as an instruction--tool--environment triplet, executes LLMs through a shared, task-agnostic harness in isolated per-task runtimes, and preserves native success criteria. For tasks that rely on dynamic resources, an optional offline mode replaces live access with fixed, pre-collected snapshots. Its evaluation protocol further standardizes efficiency measurement, execution records, and trace-based failure attribution. We migrate 7 benchmarks spanning 24 domains and evaluate 15 models in more than 400K rollouts consuming 5B tokens. Comparisons with source implementations show large bidirectional score changes and model-ranking reversals, while matched online and offline runs reveal substantial sensitivity to accessible evidence and its representation. Under the shared UniACE configuration, efficiency and failure profiles expose task-dependent model behaviors hidden by task-success scores alone. These findings motivate reporting agent benchmark outcomes as properties of an explicit evaluation configuration, enabling more interpretable and reproducible cross-benchmark comparisons. Codes and benchmarks at are available at https://github.com/whfeLingYu/A-Unified-Framework-for-the-Evaluation-of-LLM-Agentic-Capabilities, https://huggingface.co/datasets/whfeLingYu/Unified_Agent_Framework. --- ### A Unified Framework for the Evaluation of LLM Agentic Capabilities Authors: Pengyu Zhu, Lijun Li, Yaxing Lyu, Qianxin Luo, Jingyi Yang, Yi Liu, Tingfeng Hui, Xinyu Yuan, Li Sun, Sen Su, Jing Shao Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27898v2 As LLMs are increasingly deployed as agents, reliable assessment of their agentic capabilities has become essential. However, reported benchmark scores often jointly reflect model capability and the implementation choices each benchmark is packaged with, making cross-benchmark results difficult to interpret as clean measurements of the underlying model. In this work, we present a unified framework for the fair evaluation of LLM agentic capabilities. Driven by a unified configuration system, the framework integrates diverse benchmarks into a standardized instruction-tool-environment format, executes agents through a fixed ReAct-style architecture within a controllable sandbox, and provides an optional offline setting that replaces volatile live environments with curated snapshots, so that framework effects and environment effects can be analyzed separately. Building on this, we unify the evaluation methodology under each benchmark's original task-success criteria, while introducing unified metrics for resource consumption and a taxonomy for decision- and execution-level failure attribution. Within this framework, we adapt 7 widely used benchmarks spanning 24 domains across single-agent, multi-agent, and safety-critical scenarios, and conduct a large-scale empirical analysis over 400K rollouts and 5B tokens on 15 models. The results show that scaffold choice and environmental volatility materially shift benchmark outcomes in both directions, allowing our framework to disentangle intrinsic LLM capabilities from framework- and environment-induced artifacts. We further demonstrate its extensibility as a secure testbed for safety-critical domains. Codes and benchmarks at are available at https://github.com/whfeLingYu/A-Unified-Framework-for-the-Evaluation-of-LLM-Agentic-Capabilities, https://huggingface.co/datasets/whfeLingYu/Unified_Agent_Framework. --- ### FinBoardBench: Benchmarking Dynamic Wealth Management and Strategic Financial Reasoning of LLMs via Board Game Simulations Authors: Xuesi Hu, Peng Wang, Jinpeng Miao, Xilin Tao, Caiwei Li, Yue Ma, Jie He, Qiancheng Zhang, Yuntao Zou, Dagang Li Published: 2026-05-27 Categories: cs.CL, cs.CE arXiv: https://arxiv.org/abs/2605.27896v1 Recently, large language models (LLMs) have achieved superior performance in static financial reasoning and simple dynamic trading tasks. However, existing static financial benchmarks are insufficient to assess the dynamic wealth management and financial decision-making capabilities of LLMs in real-world environments. To bridge this gap, we present FinBoardBench, an evaluation suite based on three classic financial board games: Cashflow, Acquire, and Monopoly. FinBoardBench assesses a comprehensive set of financial skills, including personal cash flow management with debt balancing, corporate investment and acquisition forecasting, and competitive trade negotiations with asset auctions. Our experiments with 9 advanced LLMs reveal that while exhibiting basic long-term planning and investment logic, they fail to effectively leverage complex interactions for profit, and their strong static reasoning performance does not transform into successful dynamic decision-making. Notably, they tend to prioritize immediate asset acquisition over maintaining sufficient liquidity, making them vulnerable to financial crises triggered by random events. We hope that FinBoardBench can provide a valuable reference for more intelligent LLM-based decision-making systems in the future. --- ### Towards Unified Vision-Language Models with Incomplete Multi-Modal Inputs Authors: Xiang Fang, Wanlong Fang, Changshuo Wang, Keke Tang, Daizong Liu, Siyi Wang, Wei Ji Published: 2026-05-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.27894v1 Video-Language Models (VLMs) have demonstrated impressive multi-modal reasoning capabilities across diverse computer vision applications. However, these VLMs are task-specific and assume that both video and language inputs are complete. However, real-world VLM applications might face challenges due to deactivated sensors (e.g., cameras are unavailable due to data privacy), yielding modality-incomplete data and leading to inconsistency between training and testing data. While straightforward incomplete input can boast training generalization-ability and lead to training failure, its potential risks to VLMs regarding safety and trustworthiness have been largely neglected. To this end, we make the first attempt to propose a unified incomplete video-language model to process the incomplete multi-modal inputs. Extensive experimental results show that our method can serve as a plug-and-play module for previous works to improve their performance in various multi-modal tasks. --- ### SIGMA: Bridging Structural and Distributional Gaps for Vision Foundation Model Adaptation Authors: Lingyu Xiong, Jinjin Shi, Xuran Xu, Cong Luo, Runyu Shi, Ying Huang Published: 2026-05-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.27893v1 Vision Foundation Models (VFMs) have demonstrated impressive representational capabilities. However, adapting them to downstream tasks via full fine-tuning incurs prohibitive computational and storage overhead. Parameter-Efficient Fine-Tuning (PEFT) has emerged as a compelling alternative, aiming to achieve performance parity with full fine-tuning at minimal training costs. Nonetheless, applying PEFT to VFMs for dense prediction tasks remains challenging due to the structural and distributional gaps. To bridge these gaps, we propose \textbf{S}cale-\textbf{I}ntegrated \textbf{G}lobal \textbf{M}odulation \textbf{A}dapter (\textbf{SIGMA}), a novel lightweight PEFT method, which consists of two modules: scale-adaptive fusion and semantic modulation. Specifically, the scale-adaptive fusion module is utilized to bridge structural gaps by enhancing the extraction of multi-granularity visual information. Furthermore, SIGMA introduces semantic modulation on the fusion features to perform global feature alignment to further eliminate the distribution gap. This design facilitates unified spatial and distributional adaptation, requiring only 1.72\% trainable parameters relative to the VFM backbone. Comprehensive experiments across various downstream dense tasks and multiple VFM backbones demonstrate that SIGMA achieves consistent and superior performance over state-of-the-art PEFT methods. --- ### FedEHR-Gen: Federated Synthetic Time-Series EHR Generation via Latent Space Alignment and Distribution-Aware Aggregation Authors: Jun Bai, Ziyang Song, Yue Li Published: 2026-05-27 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.27892v1 Synthetic Electronic Health Record (EHR) generation provides a promising avenue for data augmentation and cross-hospital modeling in privacy-constrained healthcare settings. However, most existing EHR generative models are centralized and require pooling data across hospitals, which is often infeasible when real-world data sharing is restricted. While federated EHR generation offers a natural solution, direct federated modeling often collapses or diverges due to the high dimensionality, sparsity, and cross-hospital heterogeneity of EHR data. In this work, we propose FedEHR-Gen, the first federated framework for synthetic time-series EHR generation across distributed hospitals. FedEHR-Gen uses a two-stage learning paradigm. First, we introduce a federated autoencoder that projects high-dimensional and sparse EHR features onto a compact latent space. To ensure semantic consistency across hospitals, we develop a layer-wise matching aggregation mechanism that aligns local encoders into a unified global latent space. Second, operating on this aligned latent space, we train a federated temporal conditional variational autoencoder (TCVAE) with distribution-aware aggregation, enabling stable temporal generative modeling under severe cross-hospital heterogeneity. Extensive experiments on the eICU and MIMIC-III datasets demonstrate that FedEHR-Gen achieves generation fidelity, downstream utility, and privacy risk comparable to centralized training, while consistently outperforming the standard federated baseline. --- ### SmartDirector: Keyframe-Conditioned Cinematic Video Generation with Narrative Pacing Control Authors: Zhida Zhang, Jie Ma, Zhan Peng, Haoxue Wu, Yang Han, Jun Liang, Jie Cao, Jing Li Published: 2026-05-27 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.27891v1 The narrative quality of a video fundamentally determines its perceptual value. Although existing video generation methods can produce visually appealing content, they predominantly rely on sparse conditioning signals such as text prompts or first/last frames, which limits precise control over narrative structure and temporal pacing. In this paper, we propose SmartDirector, a framework that enhances the narrative capacity of video generation models through multiple keyframes. SmartDirector supports flexible generation scenarios including single-shot generation, multi-shot narrative synthesis, and video extension. The framework operates in two stages: Director-Gen generates a low-resolution video conditioned on the provided keyframes, and Director-SR refines the output by exploiting high-resolution keyframes as semantic anchors to recover fine-grained details. To enable robust multi-keyframe training, we construct a data pipeline that curates single-shot and multi-shot sequences from movies. Extensive experiments demonstrate that SmartDirector substantially outperforms existing state-of-the-art approaches. We will release the code to facilitate further research. --- ### PortBench: A Correlation-Aware, Full-Pipeline Benchmark for LLM-Driven Portfolio Management Authors: Yuxuan Zhao, Sijia Chen, Ningxin Su Published: 2026-05-27 Categories: cs.AI, q-fin.PM arXiv: https://arxiv.org/abs/2605.27887v5 Large language models (LLMs) have shown strong performance across diverse financial tasks, yet portfolio management (PM) remains poorly benchmarked. Existing benchmarks exhibit two gaps: they are often equity-only and ignore cross-asset correlations; they fail to evaluate the complete PM decision pipeline. We introduce PortBench, a benchmark spanning six heterogeneous asset classes from 2015 to 2025. PortBench comprises a static QA dataset of 6,269 questions across seven task templates and a dynamic five-stage allocation pipeline. To evaluate these layers, we introduce two metrics: a dual-layer correlation score for inter-class hedging and intra-class concentration, and CEPS, which quantifies how reasoning errors compound across pipeline stages. We further evaluate under three stress windows and three risk profiles, and support real-time evaluation to mitigate pretraining contamination on historical markets. Across ten frontier LLMs, strong financial QA performance fails to translate into superior portfolio performance: only 32.5\% of 120 evaluations beat equal weighting on Sharpe across four market periods. Our source code is available at \href{https://github.com/AgenticFinLab/portbench}{this https URL}. --- ### PortBench: A Correlation-Aware, Full-Pipeline Benchmark for LLM-Driven Portfolio Management Authors: Yuxuan Zhao, Sijia Chen, Ningxin Su Published: 2026-05-27 Categories: cs.AI, q-fin.PM arXiv: https://arxiv.org/abs/2605.27887v2 Large language models (LLMs) have shown strong performance across diverse financial tasks, yet portfolio management (PM), a critical financial decision-making task, remains poorly benchmarked. Existing benchmarks exhibit two main gaps: they ignore cross-asset correlation structures, thereby failing to distinguish genuinely diversified portfolios from concentrated ones, and fail to evaluate the complete PM decision pipeline in real-world scenarios. We introduce PortBench, a benchmark spanning six heterogeneous asset classes over ten years. PortBench consists of two complementary layers: a static QA dataset of 6,269 correlation-based questions across seven task templates, and a dynamic five-stage allocation pipeline that mirrors the full PM decision cycle. To evaluate these layers, we introduce two dedicated metrics: a dual-layer correlation score that measures whether proposed portfolios exploit inter-class hedging and avoid intra-class concentration, and CEPS, a metric that quantifies how reasoning errors compound across pipeline stages. We further assess strategy robustness and investor alignment under three historical stress regimes and risk profiles. Evaluating ten frontier LLMs, we find that despite strong performance on static financial QA, 90\% of model-profile combinations fail to outperform a basic equal-weight allocation, and models that satisfy every procedural constraint still suffer catastrophic drawdowns under stress. Our source code is available at \href{https://github.com/AgenticFinLab/portbench}{this https URL}. --- ### Reflective Dialogue between Teacher and Solver Agents for Video Question Answering Authors: Takuya Murakawa, Toru Tamaki Published: 2026-05-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.27885v1 Various approaches have been proposed to adapt Vision-Language Models (VLMs) to specialized domains for Video Question Answering, including fine-tuning and in-context learning. However, acquiring task-specific knowledge at the inference phase from only a small labeled support set without fine-tuning remains a challenge. In this paper, we propose a method that achieves adaptation solely through inference-time context injection. Our method first constructs a Reflective Dialogue (RD) -- a multi-turn conversation between two agents, in which Teacher poses each support question and delivers correctness feedback, and Solver answers and provides visual grounding explanations (or reflections) for both correct and incorrect answers. This dialogue history is then used as context at the inference phase. Experiments on the EgoCross benchmark demonstrate that our method outperforms both a baseline zero-shot setting and a standard in-context learning approach that passes support set examples directly, achieving 3rd place in the Open-source Track of the 1st Cross-Domain EgoCross Challenge at the CVPR 2026 EgoVis Workshop, for which this paper also serves as a technical report. --- ### A Road-Conditioned Traffic Movie Prediction Network with Spatiotemporal and Structure-Consistent Learning Authors: Joshua Kofi Asamoah, Blessing Agyei Kyem, Armstrong Aboah Published: 2026-05-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.27884v1 City-wide traffic forecasting is important for congestion management, route guidance, and intelligent transportation systems, but accurate prediction remains challenging when future traffic must be generated as spatial maps over an entire urban network. Existing traffic movie prediction methods have improved frame-level accuracy, yet many still treat forecasting mainly as image reconstruction. This can produce traffic maps that are numerically close to the ground truth but weakly constrained by road layout, connectivity, travel direction, and congestion propagation, especially in cross-city settings where both traffic behavior and road structure change. To address this limitation, this study proposes RCSNet, a road-conditioned spatiotemporal network that reformulates traffic movie prediction as topology-guided future-state generation. RCSNet extracts road-aware representations from static road maps, models multi-horizon traffic dynamics from historical observations, aligns directional traffic features with local road structure, and progressively generates future traffic maps for improved temporal consistency. A structure-consistent learning objective further encourages predictions to remain accurate, road-aligned, and spatially stable. Experiments across multiple cities show that RCSNet improves both forecasting accuracy and structural consistency. In same-city forecasting on Berlin, Antwerp, and Moscow, RCSNet reduces average MAE, MSE, and RMSE by 11.5%, 10.0%, and 5.1%, respectively, compared with the closest baseline. In cross-city testing on unseen Chicago and Bangkok, it reduces RMSE by 10.6% and 10.5% without target-city fine-tuning. Additional horizon-wise, road-structure, explainability, statistical, and efficiency analyses show that RCSNet produces more accurate, transferable, road-aligned, and computationally efficient traffic forecasts. --- ### VibeSearchBench: Benchmarking Long-horizon Proactive Search in the Wild Authors: Xiaohongshu Inc Published: 2026-05-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.27882v2 LLM-based agents score well on search benchmarks, yet real users consistently find results unsatisfying, revealing a persistent evaluation-experience gap. We attribute this gap to existing benchmarks' reliance on over-specified queries, single-turn interactions, and fixed-schema evaluation, none of which reflect real search behavior where users and agents collaboratively refine vague intent through multi-turn dialogue. We term this paradigm VibeSearch and introduce VibeSearchBench, a benchmark comprising 200 manually curated bilingual (Chinese and English) tasks across 20 domains, split into VibeSearch-Pro (professional) and VibeSearch-Daily (daily-life) subsets. Each task pairs a user persona with a schema-free ground-truth knowledge graph, and is evaluated through a progressive-disclosure user simulator and a graph-matching evaluation framework. We benchmark seven frontier models under both the ReAct framework and the OpenClaw agent harness. Results show that all models remain substantially inadequate for VibeSearch (best F1: 30.30), highlighting the need for fundamental advances in long-context reasoning, proactive intent elicitation, and structured knowledge construction. --- ### Retrieval, Reward, and Training Protocols: What Matters in Training Search Agents? Authors: Yibo Zhao, Zichen Ding, Jiayi Wu, Zun Wang, Xiang Li Published: 2026-05-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.27881v2 Search agents powered by large language models can autonomously decompose queries, retrieve information, and synthesize answers through multi-step reasoning. However, the rapid growth of training methods has outpaced controlled comparison: existing works differ in retrieval corpora, reward designs, and training protocols, making it unclear what actually drives improvements. We present a controlled empirical study that isolates three under-explored dimensions of search agent training. First, we identify a critical data-coverage issue in the widely used Wikipedia 2018 corpus and show that correcting it alone yields larger gains than the differences between training algorithms. Second, we systematically compare outcome-based and process-based reward methods across three base models, finding that the simplest outcome-based approach achieves competitive or superior performance in most settings, and that process-level credit assignment can over-correct agent behavior. Third, we analyze training data diversity, off-policy data utilization, and search budget scaling, distilling practical guidelines for training effective search agents. Our code is available at https://github.com/YiboZhao624/SearchAgentReview. --- ### Towards Faithful Agentic XAI: A Verification Method and an Open-World Benchmark for Better Model Faithfulness Authors: Jaechang Kim, Sunung Mun, Seungjoon Lee, Jaewoong Cho, Jungseul Ok Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27879v1 Explainable AI (XAI) helps users interpret model behavior and identify potential faults. Agentic XAI systems use Large Language Models (LLMs) to make explanations more accessible through natural-language interaction, but they can also produce plausible yet unfaithful explanations. This risk arises because unreliable XAI outputs for complex models can be amplified by LLMs and mislead users. We propose Faithful Agentic XAI (FAX), a framework that improves explanation faithfulness through explicit verification. FAX decomposes draft explanations into claims and cross-checks them against inherently faithful tools, filtering unsupported or contradictory claims before final generation. We also introduce CRAFTER-XAI-Bench, an open-world reinforcement learning benchmark with complex policies, diverse goals, and challenging scenarios for assessing model-specific faithfulness. On CRAFTER-XAI-Bench, FAX improves simulation faithfulness from 0.20 for the strongest baseline to 0.46 while maintaining high informativeness, relevance, and fluency. On three tabular benchmarks, FAX performs competitively with prior Agentic XAI baselines, but our analysis shows that these settings can conflate task accuracy with model-specific faithfulness. These findings show that explicit verification is essential for faithful Agentic XAI and that that faithfulness benchmarks must be designed to test explanations against the behavior of the target model itself. --- ### Narrative Flattening: How Post-Training Compresses Thematic, Affective, and Stylistic Variation in LLM Fiction Authors: Zehan Li, Yutong Zhu, Siyang Wu, Honglin Bao, James A. Evans Published: 2026-05-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.27878v2 Large language models produce fluent fiction, yet their creative output is widely seen as flat. We ask where this quality originates in the training and whether it affects different domains of human fiction equally. We construct a matched story-continuation paradigm across StoryStar (public-platform), TMAS (prompt-guided), and The New Yorker (professional literary)-and compare continuations from four OLMo 32B checkpoints (Base, SFT, DPO, RLVR) against matched human text. Because these checkpoints share architecture, scale, tokenizer, and pretraining, the design isolates the post-training effect. We measure each continuation along three sentence-level dimensions: thematic motion, affective prevalence, and linguistic diversity. Across all three, post-training compresses dynamic variation: thematic transitions become more uniform, high-intensity emotions give way to neutrality, and stylistic diversity across stories shrinks. We term this progressive loss narrative flattening. The effect is directionally stable across story domains but gap size depends on the human baseline: professional literary fiction is compressed most, while public-platform and prompt-guided stories show smaller gaps, consistent with their human baselines sitting closer to the model's default rhythm. Post-trained endpoints converge across domains, suggesting alignment produces a continuation regime largely insensitive to the source domain's narrative texture. --- ### Narrative Flattening: How Post-Training Compresses Thematic, Affective, and Stylistic Variation in LLM Fiction Authors: Zehan Li, Yutong Zhu, Siyang Wu, Honglin Bao, James A. Evans Published: 2026-05-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.27878v1 Large language models produce fluent fiction, yet their creative output is widely seen as flat. We ask where this quality originates in the training and whether it affects different domains of human fiction equally. We construct a matched story-continuation paradigm across StoryStar (public-platform), TMAS (prompt-guided), and The New Yorker (professional literary)-and compare continuations from four OLMo 32B checkpoints (Base, SFT, DPO, RLVR) against matched human text. Because these checkpoints share architecture, scale, tokenizer, and pretraining, the design isolates the post-training effect. We measure each continuation along three sentence-level dimensions: thematic motion, affective prevalence, and linguistic diversity. Across all three, post-training compresses dynamic variation: thematic transitions become more uniform, high-intensity emotions give way to neutrality, and stylistic diversity across stories shrinks. We term this progressive loss narrative flattening. The effect is directionally stable across story domains but gap size depends on the human baseline: professional literary fiction is compressed most, while public-platform and prompt-guided stories show smaller gaps, consistent with their human baselines sitting closer to the model's default rhythm. Post-trained endpoints converge across domains, suggesting alignment produces a continuation regime largely insensitive to the source domain's narrative texture. --- ### SPAR: Support-Preserving Action Rectification Authors: Jiaxin Zhao, Weihang Pan, Xun Liang, Binbin Lin Published: 2026-05-27 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.27877v1 Offline policy improvement faces an inherent conflict between maximizing value and fitting the data distribution. While in-sample weighted regression is stable, it suffers from over-conservatism that suppresses high-value actions in the distribution tail; conversely, gradient-based approaches often exhibit a fitting-optimization conflict of gradients, which drives the policy off the data manifold. To address this, we propose Support-Preserving Action Rectification (SPAR), which reframes global learning as a local residual rectification anchored to a frozen pure behavior cloning policy. This framework performs fine-grained fitting and local policy improvement in the residual space, thereby contracting the search space. We further introduce Latent Self-Imitation, utilizing a latent-sampling weighted-regression mechanism to address fitting-improvement gradient conflict in the residual space. Theoretically, we prove this mechanism eliminates the manifold-normal drift of standard value gradients, while extensive D4RL experiments show SPAR extracts significant gains from suboptimal baselines to achieve state-of-the-art performance. --- ### Syllabic-Structure Decoder for Automatic Speech Recognition in Vietnamese Authors: Nghia Hieu Nguyen, Quan Ngoc Hoang, Long Hoang Huu Nguyen, Kiet Van Nguyen, Ngan Luu-Thuy Nguyen Published: 2026-05-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.27874v1 Most Automatic Speech Recognition (ASR) systems formulate transcription as a prediction problem over orthographic units such as characters, subwords, or words. Although effective, such representations do not explicitly reflect the phonetic structure of speech and often require large vocabularies to maintain adequate coverage. In this work, we are motivated from the phonemic features of Vietnamese to propose a Syllabic-Structure Decoder for ASR, which models speech at the phoneme level instead of the orthographic level. Our approach explicitly captures the phonological composition of syllables, enabling the decoder to generate valid syllabic structures from a compact phonemic inventory. This design more closely aligns with the phonetic realization of speech while significantly reducing vocabulary size. Experimental results on two benchmarks: LSVSC, representing standard speech, and UIT-ViMD, a multi-dialect corpus containing diverse regional pronunciations, show that our method consistently outperforms strong previous baselines, especially pretrained baselines such as PhoWhisper and Wav2Vec2, despite using a substantially smaller vocabulary and no additional training resources. These results highlight the effectiveness of phoneme-based syllabic modeling for ASR in this language. Code for experimental reproducibility will be publicly available upon the acceptance of this paper. --- ### AIBuildAI-2: A Knowledge-Enhanced Agent for Automatically Building AI Models Authors: Ruiyi Zhang, Peijia Qin, Qi Cao, Li Zhang, Pengtao Xie Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27873v1 AI models underpin data-centric applications from image and text processing to scientific discovery in biology, physics, and chemistry. Yet developing them remains heavily manual, requiring practitioners to design architectures, build training pipelines, and iteratively refine solutions, making it challenging for natural scientists without specialized AI engineering expertise to build the high-performing models their research demands. To reduce this burden and broaden access to AI for scientific discovery, agents that automatically build AI models have been proposed. However, the performance of these agents is largely limited by the parametric knowledge of their underlying large language models, which is static, often outdated, and sparse on practical AI model engineering know-how. To address this limitation, we introduce AIBuildAI-2, a knowledge-enhanced agent with an external, evolving knowledge system for automatically building AI models. The knowledge system of AIBuildAI-2 is hierarchical, organizing curated AI development knowledge into high-level knowledge instructions over topical categories and low-level knowledge documents under each category, from which the agent dynamically loads only the context relevant to its current state and the AI task being solved, grounding each design and implementation decision in concrete, externally verifiable expertise. The system is initialized by collecting and cleaning AI-development-related documents from the web and organizing them into the corresponding categories, and continually evolves from the agent's own experience by distilling each completed run on an AI task into structured takeaways that are written back into the knowledge system. AIBuildAI-2 achieves state-of-the-art results, ranking first on MLE-Bench with a 70.7% medal rate and placing in the top 6.6% among 4,370 human-expert teams in a heart disease prediction competition. --- ### GRADE: Generalizable Reasoning-Aware Dialogue Evaluation for AI Tutors Authors: Parth Bhalerao, Jeromy Chang, David Chou, Oana Ignat Published: 2026-05-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.27866v2 Evaluating AI tutor responses requires more than factual correctness: tutors must identify mistakes, locate errors, provide guidance, and offer actionable next steps. We present GRADE, a systematic study of open-source models for pedagogical ability assessment in student-tutor dialogues. Building on the BEA 2025 TutorMind setting, we evaluate 120 configurations across five language models, zero-shot inference, LoRA fine-tuning, synthetic augmentation, CoT+Reasoning, and single-task versus multitask formulations. Gemma3-12B performs best for single-task evaluation, while Gemma3-27B in 8-bit precision is more reliable for multitask prediction. We find that augmentation helps models that struggle with the original data, verification adds limited gains despite higher cost, and CoT+Reasoning is more useful for synthetic data generation than direct classification. We further show that LoRA fine-tuning on structured classification objectives interferes with instruction-following behavior under thinking mode, redirecting generation away from the required evaluation format. Carbon analysis shows that model choice and reasoning mode substantially affect emissions. Overall, GRADE shows that carefully selected open-source LoRA pipelines can match or surpass proprietary and ensemble-based systems on key pedagogical dimensions, with code and data available at https://github.com/pvbgeek/GRADE. --- ### MERIT: Matching Expertise via Rubric-Informed Training for Reviewer Assignment Authors: Zixuan Yang, Yibo Zhao, Weicong Liu, Xiang Li Published: 2026-05-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.27865v2 Matching submissions with suitable reviewers at scale is a growing challenge for major venues, yet existing approaches either rely on coarse proxy signals that conflate general relatedness with true suitability, or require expensive human annotations that are difficult to scale for training. We propose MERIT, a two-stage framework that bridges this gap by converting criterion-level expertise matching into scalable suitability supervision. In the first stage, we train a reviewer assessor via reinforcement learning to identify the expertise dimensions a paper requires, match them against the reviewer's prior work, and produce a suitability decision, with rewards provided by an LLM judge guided by paper-specific expertise rubrics. In the second stage, we distill the assessor's predictions into an embedding-based retriever for efficient large-scale assignment. Experiments show that our 4B reviewer assessor outperforms larger general-purpose LLMs on suitability classification, and the resulting retriever achieves state-of-the-art performance across LR-Bench and the CMU Gold dataset. Our code is available at https://github.com/Luli3220/MERIT. --- ### MERIT: Matching Expertise via Rubric-Informed Training for Reviewer Assignment Authors: Zixuan Yang, Yibo Zhao, Weicong Liu, Xiang Li Published: 2026-05-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.27865v1 Matching submissions with suitable reviewers at scale is a growing challenge for major venues, yet existing approaches either rely on coarse proxy signals that conflate general relatedness with true suitability, or require expensive human annotations that are difficult to scale for training. We propose MERIT, a two-stage framework that bridges this gap by converting criterion-level expertise matching into scalable suitability supervision. In the first stage, we train a reviewer assessor via reinforcement learning to identify the expertise dimensions a paper requires, match them against the reviewer's prior work, and produce a suitability decision, with rewards provided by an LLM judge guided by paper-specific expertise rubrics. In the second stage, we distill the assessor's predictions into an embedding-based retriever for efficient large-scale assignment. Experiments show that our 4B reviewer assessor outperforms larger general-purpose LLMs on suitability classification, and the resulting retriever achieves state-of-the-art performance across LR-Bench and the CMU Gold dataset. Our code is available at https://github.com/Luli3220/MERIT. --- ### FundaPod: A Multi-Persona Agent Pod Architecture with Knowledge Graph Memory for AI-Assisted Fundamental Investment Research Authors: Di Zhu, Lei Nico Zheng, Zihan Chen Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27864v5 Large language models (LLMs) are increasingly applied in finance, yet most existing work emphasizes trading signals or financial NLP tasks centered on prediction. Institutional fundamental research, by contrast, requires human analysts or AI agents to gather evidence, identify business drivers, compare competing viewpoints, and generate investment memos. Its broader goal is not merely to predict outcomes, but to produce investment plans that are transparent, reusable, and verifiable, while contributing to the cumulative development of investment knowledge. We present FundaPod, a multi-persona agent pod architecture for AI-assisted fundamental investment research. We argue that fundamental research is a human-centric decision-support task that is qualitatively distinct from trading-signal generation, and is therefore better served by an independence-preserving architecture. In FundaPod, AI agents with different personas, such as value investors or macro strategists, conduct research independently under a shared provenance contract. Their disagreements are then surfaced post hoc for adjudication by the human portfolio manager (PM) through a knowledge-graph memory system. This paper contributes five design principles for human-AI hybrid systems supporting fundamental research, grounded in design-science practice and theories of cognitive isolation and human-machine coordination. It also describes four architectural mechanisms: a persona distillation pipeline that turns public investor materials into deployable agents; a declarative skill registry that lets the planner derive typed task graphs; a grounded evidence model that links memo claims to verifiable sources; and a knowledge-graph "second brain" that connects tickers, memos, analysts, and themes. We demonstrate the architecture through a complete case study and a persona-based memo comparison. --- ### FundaPod: A Multi-Persona Agent Pod Platform with Knowledge Graph Memory for AI-Assisted Fundamental Investment Research Authors: Di Zhu, Lei Nico Zheng, Zihan Chen Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27864v4 Large language models (LLMs) are increasingly applied in finance, yet most existing work emphasizes trading signals or financial NLP tasks centered on prediction. Institutional fundamental research, by contrast, requires human analysts or AI agents to gather evidence, identify business drivers, compare competing viewpoints, and generate investment memos. Its broader goal is not merely to predict outcomes, but to produce investment plans that are transparent, reusable, and verifiable, while contributing to the cumulative development of investment knowledge. We present FundaPod, a multi-persona agent platform for AI-assisted fundamental investment research. We argue that fundamental research is a human-centric decision-support task that is qualitatively distinct from trading-signal generation, and is therefore better served by an independence-preserving architecture. In FundaPod, AI agents with different personas, such as value investors or macro strategists, conduct research independently under a shared provenance contract. Their disagreements are then surfaced post hoc for adjudication by the human portfolio manager (PM) through a knowledge-graph memory system. This paper contributes five design principles for human-AI hybrid systems supporting fundamental research, grounded in design-science practice and theories of cognitive isolation and human-machine coordination. It also describes four architectural mechanisms: a persona distillation pipeline that turns public investor materials into deployable agents; a declarative skill registry that lets the planner derive typed task graphs; a grounded evidence model that links memo claims to verifiable sources; and a knowledge-graph "second brain" that connects tickers, memos, analysts, and themes. We demonstrate the architecture through a complete case study and a persona-based memo comparison. --- ### From Detection to Mechanism: Cross-Attention Graph Neural Networks Enable Drug-Drug Interaction Type Prediction An Ablation Study with Acetylsalicylic Acid Validation Authors: Juergen Dietrich Published: 2026-05-27 Categories: cs.LG, cs.AI, q-bio.QM arXiv: https://arxiv.org/abs/2605.27861v1 Predicting whether two drugs interact (binary detection) is a substantially dif- ferent task from predicting the mechanism type of that interaction (multi-class classification). This study presents a systematic ablation study of three Graph Neural Network (GNN) architectures for drug-drug interaction (DDI) prediction on a publicly available benchmark dataset comprising 38,337 positive pairs across 86 interaction types. Three architectures are compared under identical training conditions (n = 61,339 pairs): a siamese dual Message Passing Neural Network (MPNN) with concatenation (Concat), a dual MPNN with four-head cross-attention (CrossAtt), and a ternary MPNN incorporating an interaction graph (Ternary). CrossAtt improves multi-class F1-macro by +0.186 absolute (+45%) over Concat, while improving binary AUC by only +0.012 (+1.3%) - confirming that atom-level inter-molecular communication specifically enables mechanism-type classification. The ternary architecture underperforms despite equivalent training data, with its failure consistent with a training instability hypothesis. Validation on ten acetylsali- cylic acid (ASA) drug pairs, held out prior to training, demonstrates 10/10 correct DDI-type predictions for CrossAtt versus 0/10 for Ternary. Two consistent failure cases are identified across all architectures, linking to structural limits established in a companion toxicity study. --- ### C-MIG: Multi-view Information Gain-based Retrieval-Augmented Generation for Clinical Diagnosis Reasoning Authors: Yuwei Miao, Gen Li, Yunsheng Zeng, Xiandong Li, Yujin Wang, Siyu Chen, Luning Wang, Yunhao Qiao, Junfeng Wang, Jianwei Lv, Bo Yuan Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27860v2 Retrieval-augmented generation combined with reinforcement learning has shown promise for grounding large language models in trustworthy medical evidence. However, existing methods rely on exact-match binary rewards, which in clinical diagnosis cause two issues: (i) semantically relevant but non-verbatim steps receive zero signal, discarding valuable learning signals; and (ii) uni-dimensional rewards cannot effectively supervise heterogeneous reasoning capabilities. To address these issues, we propose C-MIG, a Multi-view Information Gain-based retrieval-augmented generation framework for Clinical diagnosis. C-MIG estimates information gain under a frozen reference model from two complementary views, retrieved-document and document-refinement, to jointly guide what to retrieve and how to refine, alleviating the issues of valuable reward signal loss and credit assignment. We further design a multi-subquery retrieval augmentation strategy that improves knowledge recall coverage in clinical diagnostic scenarios. Comprehensive experiments on four medical benchmarks demonstrate that C-MIG achieves the best performance among all RAG-RL methods on both in-domain and out-of-domain sets, and outperforms state-of-the-art general-purpose LLMs for clinical diagnosis. --- ### DecomposeRL: Learning to Ask Useful, Informative, and Diverse Questions for Semi-Supervised, Traceable Claim Verification Authors: Shubhashis Roy Dipta, Ankur Padia, Francis Ferraro Published: 2026-05-27 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.27858v1 Claim verification splits between end-to-end classifiers that are accurate but yields no inspectable traces, and decomposition-based methods produce inspectable traces but lag performance on benchmark datasets. We propose DecomposeRL an accurate claim-verifier that produce inspectable traces. DecomposeRL frames decomposition as an RL policy trained with GRPO and a multi-faceted reward ensemble, enabling both fully supervised and semi-supervised learning from unlabeled claims. DecomposeRL addresses the prohibitive training cost of GRPO with a data-curation funnel that distills 115K fact-verification claims into a compact, learning-signal-dense subset of 5K claims. We show that a DecomposeRL-7B policy trained with full supervision on only ~5K curated claims achieves 86.3 in-domain and 69.8 out-of-domain balanced accuracy across 11 claim-verification benchmarks containing biomedical, political, scientific, and general-domain claims. Despite being 4x smaller, it matches 32B baselines and GPT-4.1-mini, and it further outperforms baselines in a semi-supervised setting with only 10% labeled claims data. Code, data, and models are available at https://dipta007.github.io/DecomposeRL --- ### Fine-Tuned LLM as a Complementary Predictor Improving Ads System Authors: Hui Yang, Daiwei He, Kevin Jiang, Taejin Park, Kungang Li, Jiajun Luo, Yuying Chen, Xinyi Zhang, Sihan Wang, Haoyu He, Yu Liu, Lakshmi Manoharan et al. Published: 2026-05-27 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2605.27856v1 Recommendation systems power engagement and monetization across feeds, ads, and short-video platforms, but translating the latest advances in Large Language Models into Recommendation Systems (RecSys) gains remains rare, particularly in advertising and production-scale real-world industry setups. Prior real-world LLM successes typically fall into three buckets: (a) generative retrieval that directly predicts the next items for candidate generation, (b) late-stage re-ranking that uses LLMs, and (c) auxiliary signal enrichment with LLMs. We introduce a complementary paradigm for ads: a fine-tuned open-source LLM used not as a ranker, but as an ads-specific ancillary predictor, forecasting likely advertisers from user profiles and histories. This LLM-driven advertiser prediction augments conventional candidate generation and provides informative priors to downstream ranking. Developed in a large-scale production advertising system, our approach produces substantial offline improvements and measurable online business impact, demonstrating that LLM world knowledge and predictive capacity can be efficiently harnessed. Beyond validating LLMs for ads applications, our results show that targeted ancillary predictions can unlock end-to-end gains across both retrieval and late-stage ranking, offering a practical path to LLM-enhanced recommendation at scale. --- ### MolLingo: Molecule-Native Representations for LLM-Powered Scientific Agents Authors: Thao Nguyen, Heng Ji Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27853v1 We present MolLingo, a multi-agent system that emulates the reasoning process of a chemist to automate molecular design. Existing LLM-based approaches either operate as standalone generative models without access to external tools or lack the multi-agent coordination and shared memory needed for iterative, evidence-driven reasoning across the molecular design pipeline. MolLingo addresses this by coordinating a Literature Agent, a Chemist Agent, and an Orchestrator through a shared memory module, with each agent equipped with domain-specific tools. To enable effective molecular reasoning, we introduce BRICS-based Fragment Enumeration (BFE), a synthesis-aware molecular fragmentation method that decomposes molecules into chemically meaningful building blocks represented as block-based SMILES paired with common chemical names. This representation bridges molecular structure and LLM semantic space, enabling block-level reasoning and editing that is difficult with raw SMILES alone. As a case study in early-stage therapeutic design, MolLingo further grounds the Chemist Agent's reasoning in binding site geometry and residue-level protein context derived from molecular docking to optimize molecules for stronger target binding. Across four benchmarks, MolLingo consistently outperforms frontier LLMs and specialized baselines, including a fourfold docking score improvement over GPT-5.4 despite using the same underlying model, consistent drug property optimization gains across multiple LLM backbones, and state-of-the-art results on TOMG-Bench, surpassing both frontier LLMs and the RL-based optimization method RePO. Our results suggest that LLMs are already capable molecular design assistants when guided through chemically meaningful representations and biologically grounded structural context. Code is available at: https://anonymous.4open.science/status/MolLingo-7450. --- ### ClothTransformer: Unified Latent-Space Transformers for Scalable Cloth Simulation Authors: Yu Zhang, Yidi Shao, Wenqi Ouyang, Yushi Lan, Zhexin Liang, Chengrui Wu, Xudong Xu, Xingang Pan Published: 2026-05-27 Categories: cs.GR, cs.CV arXiv: https://arxiv.org/abs/2605.27852v5 Unified and scalable Transformers have recently achieved remarkable success in modeling diverse phenomena traditionally associated with computer graphics, such as 3D visual effects, rendering processes, and motion in videos. In this work, we take a step further by investigating whether modern Transformer techniques can tackle the challenging task of cloth simulation. To this end, we present ClothTransformer, a framework that reformulates cloth simulation as autoregressive sequence modeling in a learned latent space. Existing neural cloth simulators are largely specialized to single scenarios, intrinsically coupled to the mesh discretization, and lack robust collision handling. Our approach addresses these limitations through three contributions: (1) a unified Transformer architecture that handles diverse scenarios -- body-driven garments, robotic manipulation, and free-fall collisions -- under a single model and achieves approximately $4$--$9{\times}$ lower error than prior state-of-the-art methods across all scenarios; (2) a scalable latent-space formulation that compresses arbitrary-resolution meshes into a fixed-size set of latent tokens, making temporal dynamics computation independent of mesh resolution; and (3) a diverse-scenario high-fidelity penetration-free dataset of ${\sim}$493.4k frames spanning all three settings, which enables a differentiable Continuous Collision Detection (CCD) module to suppress penetration artifacts. Project Page: https://yucrazing.github.io/clothtransformer/ --- ### When Context Flips, Safety Breaks: Diagnosing Brittle Safety in Aligned Language Models Authors: Dasol Choi, Alex Kwon Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27851v1 Safety benchmark scores provide incomplete evidence of deployment readiness: aligned language models often adhere to rigid rules even when a situational update flips which action is safe. We term this failure brittle safety. To diagnose it, we introduce context-flip evaluation, testing 12 models across a safety benchmark (PacifAIst) and two commonsense controls using paired variants where the nominally safe action produces harm. Three findings emerge. First, brittle safety is safety-specific: all 12 models exhibit a safety-commonsense gap (mean +17.4 pp). Baseline accuracy fails to predict brittleness: among models above 90% baseline accuracy, brittleness rates range from 13.7% to 90.0%. Second, failures stem from policy override rather than miscomprehension: despite acknowledging the context change in every case, models persist via three distinct mechanisms that vary by update type and model family. Third, on a hand-audited probe of catastrophic consequence-flip scenarios, standard action-level guardrails catch none, while a state-aware validator catches all without false alarms on correct interventions. This indicates that action-level content moderation is systematically blind to consequence-flips, motivating state-aware architectural alternatives. We release our protocol, perturbed benchmarks, and deployment probe. --- ### TCP-MCP: Landscape-Guided Co-Evolution of Prompts and Communication Topologies for Multi-Agent Systems Authors: Yi Ding, Zijie Xuan, Haowei Zhou, Zhenyu Ju, Xiaoxiao Dong, Jingwen Zhang, Xingyu Zhu, Leixin Sun, Haochi Zhang Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27850v1 Effective multi-agent systems cannot be designed by selecting prompts or communication graphs in isolation. Agent behavior depends on the information an agent receives, while the usefulness of a communication edge depends on how the receiving agent interprets and uses that information. We propose \textbf{TCP-MCP} (Topology-Coupled Prompting for Multi-Agent Collaborative Problem-Solving), a co-evolution framework that searches agent prompts and communication topologies as a unified genome. TCP-MCP uses an initialization-time landscape probe to calibrate early search behavior, and then relies on Pareto-front diagnostics to adapt exploration under three objectives: task performance, token cost, and structural complexity. Using the same DeepSeek-V3.2 backbone across all methods, TCP-MCP achieves 82.66\%, 89.96\%, and 96.61\% accuracy on MMLU-Pro, MMLU, and GSM8K, respectively. Across the three benchmarks, it consistently outperforms automated graph-generation baselines and achieves competitive accuracy relative to debate-style systems, while using up to 5.69$\times$ fewer tokens than those systems at the reported operating points. These results show that jointly evolving prompts and communication structure provides a practical route to cost-aware and task-adaptive multi-agent system design in controlled evaluations. --- ### FPMoE: A Sparse Mixture-of-Experts Approach to Functional Code Generation Authors: Loc Pham, Lang Hong Nguyet Anh, Thanh Le-Cong Published: 2026-05-27 Categories: cs.PL, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.27849v1 Despite rapid progress in LLM-based code generation, existing models are predominantly trained on imperative languages, leaving functional programming languages (FPLs) such as Haskell, OCaml, and Scala chronically underexplored, with even frontier models performing substantially worse on FPLs. Fine-tuning is a natural remedy, but our experiments show that per-language fine-tuning fails to capture shared functional abstractions, while merged multi-language fine-tuning introduces cross-language interference. To address this, we introduce FPMoE, a lightweight, open-source code generation model built on a sparse Mixture-of-Experts (MoE) architecture with three language-specific routed experts (one each for Haskell, OCaml, and Scala) and a shared expert that captures cross-language functional patterns such as monadic reasoning and type-directed programming. This design resolves both failure modes simultaneously: dedicated experts eliminate interference, while the shared expert preserves abstractions that per-language models miss. On FPEval, FPMoE substantially outperforms fine-tuned baselines and, with only 3B active parameters, matches the performance of much larger models including DeepSeek-Coder-6.7B, Qwen2.5-Coder-14B-Instruct, and Qwen3-Coder-30B-A3B. --- ### EAPO: Entropy-Driven Adaptive Positive-Negative Sample Weighting for Policy Optimization in Open-Ended QA Authors: Yunsheng Zeng, Gen Li, Yuwei Miao, Xiandong Li, Yujin Wang, Siyu Chen, Luning Wang, Yunhao Qiao, Junfeng Wang, Jianwei Lv, Bo Yuan Published: 2026-05-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.27846v1 Large Reasoning Models are typically trained via reinforcement learning from verifiable rewards (RLVR). However, existing approaches adopt fixed weights for positive and negative samples, and the conclusions hardly generalize to open-ended question answering (QA). In this paper, we systematically investigate the roles of positive and negative samples in reinforcement learning for open-ended QA. We propose a reward-mean-based strategy for distinguishing positive from negative samples, and observe that negative samples predominantly govern response diversity and the performance upper bound, whereas positive samples primarily determine response quality and convergence stability. Building on these observations, we propose EAPO, an Entropy-driven Adaptive Policy Optimization method that adaptively computes the weighting coefficients of positive samples based on the ratio of the current policy entropy to the initial entropy. During the entropy-decreasing phase, the weight assigned to positive samples is reduced to preserve exploration, whereas during the entropy-increasing phase it is amplified to reinforce stability, thereby mitigating entropy collapse. Experiments on two publicly available open-ended medical QA datasets demonstrate that EAPO consistently and substantially outperforms fixed-weight baselines in both response diversity and stability. --- ### Snippet-Driven Supply Chain Discovery with LLMs: Scaling Visibility in China Authors: Hiroto Fukada, Takayuki Mizuno Published: 2026-05-27 Categories: cs.SI, cs.AI, physics.soc-ph arXiv: https://arxiv.org/abs/2605.27845v1 Financial and economic research often relies on structured supply-chain disclosures and commercial databases. In China, supplier--customer disclosure is typically limited to major partners of listed firms, leaving unlisted firms and long-tail inter-firm links poorly captured in structured data. Public web evidence can partly complement this gap through corporate, government, and trade-media disclosures; however, full-text web mining at scale is costly because pages are often inaccessible or expensive to process with large language models (LLMs). We propose a snippet-driven method for constructing a supply chain knowledge graph (SCKG), with firms as nodes and inter-firm relationships as edges. Web search snippets are query-biased summaries returned with search results. We use them as a scalable first-pass evidence layer for LLM-based relationship extraction. We evaluate the pipeline in terms of extraction efficiency and coverage. For extraction efficiency, exhaustive full-text chunking discovers 19.8$\times$ more unique relationships than snippets, but requires 251.2$\times$ more input tokens and yields higher redundancy. For coverage, we use 130,685 Chinese firms as search seeds, covering Shanghai/Shenzhen-listed firms and large unlisted firms as of 2024. In the listed-firm subset, the resulting SCKG covers 7.2$\times$ more firms and 9.3$\times$ more relationships than the CSMAR disclosure-based benchmark, while revealing heavy-tailed degree patterns. Retained provenance metadata make the SCKG an auditable complement to disclosure-based databases. --- ### A self-supervised learning approach to deep filter banks for texture recognition Authors: Joao B. Florindo, Lucas O. Lyra, Antonio E. Fabris Published: 2026-05-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.27843v1 An important challenge in texture recognition is the limited amount of data for training frequently found in real-world applications. In computer vision in general, a successful strategy to mitigate this issue is the use of a pretraining stage where the neural network learns to identify relations between parts of the data in a self-supervised manner. A well-established framework in this direction is masked autoencoder. Nevertheless, these models usually rely on computationally intensive architectures, such as vision transformers. In the particular case of texture images, most of the relevant information is compacted within a delimited area around each pixel, which suggests that capturing long-range dependence via the attention mechanism may be unnecessary. Based on that assumption, here we propose a framework where the pretraining model is a convolutional autoencoder. To leverage the rich information conveyed by texture patterns, we employ deep filters coupled with Fisher vector pooling. In this way, we improve the performance of texture recognition without adding significant computational burden. Our approach is compared with several state-of-the-art methods in different texture databases, confirming its potential both in terms of classification accuracy and computational complexity. --- ### LoSATok: Low-Dimensional Semantic-Acoustic Tokenizer for Cross-Domain Audio Understanding and Generation Authors: Zhisheng Zhang, Xiang Li, Yixuan Zhou, Jing Peng, Guoyang Zeng, Zhiyong Wu Published: 2026-05-27 Categories: eess.AS, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2605.27840v2 Audio tokenizers are fundamental to unifying audio understanding and generation. Understanding requires high-level semantics, while generation demands semantic and acoustic details. Existing unified tokenizers jointly encode both in high-dimensional continuous latents, which increases the modeling burden of Diffusion Transformers (DiTs) for generation. We propose LoSATok, a low-dimensional audio tokenizer for cross-domain audio understanding and generation. Motivated by the observation that 1280-dimensional semantic encoder features are compressible, we introduce a Semantic Bottleneck that compresses them into 128 dimensions, regularized by the proposed time-relation loss for temporal feature consistency. We further design a dual-level semantic supervision method that leverages both high- and low-dimensional semantic signals, enabling the tokenizer to jointly capture semantics and acoustic details within a compact latent space. Experiments on speech, music, and general audio show that SemBo preserves strong low-dimensional semantic capacity and LoSATok retains competitive understanding performance compared with several semantic representations, while consistently improving DiT modeling performance on speech, music, and audio generation. These results demonstrate that LoSATok's low-dimensional representations can effectively support audio understanding and generation. Our code is provided at https://github.com/wxzyd123/LoSATok. --- ### LoSATok: Low-dimensional Semantic-Acoustic Tokenizer for Cross-Domain Audio Understanding and Generation Authors: Zhisheng Zhang, Xiang Li, Yixuan Zhou, Jing Peng, Guoyang Zeng, Zhiyong Wu Published: 2026-05-27 Categories: eess.AS, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2605.27840v1 Audio tokenizers are fundamental to unifying audio understanding and generation. Understanding requires high-level semantics, while generation demands semantic and acoustic details. Existing unified tokenizers jointly encode both in high-dimensional continuous latents, which increases the modeling burden of Diffusion Transformers (DiTs) for generation. We propose LoSATok, a low-dimensional audio tokenizer for cross-domain audio understanding and generation. Motivated by the observation that 1280-dimensional semantic encoder features are compressible, we introduce a Semantic Bottleneck that compresses them into 128 dimensions, regularized by the proposed time-relation loss for temporal feature consistency. We further design a dual-level semantic supervision method that leverages both high- and low-dimensional semantic signals, enabling the tokenizer to jointly capture semantics and acoustic details within a compact latent space. Experiments on speech, music, and general audio show that SemBo preserves strong low-dimensional semantic capacity and LoSATok retains competitive understanding performance compared with several semantic representations, while consistently improving DiT modeling performance on speech, music, and audio generation. These results demonstrate that LoSATok's low-dimensional representations can effectively support audio understanding and generation. Our code is provided at https://github.com/wxzyd123/LoSATok. ---