# Research Papers — 2026-04 · bundle page 187 of 233 > Full-text markdown of papers #9301–9350 of 11612 published in 2026-04, newest first. AI-readable mirrors of the full 11612-paper 2026-04 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-04/page/187 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/186.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/188.md) ### Graph-Based Chain-of-Thought Pruning for Reducing Redundant Reflections in Reasoning LLMs Authors: Hongyuan Yuan, Xinran He, Run Shao, Bolei He, Xianwei Xue, Mengke Chen, Qiutong Pan, Haiwei Wang, Haifeng Li Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05643v2 Extending CoT through RL has been widely used to enhance the reasoning capabilities of LLMs. However, due to the sparsity of reward signals, it can also induce undesirable thinking patterns such as overthinking, i.e., generating redundant intermediate reasoning content. In this work, we argue that a major source of such redundancy is inefficient reflection, which often manifests in two problematic patterns: Indiscriminate Reflection, where the model performs broad, low-impact checks throughout reasoning, and Repetitive Reflection, where it repeatedly re-verifies an already established conclusion. To address this, we introduce a graph-based CoT optimization framework. Specifically, we convert each linear CoT into a directed acyclic graph (DAG) with explicit dependency edges, and design a dual pruning strategy: branch-level pruning removes weakly contributing reflection branches, while depth-level pruning eliminates late-stage re-verification. We distill this behavior via a three-stage pipeline: (1) SFT to initialize the policy on pruned concise traces, (2) DPO to prefer correct but less redundant trajectories, and (3) GRPO with length penalty to jointly optimize answer correctness and efficiency. Experiments show that our approach reduces the average reasoning tokens by 42\% while maintaining or improving accuracy. --- ### Parametric Nonconvex Optimization via Convex Surrogates Authors: Renzi Wang, Panagiotis Patrinos, Alberto Bemporad Published: 2026-04-07 Categories: math.OC, cs.LG, eess.SY arXiv: https://arxiv.org/abs/2604.05640v1 This paper presents a novel learning-based approach to construct a surrogate problem that approximates a given parametric nonconvex optimization problem. The surrogate function is designed to be the minimum of a finite set of functions, given by the composition of convex and monotonic terms, so that the surrogate problem can be solved directly through parallel convex optimization. As a proof of concept, numerical experiments on a nonconvex path tracking problem confirm the approximation quality of the proposed method. --- ### PanopticQuery: Unified Query-Time Reasoning for 4D Scenes Authors: Ruilin Tang, Yang Zhou, Zhong Ye, Wenxi Liu, Yan Huang, Shengfeng He Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05638v1 Understanding dynamic 4D environments through natural language queries requires not only accurate scene reconstruction but also robust semantic grounding across space, time, and viewpoints. While recent methods using neural representations have advanced 4D reconstruction, they remain limited in contextual reasoning, especially for complex semantics such as interactions, temporal actions, and spatial relations. A key challenge lies in transforming noisy, view-dependent predictions into globally consistent 4D interpretations. We introduce PanopticQuery, a framework for unified query-time reasoning in 4D scenes. Our approach builds on 4D Gaussian Splatting for high-fidelity dynamic reconstruction and introduces a multi-view semantic consensus mechanism that grounds natural language queries by aggregating 2D semantic predictions across multiple views and time frames. This process filters inconsistent outputs, enforces geometric consistency, and lifts 2D semantics into structured 4D groundings via neural field optimization. To support evaluation, we present Panoptic-L4D, a new benchmark for language-based querying in dynamic scenes. Experiments demonstrate that PanopticQuery sets a new state of the art on complex language queries, effectively handling attributes, actions, spatial relationships, and multi-object interactions. A video demonstration is available in the supplementary materials. --- ### Towards Athlete Fatigue Assessment from Association Football Videos Authors: Xavier Bou, Nathan Correger, Alexandre Cloots, Cédric Gavage, Silvio Giancola, Cédric Schwartz, François Delvaux, Rudi Cloots, Marc Van Droogenbroeck, Anthony Cioppa Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05636v1 Fatigue monitoring is central in association football due to its links with injury risk and tactical performance. However, objective fatigue-related indicators are commonly derived from subjective self-reported metrics, biomarkers derived from laboratory tests, or, more recently, intrusive sensors such as heart monitors or GPS tracking data. This paper studies whether monocular broadcast videos can provide spatio-temporal signals of sufficient quality to support fatigue-oriented analysis. Building on state-of-the-art Game State Reconstruction methods, we extract player trajectories in pitch coordinates and propose a novel kinematics processing algorithm to obtain temporally consistent speed and acceleration estimates from reconstructed tracks. We then construct acceleration--speed (A-S) profiles from these signals and analyze their behavior as fatigue-related performance indicators. We evaluate the full pipeline on the public SoccerNet-GSR benchmark, considering both 30-second clips and a complete 45-minute half to examine short-term reliability and longer-term temporal consistency. Our results indicate that monocular GSR can recover kinematic patterns that are compatible with A-S profiling while also revealing sensitivity to trajectory noise, calibration errors, and temporal discontinuities inherent to broadcast footage. These findings support monocular broadcast video as a low-cost basis for fatigue analysis and delineate the methodological challenges for future research. --- ### From Uniform to Learned Knots: A Study of Spline-Based Numerical Encodings for Tabular Deep Learning Authors: Manish Kumar, Anton Frederik Thielmann, Christoph Weisser, Benjamin Säfken Published: 2026-04-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.05635v2 Numerical preprocessing remains a critical component of tabular deep learning, as the representation of continuous features can strongly affect downstream performance. We systematically study spline-based numerical encodings, including B-splines, M-splines, and integrated splines (I-splines), under uniform, quantile-based, target-aware, and learnable-knot placement. For the learnable variants, we adopt a differentiable knot parameterization that enables stable end-to-end optimization of knot locations jointly with the backbone. We evaluate these encodings on a diverse collection of public regression and classification datasets using MLP, ResNet, and FT-Transformer backbones, and compare them against common numerical preprocessing baselines. Our results show that the effectiveness of numerical encoding depends strongly on the task, encoding size, and backbone. For classification, piecewise-linear encoding (PLE) is the most robust choice overall, while spline-based encodings remain competitive. For regression, no single encoding dominates, with performance depending on the spline family and knot-placement strategy, and larger gains generally observed for MLP and ResNet than for FT-Transformer. Learnable-knot variants can be optimized stably but may substantially increase training cost. Overall, numerical encodings should therefore be assessed jointly in terms of predictive performance and computational overhead. The implementation is publicly available at https://github.com/mkumar73/tdl-numerical-encodings/. --- ### From Uniform to Learned Knots: A Study of Spline-Based Numerical Encodings for Tabular Deep Learning Authors: Manish Kumar, Anton Frederik Thielmann, Christoph Weisser, Benjamin Säfken Published: 2026-04-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.05635v1 Numerical preprocessing remains an important component of tabular deep learning, where the representation of continuous features can strongly affect downstream performance. Although its importance is well established for classical statistical and machine learning models, the role of explicit numerical preprocessing in tabular deep learning remains less well understood. In this work, we study this question with a focus on spline-based numerical encodings. We investigate three spline families for encoding numerical features, namely B-splines, M-splines, and integrated splines (I-splines), under uniform, quantile-based, target-aware, and learnable-knot placement. For the learnable-knot variants, we use a differentiable knot parameterization that enables stable end-to-end optimization of knot locations jointly with the backbone. We evaluate these encodings on a diverse collection of public regression and classification datasets using MLP, ResNet, and FT-Transformer backbones, and compare them against common numerical preprocessing baselines. Our results show that the effect of numerical encodings depends strongly on the task, output size, and backbone. For classification, piecewise-linear encoding (PLE) is the most robust choice overall, while spline-based encodings remain competitive. For regression, no single encoding dominates uniformly. Instead, performance depends on the spline family, knot-placement strategy, and output size, with larger gains typically observed for MLP and ResNet than for FT-Transformer. We further find that learnable-knot variants can be optimized stably under the proposed parameterization, but may substantially increase training cost, especially for M-spline and I-spline expansions. Overall, the results show that numerical encodings should be assessed not only in terms of predictive performance, but also in terms of computational overhead. --- ### PECKER: A Precisely Efficient Critical Knowledge Erasure Recipe For Machine Unlearning in Diffusion Models Authors: Zhiyong Ma, Zhitao Deng, Huan Tang, Jialin Chen, Zhijun Zheng, Zhengping Li, Qingyuan Chuai Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05634v2 Machine unlearning (MU) has become a critical technique for GenAI models' safe and compliant operation. While existing MU methods are effective, most impose prohibitive training time and computational overhead. Our analysis suggests the root cause lies in poorly directed gradient updates, which reduce training efficiency and destabilize convergence. To mitigate these issues, we propose PECKER, an efficient MU approach that matches or outperforms prevailing methods. Within a distillation framework, PECKER introduces a saliency mask to prioritize updates to parameters that contribute most to forgetting the targeted data, thereby reducing unnecessary gradient computation and shortening overall training time without sacrificing unlearning efficacy. Our method generates samples that unlearn related class or concept more quickly, while closely aligning with the true image distribution on CIFAR-10 and STL-10 datasets, achieving shorter training times for both class forgetting and concept forgetting. --- ### SGANet: Semantic and Geometric Alignment for Multimodal Multi-view Anomaly Detection Authors: Letian Bai, Chengyu Tao, Juan Du Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05632v1 Multi-view anomaly detection aims to identify surface defects on complex objects using observations captured from multiple viewpoints. However, existing unsupervised methods often suffer from feature inconsistency arising from viewpoint variations and modality discrepancies. To address these challenges, we propose a Semantic and Geometric Alignment Network (SGANet), a unified framework for multimodal multi-view anomaly detection that effectively combines semantic and geometric alignment to learn physically coherent feature representations across viewpoints and modalities. SGANet consists of three key components. The Selective Cross-view Feature Refinement Module (SCFRM) selectively aggregates informative patch features from adjacent views to enhance cross-view feature interaction. The Semantic-Structural Patch Alignment (SSPA) enforces semantic alignment across modalities while maintaining structural consistency under viewpoint transformations. The Multi-View Geometric Alignment (MVGA) further aligns geometrically corresponding patches across viewpoints. By jointly modeling feature interaction, semantic and structural consistency, and global geometric correspondence, SGANet effectively enhances anomaly detection performance in multimodal multi-view settings. Extensive experiments on the SiM3D and Eyecandies datasets demonstrate that SGANet achieves state-of-the-art performance in both anomaly detection and localization, validating its effectiveness in realistic industrial scenarios. --- ### Beyond Behavior: Why AI Evaluation Needs a Cognitive Revolution Authors: Amir Konigsberg Published: 2026-04-07 Categories: cs.AI, cs.ET, cs.HC arXiv: https://arxiv.org/abs/2604.05631v1 In 1950, Alan Turing proposed replacing the question "Can machines think?" with a behavioral test: if a machine's outputs are indistinguishable from those of a thinking being, the question of whether it truly thinks can be set aside. This paper argues that Turing's move was not only a pragmatic simplification but also an epistemological commitment, a decision about what kind of evidence counts as relevant to intelligence attribution, and that this commitment has quietly constrained AI research for seven decades. We trace how Turing's behavioral epistemology became embedded in the field's evaluative infrastructure, rendering unaskable a class of questions about process, mechanism, and internal organization that cognitive psychology, neuroscience, and related disciplines learned to ask. We draw a structural parallel to the behaviorist-to-cognitivist transition in psychology: just as psychology's commitment to studying only observable behavior prevented it from asking productive questions about internal mental processes until that commitment was abandoned, AI's commitment to behavioral evaluation prevents it from distinguishing between systems that achieve identical outputs through fundamentally different computational processes, a distinction on which intelligence attribution depends. We argue that the field requires an epistemological transition comparable to the cognitive revolution: not an abandonment of behavioral evidence, but a recognition that behavioral evidence alone is insufficient for the construct claims the field wishes to make. We articulate what a post-behaviorist epistemology for AI would involve and identify the specific questions it would make askable that the field currently has no way to ask. --- ### A Unified Foundation Model for All-in-One Multi-Modal Remote Sensing Image Restoration and Fusion with Language Prompting Authors: Yongchuan Cui, Peng Liu Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05629v1 Remote sensing imagery suffers from clouds, haze, noise, resolution limits, and sensor heterogeneity. Existing restoration and fusion approaches train separate models per degradation type. In this work, we present Language-conditioned Large-scale Remote Sensing restoration model (LLaRS), the first unified foundation model for multi-modal and multi-task remote sensing low-level vision. LLaRS employs Sinkhorn-Knopp optimal transport to align heterogeneous bands into semantically matched slots, routes features through three complementary mixture-of-experts layers (convolutional experts for spatial patterns, channel-mixing experts for spectral fidelity, and attention experts with low-rank adapters for global context), and stabilizes joint training via step-level dynamic weight adjustment. To train LLaRS, we construct LLaRS1M, a million-scale multi-task dataset spanning eleven restoration and enhancement tasks, integrating real paired observations and controlled synthetic degradations with diverse natural language prompts. Experiments show LLaRS consistently outperforms seven competitive models, and parameter-efficient finetuning experiments demonstrate strong transfer capability and adaptation efficiency on unseen data. Repo: https://github.com/yc-cui/LLaRS --- ### YoNER: A New Yorùbá Multi-domain Named Entity Recognition Dataset Authors: Peace Busola Falola, Jesujoba O. Alabi, Solomon O. Akinola, Folashade T. Ogunajo, Emmanuel Oluwadunsin Alabi, David Ifeoluwa Adelani Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05624v1 Named Entity Recognition (NER) is a foundational NLP task, yet research in Yorùbá has been constrained by limited and domain-specific resources. Existing resources, such as MasakhaNER (a manually annotated news-domain corpus) and WikiAnn (automatically created from Wikipedia), are valuable but restricted in domain coverage. To address this gap, we present YoNER, a new multidomain Yorùbá NER dataset that extends entity coverage beyond news and Wikipedia. The dataset comprises about 5,000 sentences and 100,000 tokens collected from five domains including Bible, Blogs, Movies, Radio broadcast and Wikipedia, and annotated with three entity types: Person (PER), Organization (ORG) and Location (LOC), following CoNLL-style guidelines. Annotation was conducted manually by three native Yorùbá speakers, with an inter-annotator agreement of over 0.70, ensuring high quality and consistency. We benchmark several transformer encoder models using cross-domain experiments with MasakhaNER 2.0, and we also assess the effect of few-shot in-domain data using YoNER and cross-lingual setups with English datasets. Our results show that African-centric models outperform general multilingual models for Yorùbá, but cross-domain performance drops substantially, particularly for blogs and movie domains. Furthermore, we observed that closely related formal domains, such as news and Wikipedia, transfer more effectively. In addition, we introduce a new Yorùbá-specific language model (OyoBERT) that outperforms multilingual models in in-domain evaluation. We publicly release the YoNER dataset and pretrained OyoBERT models to support future research on Yorùbá natural language processing. --- ### DetailVerifyBench: A Benchmark for Dense Hallucination Localization in Long Image Captions Authors: Xinran Wang, Yuxuan Zhang, Xiao Zhang, Haolong Yan, Muxi Diao, Songyu Xu, Zhonghao Yan, Hongbing Li, Kongming Liang, Zhanyu Ma Published: 2026-04-07 Categories: cs.CV, cs.CL, cs.MM arXiv: https://arxiv.org/abs/2604.05623v1 Accurately detecting and localizing hallucinations is a critical task for ensuring high reliability of image captions. In the era of Multimodal Large Language Models (MLLMs), captions have evolved from brief sentences into comprehensive narratives, often spanning hundreds of words. This shift exponentially increases the challenge: models must now pinpoint specific erroneous spans or words within extensive contexts, rather than merely flag response-level inconsistencies. However, existing benchmarks lack the fine granularity and domain diversity required to evaluate this capability. To bridge this gap, we introduce DetailVerifyBench, a rigorous benchmark comprising 1,000 high-quality images across five distinct domains. With an average caption length of over 200 words and dense, token-level annotations of multiple hallucination types, it stands as the most challenging benchmark for precise hallucination localization in the field of long image captioning to date. Our benchmark is available at https://zyx-hhnkh.github.io/DetailVerifyBench/. --- ### FunRec: Reconstructing Functional 3D Scenes from Egocentric Interaction Videos Authors: Alexandros Delitzas, Chenyangguang Zhang, Alexey Gavryushin, Tommaso Di Mario, Boyang Sun, Rishabh Dabral, Leonidas Guibas, Christian Theobalt, Marc Pollefeys, Francis Engelmann, Daniel Barath Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05621v2 We present FunRec, a method for reconstructing functional 3D digital twins of indoor scenes directly from egocentric RGB-D interaction videos. Unlike existing methods on articulated reconstruction, which rely on controlled setups, multi-state captures, or CAD priors, FunRec operates directly on in-the-wild human interaction sequences to recover interactable 3D scenes. It automatically discovers articulated parts, estimates their kinematic parameters, tracks their 3D motion, and reconstructs static and moving geometry in canonical space, yielding simulation-compatible meshes. Across new real and simulated benchmarks, FunRec surpasses prior work by a large margin, achieving up to +50 mIoU improvement in part segmentation, 5-10 times lower articulation and pose errors, and significantly higher reconstruction accuracy. We further demonstrate applications on URDF/USD export for simulation, hand-guided affordance mapping and robot-scene interaction. --- ### Semantic-Topological Graph Reasoning for Language-Guided Pulmonary Screening Authors: Chenyu Xue, Yiran Liu, Mian Zhou, Jionglong Su, Zhixiang Lu Published: 2026-04-07 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.05620v1 Medical image segmentation driven by free-text clinical instructions is a critical frontier in computer-aided diagnosis. However, existing multimodal and foundation models struggle with the semantic ambiguity of clinical reports and fail to disambiguate complex anatomical overlaps in low-contrast scans. Furthermore, fully fine-tuning these massive architectures on limited medical datasets invariably leads to severe overfitting. To address these challenges, we propose a novel Semantic-Topological Graph Reasoning (STGR) framework for language-guided pulmonary screening. Our approach elegantly synergizes the reasoning capabilities of large language models (LLaMA-3-V) with the zero-shot delineation of vision foundation models (MedSAM). Specifically, we introduce a Text-to-Vision Intent Distillation (TVID) module to extract precise diagnostic guidance. To resolve anatomical ambiguity, we formulate mask selection as a dynamic graph reasoning problem, where candidate lesions are modeled as nodes and edges capture spatial and semantic affinities. To ensure deployment feasibility, we introduce a Selective Asymmetric Fine-Tuning (SAFT) strategy that updates less than 1% of the parameters. Rigorous 5-fold cross-validation on the LIDC-IDRI and LNDb datasets demonstrates that our framework establishes a new state-of-the-art. Notably, it achieves an 81.5% Dice Similarity Coefficient (DSC) on LIDC-IDRI, outperforming leading LLM-based tools like LISA by over 5%. Crucially, our SAFT strategy acts as a powerful regularizer, yielding exceptional cross-fold stability (0.6% DSC variance) and paving the way for robust, context-aware clinical deployment. --- ### Evaluation of Randomization through Style Transfer for Enhanced Domain Generalization Authors: Dustin Eisenhardt, Timothy Schaumlöffel, Alperen Kantarci, Gemma Roig Published: 2026-04-07 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.05616v1 Deep learning models for computer vision often suffer from poor generalization when deployed in real-world settings, especially when trained on synthetic data due to the well-known Sim2Real gap. Despite the growing popularity of style transfer as a data augmentation strategy for domain generalization, the literature contains unresolved contradictions regarding three key design axes: the diversity of the style pool, the role of texture complexity, and the choice of style source. We present a systematic empirical study that isolates and evaluates each of these factors for driving scene understanding, resolving inconsistencies in prior work. Our findings show that (i) expanding the style pool yields larger gains than repeated augmentation with few styles, (ii) texture complexity has no significant effect when the pool is sufficiently large, and (iii) diverse artistic styles outperform domain-aligned alternatives. Guided by these insights, we derive StyleMixDG (Style-Mixing for Domain Generalization), a lightweight, model-agnostic augmentation recipe that requires no architectural modifications or additional losses. Evaluated on the GTAV $\rightarrow$ {BDD100k, Cityscapes, Mapillary Vistas} benchmark, StyleMixDG demonstrates consistent improvements over strong baselines, confirming that the empirically identified design principles translate into practical gains. The code will be released on GitHub. --- ### Same Graph, Different Likelihoods: Calibration of Autoregressive Graph Generators via Permutation-Equivalent Encodings Authors: Laurits Fredsgaard, Aaron Thomas, Michael Riis Andersen, Mikkel N. Schmidt, Mahito Sugiyama Published: 2026-04-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.05613v1 Autoregressive graph generators define likelihoods via a sequential construction process, but these likelihoods are only meaningful if they are consistent across all linearizations of the same graph. Segmented Eulerian Neighborhood Trails (SENT), a recent linearization method, converts graphs into sequences that can be perfectly decoded and efficiently processed by language models, but admit multiple equivalent linearizations of the same graph. We quantify violations in assigned negative log-likelihood (NLL) using the coefficient of variation across equivalent linearizations, which we call Linearization Uncertainty (LU). Training transformers under four linearization strategies on two datasets, we show that biased orderings achieve lower NLL on their native order but exhibit expected calibration error (ECE) two orders of magnitude higher under random permutation, indicating that these models have learned their training linearization rather than the underlying graph. On the molecular graph benchmark QM9, NLL for generated graphs is negatively correlated with molecular stability (AUC $=0.43$), while LU achieves AUC $=0.85$, suggesting that permutation-based evaluation provides a more reliable quality check for generated molecules. Code is available at https://github.com/lauritsf/linearization-uncertainty --- ### INTERACT: An AI-Driven Extended Reality Framework for Accesible Communication Featuring Real-Time Sign Language Interpretation and Emotion Recognition Authors: Nikolaos D. Tantaroudas, Andrew J. McCracken, Ilias Karachalios, Evangelos Papatheou Published: 2026-04-07 Categories: cs.CE, cs.AI, cs.CL, cs.CV, cs.ET arXiv: https://arxiv.org/abs/2604.05605v1 Video conferencing has become central to professional collaboration, yet most platforms offer limited support for deaf, hard-of-hearing, and multilingual users. The World Health Organisation estimates that over 430 million people worldwide require rehabilitation for disabling hearing loss, a figure projected to exceed 700 million by 2050. Conventional accessibility measures remain constrained by high costs, limited availability, and logistical barriers, while Extended Reality (XR) technologies open new possibilities for immersive and inclusive communication. This paper presents INTERACT (Inclusive Networking for Translation and Embodied Real-Time Augmented Communication Tool), an AI-driven XR platform that integrates real-time speech-to-text conversion, International Sign Language (ISL) rendering through 3D avatars, multilingual translation, and emotion recognition within an immersive virtual environment. Built on the CORTEX2 framework and deployed on Meta Quest 3 headsets, INTERACT combines Whisper for speech recognition, NLLB for multilingual translation, RoBERTa for emotion classification, and Google MediaPipe for gesture extraction. Pilot evaluations were conducted in two phases, first with technical experts from academia and industry, and subsequently with members of the deaf community. The trials reported 92% user satisfaction, transcription accuracy above 85%, and 90% emotion-detection precision, with a mean overall experience rating of 4.6 out of 5.0 and 90% of participants willing to take part in further testing. The results highlight strong potential for advancing accessibility across educational, cultural, and professional settings. An extended version of this work, including full pilot data and implementation details, has been published as an Open Research Europe article (Tantaroudas et al., 2026a). --- ### ID-Selection: Importance-Diversity Based Visual Token Selection for Efficient LVLM Inference Authors: Zhaohong Huang, Wenjing Liu, Yuxin Zhang, Fei Chao, Rongrong Ji Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05601v1 Recent advances have explored visual token pruning to accelerate the inference of large vision-language models (LVLMs). However, existing methods often struggle to balance token importance and diversity: importance-based methods tend to retain redundant tokens, whereas diversity-based methods may overlook informative ones. This trade-off becomes especially problematic under high reduction ratios, where preserving only a small subset of visual tokens is critical. To address this issue, we propose ID-Selection, a simple yet effective token selection strategy for efficient LVLM inference. The key idea is to couple importance estimation with diversity-aware iterative selection: each token is first assigned an importance score, after which high-scoring tokens are selected one by one while the scores of similar tokens are progressively suppressed. In this way, ID-Selection preserves informative tokens while reducing redundancy in a unified selection process. Extensive experiments across 5 LVLM backbones and 16 main benchmarks demonstrate that ID-Selection consistently achieves superior performance and efficiency, especially under extreme pruning ratios. For example, on LLaVA-1.5-7B, ID-Selection prunes 97.2% of visual tokens, retaining only 16 tokens, while reducing inference FLOPs by over 97% and preserving 91.8% of the original performance, all without additional training. --- ### Uncovering Linguistic Fragility in Vision-Language-Action Models via Diversity-Aware Red Teaming Authors: Baoshun Tong, Haoran He, Ling Pan, Yang Liu, Liang Lin Published: 2026-04-07 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2604.05595v1 Vision-Language-Action (VLA) models have achieved remarkable success in robotic manipulation. However, their robustness to linguistic nuances remains a critical, under-explored safety concern, posing a significant safety risk to real-world deployment. Red teaming, or identifying environmental scenarios that elicit catastrophic behaviors, is an important step in ensuring the safe deployment of embodied AI agents. Reinforcement learning (RL) has emerged as a promising approach in automated red teaming that aims to uncover these vulnerabilities. However, standard RL-based adversaries often suffer from severe mode collapse due to their reward-maximizing nature, which tends to converge to a narrow set of trivial or repetitive failure patterns, failing to reveal the comprehensive landscape of meaningful risks. To bridge this gap, we propose a novel \textbf{D}iversity-\textbf{A}ware \textbf{E}mbodied \textbf{R}ed \textbf{T}eaming (\textbf{DAERT}) framework, to expose the vulnerabilities of VLAs against linguistic variations. Our design is based on evaluating a uniform policy, which is able to generate a diverse set of challenging instructions while ensuring its attack effectiveness, measured by execution failures in a physical simulator. We conduct extensive experiments across different robotic benchmarks against two state-of-the-art VLAs, including $π_0$ and OpenVLA. Our method consistently discovers a wider range of more effective adversarial instructions that reduce the average task success rate from 93.33\% to 5.85\%, demonstrating a scalable approach to stress-testing VLA agents and exposing critical safety blind spots before real-world deployment. --- ### RABC-Net: Reliability-Aware Annotation-Free Skin Lesion Segmentation for Low-Resource Dermoscopy Authors: Yujie Yao, Yuhaohang He, Junjie Huang, Zhou Liu, Jiangzhao Li, Yan Qiao, Wen Xiao, Yunsen Liang, Xiaofan Li Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05594v2 Pixel-level annotation is costly in low-resource dermoscopy. We present RABC-Net, a reliability-aware annotation-free segmentation system that combines pseudo-label reliability learning, restricted target-domain adaptation, and Reliability-Adaptive Boundary Calibration (RABC). The system decouples reliability learning from deployment: uncertainty-aware pseudo-label interaction shapes robust representations during training, while the image-only inference path is preserved and RABC performs local logit-space calibration from boundary confidence, uncertainty, and foreground probability. No manual masks are used for training or target-domain adaptation; validation labels, when available, are used only for final operating-point selection. Across ISIC-2017, ISIC-2018, and PH2, RABC-Net achieves macro-average DICE/JAC of 86.58\%/79.47\% and consistent matched-protocol results. Controlled within-study analyses show that RABC provides localized gains over nonlearned boundary correction, while the overall result comes from the full reliability-aware system. Adaptation updates only 3.50\% of model parameters, image-only inference runs at 87.4 FPS, and the selected operating points use $σ=0$ on all three datasets, indicating that learned calibration avoids extra smoothing at deployment. --- ### Label Effects: Shared Heuristic Reliance in Trust Assessment by Humans and LLM-as-a-Judge Authors: Xin Sun, Di Wu, Sijing Qin, Isao Echizen, Abdallah El Ali, Saku Sugawara Published: 2026-04-07 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.05593v1 Large language models (LLMs) are increasingly used as automated evaluators (LLM-as-a-Judge). This work challenges its reliability by showing that trust judgments by LLMs are biased by disclosed source labels. Using a counterfactual design, we find that both humans and LLM judges assign higher trust to information labeled as human-authored than to the same content labeled as AI-generated. Eye-tracking data reveal that humans rely heavily on source labels as heuristic cues for judgments. We analyze LLM internal states during judgment. Across label conditions, models allocate denser attention to the label region than the content region, and this label dominance is stronger under Human labels than AI labels, consistent with the human gaze patterns. Besides, decision uncertainty measured by logits is higher under AI labels than Human labels. These results indicate that the source label is a salient heuristic cue for both humans and LLMs. It raises validity concerns for label-sensitive LLM-as-a-Judge evaluation, and we cautiously raise that aligning models with human preferences may propagate human heuristic reliance into models, motivating debiased evaluation and alignment. --- ### AI-Driven Modular Services for Accessible Multilingual Education in Immersive Extended Reality Settings: Integrating Speech Processing, Translation, and Sign Language Rendering Authors: N. D. Tantaroudas, A. J. McCracken, I. Karachalios, E. Papatheou Published: 2026-04-07 Categories: cs.CE, cs.AI, cs.CL, cs.CY, cs.ET arXiv: https://arxiv.org/abs/2604.05591v1 This work introduces a modular platform that brings together six AI services, automatic speech recognition via OpenAI Whisper, multilingual translation through Meta NLLB, speech synthesis using AWS Polly, emotion classification with RoBERTa, dialogue summarisation via flan t5 base samsum, and International Sign (IS) rendering through Google MediaPipe. A corpus of IS gesture recordings was processed to derive hand landmark coordinates, which were subsequently mapped onto three dimensional avatar animations inside a virtual reality (VR) environment. Validation comprised technical benchmarking of each AI component, including comparative assessments of speech synthesis providers and multilingual translation models (NLLB 200 and EuroLLM 1.7B variants). Technical evaluations confirmed the suitability of the platform for real time XR deployment. Speech synthesis benchmarking established that AWS Polly delivers the lowest latency at a competitive price point. The EuroLLM 1.7B Instruct variant attained a higher BLEU score, surpassing NLLB. These findings establish the viability of orchestrating cross modal AI services within XR settings for accessible, multilingual language instruction. The modular design permits independent scaling and adaptation to varied educational contexts, providing a foundation for equitable learning solutions aligned with European Union digital accessibility goals. --- ### Foundations for Agentic AI Investigations from the Forensic Analysis of OpenClaw Authors: Jan Gruber, Jan-Niclas Hilgert Published: 2026-04-07 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2604.05589v1 Agentic Al systems are increasingly deployed as personal assistants and are likely to become a common object of digital investigations. However, little is known about how their internal state and actions can be reconstructed during forensic analysis. Despite growing popularity, systematic forensic approaches for such systems remain largely unexplored. This paper presents an empirical study of OpenClaw a widely used single-agent assistant. We examine OpenClaw's technical design via static code analysis and apply differential forensic analysis to identify recoverable traces across stages of the agent interaction loop. We classify and correlate these traces to assess their investigative value in a systematic way. Based on these observations, we propose an agent artifact taxonomy that captures recurring investigative patterns. Finally, we highlight a foundational challenge for agentic Al forensics: agent-mediated execution introduces an additional layer of abstraction and substantial nondeterminism in trace generation. The large language model (LLM), the execution environment, and the evolving context can influence tool choice and state transitions in ways that are largely absent from rule-based software. Overall, our results provide an initial foundation for the systematic investigation of agentic Al and outline implications for digital forensic practice and future research. --- ### ResearchEVO: An End-to-End Framework for Automated Scientific Discovery and Documentation Authors: Zhe Zhao, Haibin Wen, Jiaming Ma, Jiachang Zhan, Tianyi Xu, Ye Wei, Qingfu Zhang Published: 2026-04-07 Categories: cs.AI, math.OC arXiv: https://arxiv.org/abs/2604.05587v1 An important recurring pattern in scientific breakthroughs is a two-stage process: an initial phase of undirected experimentation that yields an unexpected finding, followed by a retrospective phase that explains why the finding works and situates it within existing theory. We present ResearchEVO, an end-to-end framework that computationally instantiates this discover-then-explain paradigm. The Evolution Phase employs LLM-guided bi-dimensional co-evolution -- simultaneously optimizing both algorithmic logic and overall architecture -- to search the space of code implementations purely by fitness, without requiring any understanding of the solutions it produces. The Writing Phase then takes the best-performing algorithm and autonomously generates a complete, publication-ready research paper through sentence-level retrieval-augmented generation with explicit anti-hallucination verification and automated experiment design. To our knowledge, ResearchEVO is the first system to cover this full pipeline end to end: no prior work jointly performs principled algorithm evolution and literature-grounded scientific documentation. We validate the framework on two cross-disciplinary scientific problems -- Quantum Error Correction using real Google quantum hardware data, and Physics-Informed Neural Networks -- where the Evolution Phase discovered human-interpretable algorithmic mechanisms that had not been previously proposed in the respective domain literatures. In both cases, the Writing Phase autonomously produced compilable LaTeX manuscripts that correctly grounded these blind discoveries in existing theory via RAG, with zero fabricated citations. --- ### Purify-then-Align: Towards Robust Human Sensing under Modality Missing with Knowledge Distillation from Noisy Multimodal Teacher Authors: Pengcheng Weng, Yanyu Qian, Yangxin Xu, Fei Wang Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05584v2 Robust multimodal human sensing must overcome the critical challenge of missing modalities. Two principal barriers are the Representation Gap between heterogeneous data and the Contamination Effect from low-quality modalities. These barriers are causally linked, as the corruption introduced by contamination fundamentally impedes the reduction of representation disparities. In this paper, we propose PTA, a novel "Purify-then-Align" framework that solves this causal dependency through a synergistic integration of meta-learning and knowledge diffusion. To purify the knowledge source, PTA first employs a meta-learning-driven weighting mechanism that dynamically learns to down-weight the influence of noisy, low-contributing modalities. Subsequently, to align different modalities, PTA introduces a diffusion-based knowledge distillation paradigm in which an information-rich clean teacher, formed from this purified consensus, refines the features of each student modality. The ultimate payoff of this "Purify-then-Align" strategy is the creation of exceptionally powerful single-modality encoders imbued with cross-modal knowledge. Comprehensive experiments on the large-scale MM-Fi and XRF55 datasets, under pronounced Representation Gap and Contamination Effect, demonstrate that PTA achieves state-of-the-art performance and significantly improves the robustness of single-modality models in diverse missing-modality scenarios. --- ### WRF4CIR: Weight-Regularized Fine-Tuning Network for Composed Image Retrieval Authors: Yizhuo Xu, Chaojian Yu, Yuanjie Shao, Tongliang Liu, Qinmu Peng, Xinge You Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05583v1 Composed Image Retrieval (CIR) task aims to retrieve target images based on reference images and modification texts. Current CIR methods primarily rely on fine-tuning vision-language pre-trained models. However, we find that these approaches commonly suffer from severe overfitting, posing challenges for CIR with limited triplet data. To better understand this issue, we present a systematic study of overfitting in VLP-based CIR, revealing a significant and previously overlooked generalization gap across different models and datasets. Motivated by these findings, we introduce WRF4CIR, a Weight-Regularized Fine-tuning network for CIR. Specifically, during the fine-tuning process, we apply adversarial perturbations to the model weights for regularization, where these perturbations are generated in the opposite direction of gradient descent. Intuitively, WRF4CIR increases the difficulty of fitting the training data, which helps mitigate overfitting in CIR under limited triplet supervision. Extensive experiments on benchmark datasets demonstrate that WRF4CIR significantly narrows the generalization gap and achieves substantial improvements over existing methods. --- ### High-Resolution Single-Shot Polarimetric Imaging Made Easy Authors: Shuangfan Zhou, Chu Zhou, Heng Guo, Youwei Lyu, Boxin Shi, Zhanyu Ma, Imari Sato Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05581v1 Polarization-based vision has gained increasing attention for providing richer physical cues beyond RGB images. While achieving single-shot capture is highly desirable for practical applications, existing Division-of-Focal-Plane (DoFP) sensors inherently suffer from reduced spatial resolution and artifacts due to their spatial multiplexing mechanism. To overcome these limitations without sacrificing the snapshot capability, we propose EasyPolar, a multi-view polarimetric imaging framework. Our system is grounded in the physical insight that three independent intensity measurements are sufficient to fully characterize linear polarization. Guided by this, we design a triple-camera setup consisting of three synchronized RGB cameras that capture one unpolarized view and two polarized views with distinct orientations. Building upon this hardware design, we further propose a confidence-guided polarization reconstruction network to address the potential misalignment in multi-view fusion. The network performs multi-modal feature fusion under a confidence-aware physical guidance mechanism, which effectively suppresses warping-induced artifacts and enforces explicit geometric constraints on the solution space. Experimental results demonstrate that our method achieves high-quality results and benefits various downstream tasks. --- ### THIVLVC: Retrieval Augmented Dependency Parsing for Latin Authors: Luc Pommeret, Thibault Wagret, Jules Deret Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05564v1 We describe THIVLVC, a two-stage system for the EvaLatin 2026 Dependency Parsing task. Given a Latin sentence, we retrieve structurally similar entries from the CIRCSE treebank using sentence length and POS n-gram similarity, then prompt a large language model to refine the baseline parse from UDPipe using the retrieved examples and UD annotation guidelines. We submit two configurations: one without retrieval and one with retrieval (RAG). On poetry (Seneca), THIVLVC improves CLAS by +17 points over the UDPipe baseline; on prose (Thomas Aquinas), the gain is +1.5 CLAS. A double-blind error analysis of 300 divergences between our system and the gold standard reveals that, among unanimous annotator decisions, 53.3% favour THIVLVC, showing annotation inconsistencies both within and across treebanks. --- ### Physics-Aligned Spectral Mamba: Decoupling Semantics and Dynamics for Few-Shot Hyperspectral Target Detection Authors: Luqi Gong, Qixin Xie, Yue Chen, Ziqiang Chen, Fanda Fan, Shuai Zhao, Chao Li Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05562v1 Meta-learning facilitates few-shot hyperspectral target detection (HTD), but adapting deep backbones remains challenging. Full-parameter fine-tuning is inefficient and prone to overfitting, and existing methods largely ignore the frequency-domain structure and spectral band continuity of hyperspectral data, limiting spectral adaptation and cross-domain generalization.To address these challenges, we propose SpecMamba, a parameter-efficient and frequency-aware framework that decouples stable semantic representation from agile spectral adaptation. Specifically, we introduce a Discrete Cosine Transform Mamba Adapter (DCTMA) on top of frozen Transformer representations. By projecting spectral features into the frequency domain via DCT and leveraging Mamba's linear-complexity state-space recursion, DCTMA explicitly captures global spectral dependencies and band continuity while avoiding the redundancy of full fine-tuning. Furthermore, to address prototype drift caused by limited sample sizes, we design a Prior-Guided Tri-Encoder (PGTE) that allows laboratory spectral priors to guide the optimization of the learnable adapter without disrupting the stable semantic feature space. Finally, a Self-Supervised Pseudo-Label Mapping (SSPLM) strategy is developed for test-time adaptation, enabling efficient decision boundary refinement through uncertainty-aware sampling and dual-path consistency constraints. Extensive experiments on multiple public datasets demonstrate that SpecMamba consistently outperforms state-of-the-art methods in detection accuracy and cross-domain generalization. --- ### Evaluation Before Generation: A Paradigm for Robust Multimodal Sentiment Analysis with Missing Modalities Authors: Rongfei Chen, Tingting Zhang, Xiaoyu Shen, Wei Zhang Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05558v1 The missing modality problem poses a fundamental challenge in multimodal sentiment analysis, significantly degrading model accuracy and generalization in real world scenarios. Existing approaches primarily improve robustness through prompt learning and pre trained models. However, two limitations remain. First, the necessity of generating missing modalities lacks rigorous evaluation. Second, the structural dependencies among multimodal prompts and their global coherence are insufficiently explored. To address these issues, a Prompt based Missing Modality Adaptation framework is proposed. A Missing Modality Evaluator is introduced at the input stage to dynamically assess the importance of missing modalities using pretrained models and pseudo labels, thereby avoiding low quality data imputation. Building on this, a Modality invariant Prompt Disentanglement module decomposes shared prompts into modality specific private prompts to capture intrinsic local correlations and improve representation quality. In addition, a Dynamic Prompt Weighting module computes mutual information based weights from cross attention outputs to adaptively suppress interference from missing modalities. To enhance global consistency, a Multi level Prompt Dynamic Connection module integrates shared prompts with self attention outputs through residual connections, leveraging global prompt priors to strengthen key guidance features. Extensive experiments on three public benchmarks, including CMU MOSI, CMU MOSEI, and CH SIMS, demonstrate that the proposed framework achieves state of the art performance and stable results under diverse missing modality settings. The implementation is available at https://github.com/rongfei-chen/ProMMA --- ### EpiBench: Benchmarking Multi-turn Research Workflows for Multimodal Agents Authors: Xuan Dong, Huanyang Zheng, Tianhao Niu, Zhe Han, Pengzhan Li, Bofei Liu, Zhengyang Liu, Guancheng Li, Qingfu Zhu, Wanxiang Che Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05557v1 Scientific research follows multi-turn, multi-step workflows that require proactively searching the literature, consulting figures and tables, and integrating evidence across papers to align experimental settings and support reproducible conclusions. This joint capability is not systematically assessed in existing benchmarks, which largely under-evaluate proactive search, multi-evidence integration and sustained evidence use over time. In this work, we introduce EpiBench, an episodic multi-turn multimodal benchmark that instantiates short research workflows. Given a research task, agents must navigate across papers over multiple turns, align evidence from figures and tables, and use the accumulated evidence in the memory to answer objective questions that require cross paper comparisons and multi-figure integration. EpiBench introduces a process-level evaluation framework for fine-grained testing and diagnosis of research agents. Our experiments show that even the leading model achieves an accuracy of only 29.23% on the hard split, indicating substantial room for improvement in multi-turn, multi-evidence research workflows, providing an evaluation platform for verifiable and reproducible research agents. --- ### Context-Agent: Dynamic Discourse Trees for Non-Linear Dialogue Authors: Junan Hu, Shudan Guo, Wenqi Liu, Jianhua Yin, Yinwei Wei Published: 2026-04-07 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.05552v2 Large Language Models demonstrate outstanding performance in many language tasks but still face fundamental challenges in managing the non-linear flow of human conversation. The prevalent approach of treating dialogue history as a flat, linear sequence is misaligned with the intrinsically hierarchical and branching structure of natural discourse, leading to inefficient context utilization and a loss of coherence during extended interactions involving topic shifts or instruction refinements. To address this limitation, we introduce Context-Agent, a novel framework that models multi-turn dialogue history as a dynamic tree structure. This approach mirrors the inherent non-linearity of conversation, enabling the model to maintain and navigate multiple dialogue branches corresponding to different topics. Furthermore, to facilitate robust evaluation, we introduce the Non-linear Task Multi-turn Dialogue (NTM) benchmark, specifically designed to assess model performance in long-horizon, non-linear scenarios. Our experiments demonstrate that Context-Agent enhances task completion rates and improves token efficiency across various LLMs, underscoring the value of structured context management for complex, dynamic dialogues. The dataset and code is available at GitHub. --- ### FastDiSS: Few-step Match Many-step Diffusion Language Model on Sequence-to-Sequence Generation--Full Version Authors: Dat Nguyen-Cong, Tung Kieu, Hoang Thanh-Tung Published: 2026-04-07 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.05551v1 Self-conditioning has been central to the success of continuous diffusion language models, as it allows models to correct previous errors. Yet its ability degrades precisely in the regime where diffusion is most attractive for deployment: few-step sampling for fast inference. In this study, we show that when models only have a few denoising steps, inaccurate self-conditioning induces a substantial approximation gap; this mistake compounds across denoising steps and ultimately dominate the sample quality. To address this, we propose a novel training framework that handles these errors during learning by perturbing the self-conditioning signal to match inference noise, improving robustness to prior estimation errors. In addition, we introduce a token-level noise-awareness mechanism that prevents training from saturation, hence improving optimization. Extensive experiments across conditional generation benchmarks demonstrate that our framework surpasses standard continuous diffusion models while providing up to 400x faster inference speed, and remains competitive against other one-step diffusion frameworks. --- ### AutoSOTA: An End-to-End Automated Research System for State-of-the-Art AI Model Discovery Authors: Yu Li, Chenyang Shao, Xinyang Liu, Ruotong Zhao, Peijie Liu, Hongyuan Su, Zhibin Chen, Qinglong Yang, Anjie Xu, Yi Fang, Qingbin Zeng, Tianxing Li et al. Published: 2026-04-07 Categories: cs.CL, cs.CE arXiv: https://arxiv.org/abs/2604.05550v2 Artificial intelligence research increasingly depends on prolonged cycles of reproduction, debugging, and iterative refinement to achieve State-Of-The-Art (SOTA) performance, creating a growing need for systems that can accelerate the full pipeline of empirical model optimization. In this work, we introduce AutoSOTA, an end-to-end automated research system that advances the latest SOTA models published in top-tier AI papers to reproducible and empirically improved new SOTA models. We formulate this problem through three tightly coupled stages: resource preparation and goal setting; experiment evaluation; and reflection and ideation. To tackle this problem, AutoSOTA adopts a multi-agent architecture with eight specialized agents that collaboratively ground papers to code and dependencies, initialize and repair execution environments, track long-horizon experiments, generate and schedule optimization ideas, and supervise validity to avoid spurious gains. We evaluate AutoSOTA on recent research papers collected from eight top-tier AI conferences under filters for code availability and execution cost. Across these papers, AutoSOTA achieves strong end-to-end performance in both automated replication and subsequent optimization. Specifically, it successfully discovers 105 new SOTA models that surpass the original reported methods, averaging approximately five hours per paper. Case studies spanning LLM, NLP, computer vision, time series, and optimization further show that the system can move beyond routine hyperparameter tuning to identify architectural innovation, algorithmic redesigns, and workflow-level improvements. These results suggest that end-to-end research automation can serve not only as a performance optimizer, but also as a new form of research infrastructure that reduces repetitive experimental burden and helps redirect human attention toward higher-level scientific creativity. --- ### Stop Fixating on Prompts: Reasoning Hijacking and Constraint Tightening for Red-Teaming LLM Agents Authors: Yanxu Mao, Peipei Liu, Tiehan Cui, Congying Liu, Mingzhe Xing, Datao You Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05549v2 With the widespread application of LLM-based agents across various domains, their complexity has introduced new security threats. Existing red-team methods mostly rely on modifying user prompts, which lack adaptability to new data and may impact the agent's performance. To address the challenge, this paper proposes the JailAgent framework, which completely avoids modifying the user prompt. Specifically, it implicitly manipulates the agent's reasoning trajectory and memory retrieval with three key stages: Trigger Extraction, Reasoning Hijacking, and Constraint Tightening. Through precise trigger identification, real-time adaptive mechanisms, and an optimized objective function, JailAgent demonstrates outstanding performance in cross-model and cross-scenario environments. --- ### COSMO-Agent: Tool-Augmented Agent for Closed-loop Optimization,Simulation,and Modeling Orchestration Authors: Liyuan Deng, Shujian Deng, Yongkang Chen, Yongkang Dai, Zhihang Zhong, Linyang Li, Xiao Sun, Yilei Shi, Huaxi Huang Published: 2026-04-07 Categories: cs.AI, cs.GR arXiv: https://arxiv.org/abs/2604.05547v1 Iterative industrial design-simulation optimization is bottlenecked by the CAD-CAE semantic gap: translating simulation feedback into valid geometric edits under diverse, coupled constraints. To fill this gap, we propose COSMO-Agent (Closed-loop Optimization, Simulation, and Modeling Orchestration), a tool-augmented reinforcement learning (RL) framework that teaches LLMs to complete the closed-loop CAD-CAE process. Specifically, we cast CAD generation, CAE solving, result parsing, and geometry revision as an interactive RL environment, where an LLM learns to orchestrate external tools and revise parametric geometries until constraints are satisfied. To make this learning stable and industrially usable, we design a multi-constraint reward that jointly encourages feasibility, toolchain robustness, and structured output validity. In addition, we contribute an industry-aligned dataset that covers 25 component categories with executable CAD-CAE tasks to support realistic training and evaluation. Experiments show that COSMO-Agent training substantially improves small open-source LLMs for constraint-driven design, exceeding large open-source and strong closed-source models in feasibility, efficiency, and stability. --- ### Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects Authors: Jun Zhang, Yicheng Ji, Feiyang Ren, Yihang Li, Bowen Zeng, Zonghao Chen, Ke Chen, Lidan Shou, Gang Chen, Huan Li Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05546v2 Large Vision-Language Models (LVLMs) enable sophisticated reasoning over images and videos, yet their inference is hindered by a systemic efficiency barrier known as visual token dominance. This overhead is driven by a multi-regime interplay between high-resolution feature extraction, quadratic attention scaling, and memory bandwidth constraints. We present a systematic taxonomy of efficiency techniques structured around the inference lifecycle, consisting of encoding, prefilling, and decoding. Unlike prior reviews focused on isolated optimizations, we analyze the end-to-end pipeline to reveal how upstream decisions dictate downstream bottlenecks, covering compute-bound visual encoding, the intensive prefilling of massive contexts, and the ''visual memory wall'' in bandwidth-bound decoding. By decoupling the efficiency landscape into the axes of shaping information density, managing long-context attention, and overcoming memory limits, this work provides a structured analysis of how isolated optimizations compose to navigate the trade-off between visual fidelity and system efficiency. The survey concludes by outlining four future frontiers supported by pilot empirical insights, including hybrid compression based on functional unit sensitivity, modality-aware decoding with relaxed verification, progressive state management for streaming continuity, and stage-disaggregated serving through hardware-algorithm co-design. Our literature repository is at https://github.com/SuDIS-ZJU/Efficient-LVLMs-Inference. --- ### Referring-Aware Visuomotor Policy Learning for Closed-Loop Manipulation Authors: Jiahua Ma, Yiran Qin, Xin Wen, Yixiong Li, Yuyu Sun, Yulan Guo, Liang Lin, Ruimao Zhang Published: 2026-04-07 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2604.05544v1 This paper addresses a fundamental problem of visuomotor policy learning for robotic manipulation: how to enhance robustness in out-of-distribution execution errors or dynamically re-routing trajectories, where the model relies solely on the original expert demonstrations for training. We introduce the Referring-Aware Visuomotor Policy (ReV), a closed-loop framework that can adapt to unforeseen circumstances by instantly incorporating sparse referring points provided by a human or a high-level reasoning planner. Specifically, ReV leverages the coupled diffusion heads to preserve standard task execution patterns while seamlessly integrating sparse referring via a trajectory-steering strategy. Upon receiving a specific referring point, the global diffusion head firstly generates a sequence of globally consistent yet temporally sparse action anchors, while identifies the precise temporal position for the referring point within this sequence. Subsequently, the local diffusion head adaptively interpolates adjacent anchors based on the current temporal position for specific tasks. This closed-loop process repeats at every execution step, enabling real-time trajectory replanning in response to dynamic changes in the scene. In practice, rather than relying on elaborate annotations, ReV is trained only by applying targeted perturbations to expert demonstrations. Without any additional data or fine-tuning scheme, ReV achieve higher success rates across challenging simulated and real-world tasks. --- ### Channel-wise Retrieval for Multivariate Time Series Forecasting Authors: Junhyeok Kang, Jun Seo, Soyeon Park, Sangjun Han, Seohui Bae, Hyeokjun Choe, Soonyoung Lee Published: 2026-04-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.05543v2 Multivariate time series forecasting often struggles to capture long-range dependencies due to fixed lookback windows. Retrieval-augmented forecasting addresses this by retrieving historical segments from memory, but existing approaches rely on a channel-agnostic strategy that applies the same references to all variables. This neglects inter-variable heterogeneity, where different channels exhibit distinct periodicities and spectral profiles. We propose CRAFT (Channel-wise retrieval-augmented forecasting), a novel framework that performs retrieval independently for each channel. To ensure efficiency, CRAFT adopts a two-stage pipeline: a sparse relation graph constructed in the time domain prunes irrelevant candidates, and spectral similarity in the frequency domain ranks references, emphasizing dominant periodic components while suppressing noise. Experiments on seven public benchmarks demonstrate that CRAFT outperforms state-of-the-art forecasting baselines, achieving superior accuracy with practical inference efficiency. --- ### EchoAgent: Towards Reliable Echocardiography Interpretation with "Eyes","Hands" and "Minds" Authors: Qin Wang, Zhiqing He, Yu Liu, Bowen Guo, Zeju Li, Miao Zhao, Wenhao Ju, Zhiling Luo, Xianhong Shu, Yi Guo, Yuanyuan Wang Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05541v2 Reliable interpretation of echocardiography (Echo) is crucial for assessing cardiac function, which demands clinicians to synchronously orchestrate multiple capabilities, including visual observation (eyes), manual measurement (hands), and expert knowledge learning and reasoning (minds). While current task-specific deep-learning approaches and multimodal large language models have demonstrated promise in assisting Echo analysis through automated segmentation or reasoning, they remain focused on restricted skills, i.e., eyes-hands or eyes-minds, thereby limiting clinical reliability and utility. To address these issues, we propose EchoAgent, an agentic system tailored for end-to-end Echo interpretation, which achieves a fully coordinated eyes-hands-minds workflow that learns, observes, operates, and reasons like a cardiac sonographer. First, we introduce an expertise-driven cognition engine where our agent can automatically assimilate credible Echo guidelines into a structured knowledge base, thus constructing an Echo-customized mind. Second, we devise a hierarchical collaboration toolkit to endow EchoAgent with eyes-hands, which can automatically parse Echo video streams, identify cardiac views, perform anatomical segmentation, and quantitative measurement. Third, we integrate the perceived multimodal evidence with the exclusive knowledge base into an orchestrated reasoning hub to conduct explainable inferences. We evaluate EchoAgent on CAMUS and MIMIC-EchoQA datasets, which cover 48 distinct echocardiographic views spanning 14 cardiac anatomical regions. Experimental results show that EchoAgent achieves optimal performance across diverse structure analyses, yielding overall accuracy of up to 80.00%. Importantly, EchoAgent empowers a single system with abilities to learn, observe, operate and reason like an echocardiologist, which holds great promise for reliable Echo interpretation. --- ### Learning to Edit Knowledge via Instruction-based Chain-of-Thought Prompting Authors: Jinhu Fu, Yan Bai, Longzhu He, Yihang Lou, Yanxiao Zhao, Li Sun, Sen Su Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05540v1 Large language models (LLMs) can effectively handle outdated information through knowledge editing. However, current approaches face two key limitations: (I) Poor generalization: Most approaches rigidly inject new knowledge without ensuring that the model can use it effectively to solve practical problems. (II) Narrow scope: Current methods focus primarily on structured fact triples, overlooking the diverse unstructured forms of factual information (e.g., news, articles) prevalent in real-world contexts. To address these challenges, we propose a new paradigm: teaching LLMs to edit knowledge via Chain of Thoughts (CoTs) reasoning (CoT2Edit). We first leverage language model agents for both structured and unstructured edited data to generate CoTs, building high-quality instruction data. The model is then trained to reason over edited knowledge through supervised fine-tuning (SFT) and Group Relative Policy Optimization (GRPO). At inference time, we integrate Retrieval-Augmented Generation (RAG) to dynamically retrieve relevant edited facts for real-time knowledge editing. Experimental results demonstrate that our method achieves strong generalization across six diverse knowledge editing scenarios with just a single round of training on three open-source language models. The codes are available at https://github.com/FredJDean/CoT2Edit. --- ### From Large Language Model Predicates to Logic Tensor Networks: Neurosymbolic Offer Validation in Regulated Procurement Authors: Cedric Haufe, Frieder Stolzenburg Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05539v2 We present a neurosymbolic approach, i.e. combine symbolic and subsymbolic artificial intelligence, to validating offer documents in regulated public institutions. We employ a language model to extract information and then aggregate it with an LTN (Logic Tensor Network) to make an auditable decision. In regulated public institutions, decisions must be made in a manner that is both factually correct and legally verifiable. Our neurosymbolic approach allows existing domain-specific knowledge to be linked to the semantic text understanding of language models. The decisions resulting from our pipeline can be justified by predicate values, rule truth values, and corresponding text passages. Our experiments on a real corpus show that the proposed pipeline achieves performance comparable to existing models, but its key advantage lies in its interpretability, modular predicate extraction, and explicit support for XAI (Explainable AI). --- ### A canonical generalization of OBDD Authors: Florent Capelli, YooJung Choi, Stefan Mengel, Martín Muñoz, Guy Van den Broeck Published: 2026-04-07 Categories: cs.AI, cs.DS arXiv: https://arxiv.org/abs/2604.05537v1 We introduce Tree Decision Diagrams (TDD) as a model for Boolean functions that generalizes OBDD. They can be seen as a restriction of structured d-DNNF; that is, d-DNNF that respect a vtree $T$. We show that TDDs enjoy the same tractability properties as OBDD, such as model counting, enumeration, conditioning, and apply, and are more succinct. In particular, we show that CNF formulas of treewidth $k$ can be represented by TDDs of FPT size, which is known to be impossible for OBDD. We study the complexity of compiling CNF formulas into deterministic TDDs via bottom-up compilation and relate the complexity of this approach with the notion of factor width introduced by Bova and Szeider. --- ### Turbulence-like 5/3 spectral scaling in contextual representations of language as a complex system Authors: Zhongxin Yang, Chun Bao, Yuanwei Bin, Xiang I. A. Yang, Shiyi Chen Published: 2026-04-07 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.05536v1 Natural language is a complex system that exhibits robust statistical regularities. Here, we represent text as a trajectory in a high-dimensional embedding space generated by transformer-based language models, and quantify scale-dependent fluctuations along the token sequence using an embedding-step signal. Across multiple languages and corpora, the resulting power spectrum exhibits a robust power law with an exponent close to $5/3$ over an extended frequency range. This scaling is observed consistently in contextual embeddings from both human-written and AI-generated text, but is absent in static word embeddings and is disrupted by randomization of token order. These results show that the observed scaling reflects multiscale, context-dependent organization rather than lexical statistics alone. By analogy with the Kolmogorov spectrum in turbulence, our findings suggest that semantic information is integrated in a scale-free, self-similar manner across linguistic scales, and provide a quantitative, model-agnostic benchmark for studying complex structure in language representations. --- ### SignalClaw: LLM-Guided Evolutionary Synthesis of Interpretable Traffic Signal Control Skills Authors: Da Lei, Feng Xiao, Lu Li, Yuzhan Liu Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05535v1 Traffic signal control TSC requires strategies that are both effective and interpretable for deployment, yet reinforcement learning produces opaque neural policies while program synthesis depends on restrictive domain-specific languages. We present SIGNALCLAW, a framework that uses large language models LLMs as evolutionary skill generators to synthesize and refine interpretable control skills for adaptive TSC. Each skill includes rationale, selection guidance, and executable code, making policies human-inspectable and self-documenting. At each generation, evolution signals from simulation metrics such as queue percentiles, delay trends, and stagnation are translated into natural language feedback to guide improvement. SignalClaw also introduces event-driven compositional evolution: an event detector identifies emergency vehicles, transit priority, incidents, and congestion via TraCI, and a priority dispatcher selects specialized skills. Each skill is evolved independently, and a priority chain enables runtime composition without retraining. We evaluate SignalClaw on routine and event-injected SUMO scenarios against four baselines. On routine scenarios, it achieves average delay of 7.8 to 9.2 seconds, within 3 to 10 percent of the best method, with low variance across random seeds. Under event scenarios, it yields the lowest emergency delay 11.2 to 18.5 seconds versus 42.3 to 72.3 for MaxPressure and 78.5 to 95.3 for DQN, and the lowest transit person delay 9.8 to 11.5 seconds versus 38.7 to 45.2 for MaxPressure. In mixed events, the dispatcher composes skills effectively while maintaining stable overall delay. The evolved skills progress from simple linear rules to conditional strategies with multi-feature interactions, while remaining fully interpretable and directly modifiable by traffic engineers. --- ### Experience Transfer for Multimodal LLM Agents in Minecraft Game Authors: Chenghao Li, Jun Liu, Songbo Zhang, Huadong Jian, Hao Ni, Lik-Hang Lee, Sung-Ho Bae, Guoqing Wang, Yang Yang, Chaoning Zhang Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05533v1 Multimodal LLM agents operating in complex game environments must continually reuse past experience to solve new tasks efficiently. In this work, we propose Echo, a transfer-oriented memory framework that enables agents to derive actionable knowledge from prior interactions rather than treating memory as a passive repository of static records. To make transfer explicit, Echo decomposes reusable knowledge into five dimensions: structure, attribute, process, function, and interaction. This formulation allows the agent to identify recurring patterns shared across different tasks and infer what prior experience remains applicable in new situations. Building on this formulation, Echo leverages In-Context Analogy Learning (ICAL) to retrieve relevant experiences and adapt them to unseen tasks through contextual examples. Experiments in Minecraft show that, under a from-scratch learning setting, Echo achieves a 1.3x to 1.7x speed-up on object-unlocking tasks. Moreover, Echo exhibits a burst-like chain-unlocking phenomenon, rapidly unlocking multiple similar items within a short time interval after acquiring transferable experience. These results suggest that experience transfer is a promising direction for improving the efficiency and adaptability of multimodal LLM agents in complex interactive environments. --- ### Inventory of the 12 007 Low-Dimensional Pseudo-Boolean Landscapes Invariant to Rank, Translation, and Rotation Authors: Arnaud Liefooghe, Sébastien Verel Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05530v1 Many randomized optimization algorithms are rank-invariant, relying solely on the relative ordering of solutions rather than absolute fitness values. We introduce a stronger notion of rank landscape invariance: two problems are equivalent if their ranking, but also their neighborhood structure and symmetries (translation and rotation), induce identical landscapes. This motivates the study of rank landscapes rather than individual functions. While prior work analyzed the rankings of injective function classes in isolation, we provide an exhaustive inventory of the invariant landscape classes for pseudo-Boolean functions of dimensions 1, 2, and 3, including non-injective cases. Our analysis reveals 12,007 classes in total, a significant reduction compared to rank-invariance alone. We find that non-injective functions yield far more invariant landscape classes than injective ones. In addition, complex combinations of topological landscape properties and algorithm behaviors emerge, particularly regarding deceptiveness, neutrality, and the performance of hill-climbing strategies. The inventory serves as a resource for pedagogical purposes and benchmark design, offering a foundation for constructing larger problems with controlled hardness and advancing our understanding of landscape difficulty and algorithm performance. --- ### ActivityEditor: Learning to Synthesize Physically Valid Human Mobility Authors: Chenjie Yang, Yutian Jiang, Anqi Liang, Wei Qi, Chenyu Wu, Junbo Zhang Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05529v2 Human mobility modeling is indispensable for diverse urban applications. However, existing data-driven methods often suffer from data scarcity, limiting their applicability in regions where historical trajectories are unavailable or restricted. To bridge this gap, we propose \textbf{ActivityEditor}, a novel dual-LLM-agent framework designed for zero-shot cross-regional trajectory generation. Our framework decomposes the complex synthesis task into two collaborative stages. Specifically, an intention-based agent, which leverages demographic-driven priors to generate structured human intentions and coarse activity chains to ensure high-level socio-semantic coherence. These outputs are then refined by editor agent to obtain mobility trajectories through iteratively revisions that enforces human mobility law. This capability is acquired through reinforcement learning with multiple rewards grounded in real-world physical constraints, allowing the agent to internalize mobility regularities and ensure high-fidelity trajectory generation. Extensive experiments demonstrate that \textbf{ActivityEditor} achieves superior zero-shot performance when transferred across diverse urban contexts. It maintains high statistical fidelity and physical validity, providing a robust and highly generalizable solution for mobility simulation in data-scarce scenarios. Our code is available at: https://anonymous.4open.science/r/ActivityEditor-066B. --- ### Prior-guided Fusion of Multimodal Features for Change Detection from Optical-SAR Images Authors: Xuanguang Liu, Lei Ding, Yujie Li, Chenguang Dai, Zhenchao Zhang, Mengmeng Li, Ziyi Yang, Yifan Sun, Yongqi Sun, Hanyun Wang, Lorenzo Bruzzone Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05527v2 Multimodal change detection (MMCD) identifies changed areas in multimodal remote sensing data, demonstrating significant application value in land use monitoring and urban sustainable development. However, literature MMCD approaches exhibit limitations in both cross-modal interaction and exploiting modality-specific characteristics. This leads to insufficient modeling of fine-grained change information, thus hindering the precise detection of semantic changes. To address these problems, we propose STSF-Net, a framework designed for MMCD between optical and SAR images. STSF-Net jointly models modality-specific and spatio-temporal common features to enhance change representations. Specifically, modality-specific features are exploited to capture genuine semantic change signals, while spatio-temporal common features are embedded to suppress pseudo-changes caused by differences in imaging mechanisms. Furthermore, we introduce an optical and SAR feature fusion strategy that adaptively adjusts multimodal feature importance based on semantic priors obtained from visual foundation models. Finally, we introduce the novel Delta-SN6 dataset, the first openly-accessible multiclass MMCD benchmark consisting of very-high-resolution fully polarimetric SAR and optical images. Experimental results on Delta-SN6, BRIGHT, and Wuhan datasets demonstrate that our method outperforms the state-of-the-art by 3.21%, 0.87%, and 1.32% in mIoU, respectively. --- ### Controllable Singing Style Conversion with Boundary-Aware Information Bottleneck Authors: Zhetao Hu, Yiquan Zhou, Wenyu Wang, Zhiyu Wu, Xin Gao, Jihua Zhu Published: 2026-04-07 Categories: cs.SD, cs.AI arXiv: https://arxiv.org/abs/2604.05526v1 This paper presents the submission of the S4 team to the Singing Voice Conversion Challenge 2025 (SVCC2025)-a novel singing style conversion system that advances fine-grained style conversion and control within in-domain settings. To address the critical challenges of style leakage, dynamic rendering, and high-fidelity generation with limited data, we introduce three key innovations: a boundary-aware Whisper bottleneck that pools phoneme-span representations to suppress residual source style while preserving linguistic content; an explicit frame-level technique matrix, enhanced by targeted F0 processing during inference, for stable and distinct dynamic style rendering; and a perceptually motivated high-frequency band completion strategy that leverages an auxiliary standard 48kHz SVC model to augment the high-frequency spectrum, thereby overcoming data scarcity without overfitting. In the official SVCC2025 subjective evaluation, our system achieves the best naturalness performance among all submissions while maintaining competitive results in speaker similarity and technique control, despite using significantly less extra singing data than other top-performing systems. Audio samples are available online. ---