# Research Papers — 2026-04 · bundle page 197 of 233 > Full-text markdown of papers #9801–9850 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/197 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/196.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/198.md) ### What Makes a Sale? Simulating End-to-End Seller--Buyer Retail Dynamics with LLM Agents Authors: Jeonghwan Choi, Jibin Hwang, Gyeonghun Sun, Minjeong Ban, Taewon Yun, Hyeonjae Cheon, Hwanjun Song Published: 2026-04-06 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.04468v3 Evaluating retail strategies before deployment is difficult, as outcomes are determined across multiple stages, from seller-side persuasion through buyer-seller interaction to purchase decisions. However, existing retail simulators capture only partial aspects of this process and do not model cross-stage dependencies, making it difficult to assess how early decisions affect downstream outcomes. We present RetailSim, an end-to-end retail simulation framework that models this pipeline in a unified environment, explicitly designed for simulation fidelity through diverse product spaces, persona-driven agents, and multi-turn interactions. We evaluate RetailSim with a dual protocol comprising human evaluation of behavioral fidelity and meta-evaluation against real-world economic regularities, showing that it successfully reproduces key patterns such as demographic purchasing behavior, the price-demand relationship, and heterogeneous price elasticity. We further demonstrate its practical utility via decision-oriented use cases, including persona inference, seller-buyer interaction analysis, and sales strategy evaluation, showing RetailSim's potential as a controlled testbed for exploring retail strategies. --- ### Group-DINOmics: Incorporating People Dynamics into DINO for Self-supervised Group Activity Feature Learning Authors: Ryuki Tezuka, Chihiro Nakatani, Norimichi Ukita Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04467v1 This paper proposes Group Activity Feature (GAF) learning without group activity annotations. Unlike prior work, which uses low-level static local features to learn GAFs, we propose leveraging dynamics-aware and group-aware pretext tasks, along with local and global features provided by DINO, for group-dynamics-aware GAF learning. To adapt DINO and GAF learning to local dynamics and global group features, our pretext tasks use person flow estimation and group-relevant object location estimation, respectively. Person flow estimation is used to represent the local motion of each person, which is an important cue for understanding group activities. In contrast, group-relevant object location estimation encourages GAFs to learn scene context (e.g., spatial relations of people and objects) as global features. Comprehensive experiments on public datasets demonstrate the state-of-the-art performance of our method in group activity retrieval and recognition. Our ablation studies verify the effectiveness of each component in our method. Code: https://github.com/tezuka0001/Group-DINOmics. --- ### The Topology of Multimodal Fusion: Why Current Architectures Fail at Creative Cognition Authors: Xiujiang Tan Published: 2026-04-06 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.04465v2 This paper identifies a structural limitation in current multimodal AI architectures that is topological rather than parametric. Contrastive alignment (CLIP), cross-attention fusion (GPT-4V/Gemini), and diffusion-based generation share a common geometric prior -- modal separability -- which we term contact topology. The argument rests on three pillars with philosophy as the generative center. The philosophical pillar reinterprets Wittgenstein's saying/showing distinction as a problem rather than a conclusion: where Wittgenstein chose silence, the Chinese craft epistemology tradition responded with xiang (operative schema) -- the third state emerging when saying and showing interpenetrate. A cruciform framework (dao/qi x saying/showing) positions xiang at the intersection, executing dual huacai (transformation-and-cutting) along both axes. This generates a dual-layer dynamics: chuanghua (creative transformation as spontaneous event) and huacai (its institutionalization into repeatable form). The cognitive science pillar reinterprets DMN/ECN/SN tripartite co-activation through the pathological mirror: overlap isomorphism vs. superimposition collapse in a 2D parameter space (coupling intensity x regulatory capacity). The mathematical pillar formalizes these via fiber bundles and Yang-Mills curvature, with the cruciform structure mapped to fiber bundle language. We propose UOO implementation via Neural ODEs with topological regularization, the ANALOGY-MM benchmark with error-type-ratio metric, and the META-TOP three-tier benchmark testing cross-civilizational topological isomorphism across seven archetypes. A phased experimental roadmap with explicit termination criteria ensures clean exit if falsified. --- ### DP-OPD: Differentially Private On-Policy Distillation for Language Models Authors: Fatemeh Khadem, Sajad Mousavi, Yi Fang, Yuhong Liu Published: 2026-04-06 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.04461v1 Large language models (LLMs) are increasingly adapted to proprietary and domain-specific corpora that contain sensitive information, creating a tension between formal privacy guarantees and efficient deployment through model compression. Differential privacy (DP), typically enforced via DP-SGD, provides record-level protection but often incurs substantial utility loss in autoregressive generation, where optimization noise can amplify exposure bias and compounding errors along long rollouts. Existing approaches to private distillation either apply DP-SGD to both teacher and student, worsening computation and the privacy--utility tradeoff, or rely on DP synthetic text generation from a DP-trained teacher, avoiding DP on the student at the cost of DP-optimizing a large teacher and introducing an offline generation pipeline. We propose \textbf{Differentially Private On-Policy Distillation (DP-OPD)}, a synthesis-free framework that enforces privacy solely through DP-SGD on the student while leveraging a frozen teacher to provide dense token-level targets on \emph{student-generated} trajectories. DP-OPD instantiates this idea via \emph{private generalized knowledge distillation} on continuation tokens. Under a strict privacy budget ($\varepsilon=2.0$), DP-OPD improves perplexity over DP fine-tuning and off-policy DP distillation, and outperforms synthesis-based DP distillation (Yelp: 44.15$\rightarrow$41.68; BigPatent: 32.43$\rightarrow$30.63), while substantially simplifying the training pipeline. In particular, \textbf{DP-OPD collapses private compression into a single DP student-training loop} by eliminating DP teacher training and offline synthetic text generation. Code will be released upon publication at https://github.com/khademfatemeh/dp_opd. --- ### Empirical Characterization of Rationale Stability Under Controlled Perturbations for Explainable Pattern Recognition Authors: Abu Noman Md Sakib, Zhensen Wang, Merjulah Roby, Zijie Zhang Published: 2026-04-06 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.04456v1 Reliable pattern recognition systems should exhibit consistent behavior across similar inputs, and their explanations should remain stable. However, most Explainable AI evaluations remain instance centric and do not explicitly quantify whether attribution patterns are consistent across samples that share the same class or represent small variations of the same input. In this work, we propose a novel metric aimed at assessing the consistency of model explanations, ensuring that models consistently reflect the intended objectives and consistency under label-preserving perturbations. We implement this metric using a pre-trained BERT model on the SST-2 sentiment analysis dataset, with additional robustness tests on RoBERTa, DistilBERT, and IMDB, applying SHAP to compute feature importance for various test samples. The proposed metric quantifies the cosine similarity of SHAP values for inputs with the same label, aiming to detect inconsistent behaviors, such as biased reliance on certain features or failure to maintain consistent reasoning for similar predictions. Through a series of experiments, we evaluate the ability of this metric to identify misaligned predictions and inconsistencies in model explanations. These experiments are compared against standard fidelity metrics to assess whether the new metric can effectively identify when a model's behavior deviates from its intended objectives. The proposed framework provides a deeper understanding of model behavior by enabling more robust verification of rationale stability, which is critical for building trustworthy AI systems. By quantifying whether models rely on consistent attribution patterns for similar inputs, the proposed approach supports more robust evaluation of model behavior in practical pattern recognition pipelines. Our code is publicly available at https://github.com/anmspro/ESS-XAI-Stability. --- ### Generative modeling of granular flow on inclined planes using conditional flow matching Authors: Xuyang Li, Rui Li, Teng Man, Yimin Lu Published: 2026-04-06 Categories: cs.CE, cs.LG arXiv: https://arxiv.org/abs/2604.04453v2 Granular flows govern many natural and industrial processes, yet their interior kinematics and mechanics remain largely unobservable, as experiments access only boundaries or free surfaces. Conventional numerical simulations are computationally expensive for fast inverse reconstruction, and deterministic models tend to collapse to over-smoothed mean predictions in ill-posed settings. This study, to the best of the authors' knowledge, presents the first conditional flow matching (CFM) framework for granular-flow reconstruction from sparse boundary observations. Trained on high-fidelity particle-resolved discrete element simulations, the generative model is guided at inference by a differentiable forward operator and a novel sparsity-aware gradient guidance mechanism. This mechanism avoids the gradient dilution inherent to standard mean-squared-error approaches, preserves the absolute physical scale of observation errors, enforces measurement consistency without hyperparameter tuning, and prevents unphysical velocity predictions in non-material regions. A physics decoder maps the reconstructed velocity fields to stress states and energy fluctuation quantities, including mean stress, deviatoric stress, and granular temperature. The framework accurately recovers interior flow fields from full observation to only 16\% of the informative window, and it remains effective under strongly diluted spatial resolution with only 11% of data. It also outperforms a deterministic CNN baseline in the most ill-posed reconstruction regime and provides spatially resolved uncertainty estimates through ensemble generation. These results demonstrate that conditional generative modeling offers a practical route for non-invasive inference of hidden bulk mechanics in granular media, and it suggests potential applicability for inverse problems in particulate and multiphase systems. --- ### Beyond Few-Step Inference: Accelerating Video Diffusion Transformer Model Serving with Inter-Request Caching Reuse Authors: Hao Liu, Ye Huang, Chenghuan Huang, Zhenyi Zheng, Jiangsu Du, Ziyang Ma, Jing Lyu, Yutong Lu Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04451v1 Video Diffusion Transformer (DiT) models are a dominant approach for high-quality video generation but suffer from high inference cost due to iterative denoising. Existing caching approaches primarily exploit similarity within the diffusion process of a single request to skip redundant denoising steps. In this paper, we introduce Chorus, a caching approach that leverages similarity across requests to accelerate video diffusion model serving. Chorus achieves up to 45\% speedup on industrial 4-step distilled models, where prior intra-request caching approaches are ineffective. Particularly, Chorus employs a three-stage caching strategy along the denoising process. Stage 1 performs full reuse of latent features from similar requests. Stage 2 exploits inter-request caching in specific latent regions during intermediate denoising steps. This stage is combined with Token-Guided Attention Amplification to improve semantic alignment between the generated video and the conditional prompts, thereby extending the applicability of full reuse to later denoising steps. --- ### Conversational Control with Ontologies for Large Language Models: A Lightweight Framework for Constrained Generation Authors: Barbara Gendron, Gaël Guibon, Mathieu d'Aquin Published: 2026-04-06 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.04450v1 Conversational agents based on Large Language Models (LLMs) have recently emerged as powerful tools for human-computer interaction. Nevertheless, their black-box nature implies challenges in predictability and a lack of personalization, both of which can be addressed by controlled generation. This work proposes an end-to-end method to obtain modular and explainable control over LLM outputs through ontological definitions of aspects related to the conversation. Key aspects are modeled and used as constraints; we then further fine-tune the LLM to generate content accordingly. To validate our approach, we explore two tasks that tackle two key conversational aspects: the English proficiency level and the polarity profile of the content. Using a hybrid fine-tuning procedure on seven state-of-the-art, open-weight conversational LLMs, we show that our method consistently outperforms pre-trained baselines, even on smaller models. Beyond quantitative gains, the framework remains model-agnostic, lightweight, and interpretable, enabling reusable control strategies that can be extended to new domains and interaction goals. This approach enhances alignment with strategy instructions and demonstrates the effectiveness of ontology-driven control in conversational systems. --- ### PSY-STEP: Structuring Therapeutic Targets and Action Sequences for Proactive Counseling Dialogue Systems Authors: Jihyun Lee, Yejin Min, Yejin Jeon, SungJun Yang, Hyounghun Kim, Gary Geunbae Lee Published: 2026-04-06 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.04448v1 Cognitive Behavioral Therapy (CBT) aims to identify and restructure automatic negative thoughts pertaining to involuntary interpretations of events, yet existing counseling agents struggle to identify and address them in dialogue settings. To bridge this gap, we introduce STEP, a dataset that models CBT counseling by explicitly reflecting automatic thoughts alongside dynamic, action-level counseling sequences. Using this dataset, we train STEPPER, a counseling agent that proactively elicits automatic thoughts and executes cognitively grounded interventions. To further enhance both decision accuracy and empathic responsiveness, we refine STEPPER through preference learning based on simulated, synthesized counseling sessions. Extensive CBT-aligned evaluations show that STEPPER delivers more clinically grounded, coherent, and personalized counseling compared to other strong baseline models, and achieves higher counselor competence without inducing emotional disruption. --- ### TinyNina: A Resource-Efficient Edge-AI Framework for Sustainable Air Quality Monitoring via Intra-Image Satellite Super-Resolution Authors: Prasanjit Dey, Zachary Yahn, Bianca Schoen-Phelan, Soumyabrata Dev Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.04445v1 Nitrogen dioxide (NO$_2$) is a primary atmospheric pollutant and a significant contributor to respiratory morbidity and urban climate-related challenges. While satellite platforms like Sentinel-2 provide global coverage, their native spatial resolution often limits the precision required, fine-grained NO$_2$ assessment. To address this, we propose TinyNina, a resource-efficient Edge-AI framework specifically engineered for sustainable environmental monitoring. TinyNina implements a novel intra-image learning paradigm that leverages the multi-spectral hierarchy of Sentinel-2 as internal training labels, effectively eliminating the dependency on costly and often unavailable external high-resolution reference datasets. The framework incorporates wavelength-specific attention gates and depthwise separable convolutions to preserve pollutant-sensitive spectral features while maintaining an ultra-lightweight footprint of only 51K parameters. Experimental results, validated against 3,276 matched satellite-ground station pairs, demonstrate that TinyNina achieves a state-of-the-art Mean Absolute Error (MAE) of 7.4 $μ$g/m$^3$. This performance represents a 95% reduction in computational overhead and 47$\times$ faster inference compared to high-capacity models such as EDSR and RCAN. By prioritizing task-specific utility and architectural efficiency, TinyNina provides a scalable, low-latency solution for real-time air quality monitoring in smart city infrastructures. --- ### Parameter-Efficient Semantic Augmentation for Enhancing Open-Vocabulary Object Detection Authors: Weihao Cao, Runqi Wang, Xiaoyue Duan, Jinchao Zhang, Ang Yang, Liping Jing Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04444v2 Open-vocabulary object detection (OVOD) enables models to detect any object category, including unseen ones. Benefiting from large-scale pre-training, existing OVOD methods achieve strong detection performance on general scenarios (e.g., OV-COCO) but suffer severe performance drops when transferred to downstream tasks with substantial domain shifts. This degradation stems from the scarcity and weak semantics of category labels in domain-specific task, as well as the inability of existing models to capture auxiliary semantics beyond coarse-grained category label. To address these issues, we propose HSA-DINO, a parameter-efficient semantic augmentation framework for enhancing open-vocabulary object detection. Specifically, we propose a multi-scale prompt bank that leverages image feature pyramids to capture hierarchical semantics and select domain-specific local semantic prompts, progressively enriching textual representations from coarse to fine-grained levels. Furthermore, we introduce a semantic-aware router that dynamically selects the appropriate semantic augmentation strategy during inference, thereby preventing parameter updates from degrading the generalization ability of the pre-trained OVOD model. We evaluate HSA-DINO on OV-COCO, several vertical domain datasets, and modified benchmark settings. The results show that HSA-DINO performs favorably against previous state-of-the-art methods, achieving a superior trade-off between domain adaptability and open-vocabulary generalization. --- ### DeonticBench: A Benchmark for Reasoning over Rules Authors: Guangyao Dou, Luis Brena, Akhil Deo, William Jurayj, Jingyu Zhang, Nils Holzenberger, Benjamin Van Durme Published: 2026-04-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.04443v1 Reasoning with complex, context-specific rules remains challenging for large language models (LLMs). In legal and policy settings, this manifests as deontic reasoning: reasoning about obligations, permissions, and prohibitions under explicit rules. While many recent benchmarks emphasize short-context mathematical reasoning, fewer focus on long-context, high-stakes deontic reasoning. To address this gap, we introduce DEONTICBENCH, a benchmark of 6,232 tasks across U.S. federal taxes, airline baggage policies, U.S. immigration administration, and U.S. state housing law. These tasks can be approached in multiple ways, including direct reasoning in language or with the aid of symbolic computation. Besides free-form chain-of-thought reasoning, DEONTICBENCH enables an optional solver-based workflow in which models translate statutes and case facts into executable Prolog, leading to formal problem interpretations and an explicit program trace. We release reference Prolog programs for all instances. Across frontier LLMs and coding models, best hard-subset performance reaches only 44.4% on SARA Numeric and 46.6 macro-F1 on Housing. We further study training with supervised fine-tuning and reinforcement learning for symbolic program generation. Although training improves Prolog generation quality, current RL methods still fail to solve these tasks reliably. Overall, DEONTICBENCH provides a benchmark for studying context-grounded rule reasoning in real-world domains under both symbolic and non-symbolic settings. --- ### Explainable Autonomous Cyber Defense using Adversarial Multi-Agent Reinforcement Learning Authors: Yiyao Zhang, Diksha Goel, Hussain Ahmad Published: 2026-04-06 Categories: cs.CR, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2604.04442v1 Autonomous agents are increasingly deployed in both offensive and defensive cyber operations, creating high-speed, closed-loop interactions in critical infrastructure environments. Advanced Persistent Threat (APT) actors exploit "Living off the Land" techniques and targeted telemetry perturbations to induce ambiguity in monitoring systems, causing automated defenses to overreact or misclassify benign behavior as malicious activity. Existing monolithic and multi-agent defense pipelines largely operate on correlation-based signals, lack structural constraints on response actions, and are vulnerable to reasoning drift under ambiguous or adversarial inputs. We present the Causal Multi-Agent Decision Framework (C-MADF), a structurally constrained architecture for autonomous cyber defense that integrates causal modeling with adversarial dual-policy control. C-MADF first learns a Structural Causal Model (SCM) from historical telemetry and compiles it into an investigation-level Directed Acyclic Graph (DAG) that defines admissible response transitions. This roadmap is formalized as a Markov Decision Process (MDP) whose action space is explicitly restricted to causally consistent transitions. Decision-making within this constrained space is performed by a dual-agent reinforcement learning system in which a threat-optimizing Blue-Team policy is counterbalanced by a conservatively shaped Red-Team policy. Inter-policy disagreement is quantified through a Policy Divergence Score and exposed via a human-in-the-loop interface equipped with an Explainability-Transparency Score that serves as an escalation signal under uncertainty. On the real-world CICIoT2023 dataset, C-MADF reduces the false-positive rate from 11.2%, 9.7%, and 8.4% in three cutting-edge literature baselines to 1.8%, while achieving 0.997 precision, 0.961 recall, and 0.979 F1-score. --- ### Training Transformers in Cosine Coefficient Space Authors: Mohamed Amine Bergach Published: 2026-04-06 Categories: cs.PF, cs.AI arXiv: https://arxiv.org/abs/2604.04440v2 Linear layers hold most of a transformer's parameters. We replace each linear layer with one that stores $K$ out of $mn$ two-dimensional DCT coefficients per weight matrix and reconstructs the full matrix through an inverse DCT at every forward pass; the $K$ coefficients are the trainable parameters. A 4-layer, 128-dim transformer trained from scratch on character-level Shakespeare reaches validation loss $1.604$ at $K = mn/2$, against $1.580$ for a standard dense baseline -- a gap of $+0.024$ at half the trainable parameter count, within the terminal-epoch variation of the dense run. A rank-48 LoRA factorization at the same trainable parameter count reaches only $1.801$ ($+0.221$). The structural advantage of sparse-coefficient over low-rank parameterizations at matched $K$ is qualitative. We identify rank flexibility as the mechanism. A random orthonormal basis matches the DCT within noise at $K = mn/2$, and a compression sweep through $K = mn/10$ and $K = mn/20$ shows that subspaces that can host high-rank matrices keep the loss low, while subspaces that flatten into a low-rank block (zigzag-selection variants) converge onto the observed stable rank \emph{and} the loss line of the rank-48 LoRA reference in lock-step. Among these orthonormal bases, the DCT is preferred because its separable fast transform admits a fused reconstruction kernel: the materialized weight matrix never leaves on-chip memory, so the parameter saving translates into a bandwidth saving as well. --- ### Estimating Central, Peripheral, and Temporal Visual Contributions to Human Decision Making in Atari Games Authors: Henrik Krauss, Takehisa Yairi Published: 2026-04-06 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2604.04439v2 We study how different visual information sources contribute to human decision making in dynamic visual environments. Using Atari-HEAD, a large-scale Atari gameplay dataset with synchronized eye-tracking, we introduce a controlled ablation framework as a means to reverse-engineer the contribution of peripheral visual information, explicit gaze information in the form of gaze maps, and past-state information from human behavior. We train action-prediction networks under six settings that selectively include or exclude these information sources. Across 20 games, peripheral information shows by far the strongest contribution, with median prediction-accuracy drops in the range of 35.27-43.90% when removed. Gaze information yields smaller drops of 2.11-2.76%, while past-state information shows a broader range of 1.52-15.51%, with the upper end likely more informative due to reduced peripheral-information leakage. To complement aggregate accuracies, we cluster states by true-action probabilities assigned by the different model configurations. This analysis identifies coarse behavioral regimes, including focus-dominated, periphery-dominated, and more contextual decision situations. These results suggest that human decision making in Atari depends strongly on information beyond the current focus of gaze, while the proposed framework provides a way to estimate such information-source contributions from behavior. --- ### FAVE: Flow-based Average Velocity Establishment for Sequential Recommendation Authors: Ke Shi, Yao Zhang, Feng Guo, Jinyuan Zhang, JunShuo Zhang, Shen Gao, Shuo Shang Published: 2026-04-06 Categories: cs.IR, cs.CL arXiv: https://arxiv.org/abs/2604.04427v1 Generative recommendation has emerged as a transformative paradigm for capturing the dynamic evolution of user intents in sequential recommendation. While flow-based methods improve the efficiency of diffusion models, they remain hindered by the ``Noise-to-Data'' paradigm, which introduces two critical inefficiencies: prior mismatch, where generation starts from uninformative noise, forcing a lengthy recovery trajectory; and linear redundancy, where iterative solvers waste computation on modeling deterministic preference transitions. To address these limitations, we propose a Flow-based Average Velocity Establishment (Fave) framework for one-step generation recommendation that learns a direct trajectory from an informative prior to the target distribution. Fave is structured via a progressive two-stage training strategy. In Stage 1, we establish a stable preference space through dual-end semantic alignment, applying constraints at both the source (user history) and target (next item) to prevent representation collapse. In Stage 2, we directly resolve the efficiency bottlenecks by introducing a semantic anchor prior, which initializes the flow with a masked embedding from the user's interaction history, providing an informative starting point. Then we learn a global average velocity, consolidating the multi-step trajectory into a single displacement vector, and enforce trajectory straightness via a JVP-based consistency constraint to ensure one-step generation. Extensive experiments on three benchmarks demonstrate that Fave not only achieves state-of-the-art recommendation performance but also delivers an order-of-magnitude improvement in inference efficiency, making it practical for latency-sensitive scenarios. --- ### ShieldNet: Network-Level Guardrails against Emerging Supply-Chain Injections in Agentic Systems Authors: Zhuowen Yuan, Zhaorun Chen, Zhen Xiang, Nathaniel D. Bastian, Seyyed Hadi Hashemi, Chaowei Xiao, Wenbo Guo, Bo Li Published: 2026-04-06 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.04426v1 Existing research on LLM agent security mainly focuses on prompt injection and unsafe input/output behaviors. However, as agents increasingly rely on third-party tools and MCP servers, a new class of supply-chain threats has emerged, where malicious behaviors are embedded in seemingly benign tools, silently hijacking agent execution, leaking sensitive data, or triggering unauthorized actions. Despite their growing impact, there is currently no comprehensive benchmark for evaluating such threats. To bridge this gap, we introduce SC-Inject-Bench, a large-scale benchmark comprising over 10,000 malicious MCP tools grounded in a taxonomy of 25+ attack types derived from MITRE ATT&CK targeting supply-chain threats. We observe that existing MCP scanners and semantic guardrails perform poorly on this benchmark. Motivated by this finding, we propose ShieldNet, a network-level guardrail framework that detects supply-chain poisoning by observing real network interactions rather than surface-level tool traces. ShieldNet integrates a man-in-the-middle (MITM) proxy and an event extractor to identify critical network behaviors, which are then processed by a lightweight classifier for attack detection. Extensive experiments show that ShieldNet achieves strong detection performance (up to 0.995 F-1 with only 0.8% false positives) while introducing little runtime overhead, substantially outperforming existing MCP scanners and LLM-based guardrails. --- ### HandDreamer: Zero-Shot Text to 3D Hand Model Generation using Corrective Hand Shape Guidance Authors: Green Rosh, Prateek Kukreja, Vishakha SR, Pawan Prasad B H Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04425v1 The emergence of virtual reality has necessitated the generation of detailed and customizable 3D hand models for interaction in the virtual world. However, the current methods for 3D hand model generation are both expensive and cumbersome, offering very little customizability to the users. While recent advancements in zero-shot text-to-3D synthesis have enabled the generation of diverse and customizable 3D models using Score Distillation Sampling (SDS), they do not generalize very well to 3D hand model generation, resulting in unnatural hand structures, view-inconsistencies and loss of details. To address these limitations, we introduce HandDreamer, the first method for zero-shot 3D hand model generation from text prompts. Our findings suggest that view-inconsistencies in SDS is primarily caused due to the ambiguity in the probability landscape described by the text prompt, resulting in similar views converging to different modes of the distribution. This is particularly aggravated for hands due to the large variations in articulations and poses. To alleviate this, we propose to use MANO hand model based initialization and a hand skeleton guided diffusion process to provide a strong prior for the hand structure and to ensure view and pose consistency. Further, we propose a novel corrective hand shape guidance loss to ensure that all the views of the 3D hand model converges to view-consistent modes, without leading to geometric distortions. Extensive evaluations demonstrate the superiority of our method over the state-of-the-art methods, paving a new way forward in 3D hand model generation. --- ### Is Prompt Selection Necessary for Task-Free Online Continual Learning? Authors: Seoyoung Park, Haemin Lee, Hankook Lee Published: 2026-04-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.04420v1 Task-free online continual learning has recently emerged as a realistic paradigm for addressing continual learning in dynamic, real-world environments, where data arrive in a non-stationary stream without clear task boundaries and can only be observed once. To consider such challenging scenarios, many recent approaches have employed prompt selection, an adaptive strategy that selects prompts from a pool based on input signals. However, we observe that such selection strategies often fail to select appropriate prompts, yielding suboptimal results despite additional training of key parameters. Motivated by this observation, we propose a simple yet effective SinglePrompt that eliminates the need for prompt selection and focuses on classifier optimization. Specifically, we simply (i) inject a single prompt into each self-attention block, (ii) employ a cosine similarity-based logit design to alleviate the forgetting effect inherent in the classifier weights, and (iii) mask logits for unexposed classes in the current minibatch. With this simple task-free design, our framework achieves state-of-the-art performance across various online continual learning benchmarks. Source code is available at https://github.com/efficient-learning-lab/SinglePrompt. --- ### BoxComm: Benchmarking Category-Aware Commentary Generation and Narration Rhythm in Boxing Authors: Kaiwen Wang, Kaili Zheng, Rongrong Deng, Yiming Shi, Chenyi Guo, Ji Wu Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04419v1 Recent multimodal large language models (MLLMs) have shown strong capabilities in general video understanding, driving growing interest in automatic sports commentary generation. However, existing benchmarks for this task focus exclusively on team sports such as soccer and basketball, leaving combat sports entirely unexplored. Notably, combat sports present distinct challenges: critical actions unfold within milliseconds with visually subtle yet semantically decisive differences, and professional commentary contains a substantially higher proportion of tactical analysis compared to team sports. In this paper, we present BoxComm, a large-scale dataset comprising 445 World Boxing Championship match videos with over 52K commentary sentences from professional broadcasts. We propose a structured commentary taxonomy that categorizes each sentence into play-by-play, tactical, or contextual, providing the first category-level annotation for sports commentary benchmarks. Building on this taxonomy, we introduce two novel and complementary evaluations tailored to sports commentary generation: (1) category-conditioned generation, which evaluates whether models can produce accurate commentary of a specified type given video context; and (2) commentary rhythm assessment, which measures whether freely generated commentary exhibits appropriate temporal pacing and type distribution over continuous video segments, capturing a dimension of commentary competence that prior benchmarks have not addressed. Experiments on multiple state-of-the-art MLLMs reveal that current models struggle on both evaluations. We further propose EIC-Gen, an improved baseline incorporating detected punch events to supply structured action cues, yielding consistent gains and highlighting the importance of perceiving fleeting and subtle events for combat sports commentary. --- ### Justified or Just Convincing? Error Verifiability as a Dimension of LLM Quality Authors: Xiaoyuan Zhu, Kimberly Le Truong, Riccardo Fogliato, Gokul Swamy, Weijian Zhang, Minglai Yang, Longtian Ye, Bangya Liu, Minghao Liu, Andrew Ilyas, Steven Wu Published: 2026-04-06 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2604.04418v2 As LLMs are deployed in high-stakes settings, users must judge the correctness of individual responses, often relying on model-generated justifications such as reasoning chains or explanations. Yet, no standard measure exists for whether these justifications help users distinguish correct answers from incorrect ones. We formalize this idea as error verifiability and propose $v_{\text{bal}}$, a balanced metric that measures whether justifications enable raters to accurately assess answer correctness, validated against human raters who show high agreement. We find that neither common approaches, such as post-training and model scaling, nor more targeted interventions recommended improve verifiability. We introduce two methods that succeed at improving verifiability: reflect-and-rephrase (RR) for mathematical reasoning and oracle-rephrase (OR) for factual QA, both of which improve verifiability by incorporating domain-appropriate external information. Together, our results establish error verifiability as a distinct dimension of response quality that does not emerge from accuracy improvements alone and requires dedicated, domain-aware methods to address. --- ### STEER: Structured Event Evidence for Video Reasoning via Multi-Objective Reinforcement Learning Authors: Zinuo Li, Yongxin Guo, Jun Liu, Jiawei Zhan, Xi Jiang, Chengjie Wang, Mohammed Bennamoun, Farid Boussaid, Feng Zheng, Qiuhong Ke Published: 2026-04-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.04415v3 Human understanding of video dynamics relies on forming structured representations of entities, actions, and temporal relations before engaging in abstract reasoning. In contrast, existing Video-LLMs apply unstructured chain-of-thought directly to raw visual tokens, where critical temporal cues are buried in verbose narration and event-level structure is largely overlooked. We propose Structured Event Evidence, which represents a video as a compact, time-ordered event schema capturing salient events with key attributes and inter-event temporal dependencies, enabling evidence-grounded reasoning through a constrained verification process. This design promotes concise, interpretable reasoning while reducing the drift typical of unconstrained chain-of-thought. To train models under this paradigm, we introduce STEER-60K, a dataset with a four-stage progressive pipeline: evidence training, format warm-start, thinking warm-start, and RL post-training. During RL, CoT length and task accuracy often conflict while rewards for hard samples are too sparse, causing the policy to neglect challenging instances. We formulate this as a multi-objective Pareto optimality problem and propose Pareto-Frontier guided Advantage Balancing (P-FAB), which dynamically resolves reward conflicts and identifies balanced optimization directions along the Pareto frontier. The resulting model STEER-4B rivals 7B-scale baselines on video understanding tasks with half the input frames Code and data will be released. --- ### Eliminating Vendor Lock-In in Quantum Machine Learning via Framework-Agnostic Neural Networks Authors: Poornima Kumaresan, Shwetha Singaravelu, Lakshmi Rajendran, Santhosh Sivasubramani Published: 2026-04-06 Categories: cs.ET, cs.LG, quant-ph arXiv: https://arxiv.org/abs/2604.04414v1 Quantum machine learning (QML) stands at the intersection of quantum computing and artificial intelligence, offering the potential to solve problems that remain intractable for classical methods. However, the current landscape of QML software frameworks suffers from severe fragmentation: models developed in TensorFlow Quantum cannot execute on PennyLane backends, circuits authored in Qiskit Machine Learning cannot be deployed to Amazon Braket hardware, and researchers who invest in one ecosystem face prohibitive switching costs when migrating to another. This vendor lock-in impedes reproducibility, limits hardware access, and slows the pace of scientific discovery. In this paper, we present a framework-agnostic quantum neural network (QNN) architecture that abstracts away vendor-specific interfaces through a unified computational graph, a hardware abstraction layer (HAL), and a multi-framework export pipeline. The core architecture supports simultaneous integration with TensorFlow, PyTorch, and JAX as classical co-processors, while the HAL provides transparent access to IBM Quantum, Amazon Braket, Azure Quantum, IonQ, and Rigetti backends through a single application programming interface (API). We introduce three pluggable data encoding strategies (amplitude, angle, and instantaneous quantum polynomial encoding) that are compatible with all supported backends. An export module leveraging Open Neural Network Exchange (ONNX) metadata enables lossless circuit translation across Qiskit, Cirq, PennyLane, and Braket representations. We benchmark our framework on the Iris, Wine, and MNIST-4 classification tasks, demonstrating training time parity (within 8\% overhead) compared to native framework implementations, while achieving identical classification accuracy. --- ### Responses Fall Short of Understanding: Revealing the Gap between Internal Representations and Responses in Visual Document Understanding Authors: Haruka Kawasaki, Ryota Tanaka, Kyosuke Nishida Published: 2026-04-06 Categories: cs.CL, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2604.04411v1 Visual document understanding (VDU) is a challenging task for large vision language models (LVLMs), requiring the integration of visual perception, text recognition, and reasoning over structured layouts. Although recent LVLMs have shown progress on VDU benchmarks, their performance is typically evaluated based on generated responses, which may not necessarily reflect whether the model has actually captured the required information internally. In this paper, we investigate how information required to solve VDU tasks is represented across different layers of LLMs within LVLMs using linear probing. Our study reveals that (1) there is a clear gap between internal representations and generated responses, and (2) information required to solve the task is often encoded more linearly from intermediate layers than from the final layer. Motivated by these findings, we explore fine-tuning strategies that target intermediate layers. Experiments show that fine-tuning intermediate layers improves both linear probing accuracy and response accuracy while narrowing the gap. --- ### Relative Density Ratio Optimization for Stable and Statistically Consistent Model Alignment Authors: Hiroshi Takahashi, Tomoharu Iwata, Atsutoshi Kumagai, Sekitoshi Kanai, Masanori Yamada, Kosuke Nishida, Kazutoshi Shinoda Published: 2026-04-06 Categories: cs.LG, cs.AI, cs.CL, stat.ML arXiv: https://arxiv.org/abs/2604.04410v2 Aligning language models with human preferences is essential for ensuring their safety and reliability. Although most existing approaches assume specific human preference models such as the Bradley-Terry model, this assumption may fail to accurately capture true human preferences, and consequently, these methods lack statistical consistency, i.e., the guarantee that language models converge to the true human preference as the number of samples increases. In contrast, direct density ratio optimization (DDRO) achieves statistical consistency without assuming any human preference models. DDRO models the density ratio between preferred and non-preferred data distributions using the language model, and then optimizes it via density ratio estimation. However, this density ratio is unstable and often diverges, leading to training instability of DDRO. In this paper, we propose a novel alignment method that is both stable and statistically consistent. Our approach is based on the relative density ratio between the preferred data distribution and a mixture of the preferred and non-preferred data distributions. Our approach is stable since this relative density ratio is bounded above and does not diverge. Moreover, it is statistically consistent and yields significantly tighter convergence guarantees than DDRO. We experimentally show its effectiveness with Qwen 2.5 and Llama 3. --- ### Relative Density Ratio Optimization for Stable and Statistically Consistent Model Alignment Authors: Hiroshi Takahashi, Tomoharu Iwata, Atsutoshi Kumagai, Sekitoshi Kanai, Masanori Yamada, Kosuke Nishida, Kazutoshi Shinoda Published: 2026-04-06 Categories: cs.LG, cs.AI, cs.CL, stat.ML arXiv: https://arxiv.org/abs/2604.04410v1 Aligning language models with human preferences is essential for ensuring their safety and reliability. Although most existing approaches assume specific human preference models such as the Bradley-Terry model, this assumption may fail to accurately capture true human preferences, and consequently, these methods lack statistical consistency, i.e., the guarantee that language models converge to the true human preference as the number of samples increases. In contrast, direct density ratio optimization (DDRO) achieves statistical consistency without assuming any human preference models. DDRO models the density ratio between preferred and non-preferred data distributions using the language model, and then optimizes it via density ratio estimation. However, this density ratio is unstable and often diverges, leading to training instability of DDRO. In this paper, we propose a novel alignment method that is both stable and statistically consistent. Our approach is based on the relative density ratio between the preferred data distribution and a mixture of the preferred and non-preferred data distributions. Our approach is stable since this relative density ratio is bounded above and does not diverge. Moreover, it is statistically consistent and yields significantly tighter convergence guarantees than DDRO. We experimentally show its effectiveness with Qwen 2.5 and Llama 3. --- ### NAIMA: Semantics Aware RGB Guided Depth Super-Resolution Authors: Tayyab Nasir, Daochang Liu, Ajmal Mian Published: 2026-04-06 Categories: eess.IV, cs.CV, cs.LG, cs.MM arXiv: https://arxiv.org/abs/2604.04407v2 Guided depth super-resolution (GDSR) is a multi-modal approach for depth map super-resolution that relies on a low-resolution depth map and a high-resolution RGB image to restore finer structural details. However, the misleading color and texture cues indicating depth discontinuities in RGB images often lead to artifacts and blurred depth boundaries in the generated depth map. Recent methods counter this by drawing priors from large pretrained models, but these priors enter the network as decoded predictions such as relative depth, surface normal, or segmentation maps, coupling restoration quality to the accuracy of the decoded prior and requiring auxiliary objectives. We propose a solution that introduces global contextual semantic priors, generated from pretrained vision transformer token embeddings, injecting them directly into the depth branch. Our Guided Token Attention (GTA) module lets multi-level depth encodings act as queries of cross-attention over semantic tokens drawn from progressively deeper layers of the pretrained visual transformer (ViT), scaled by a zero-initialized gate that admits semantic evidence only to the extent that it reduces reconstruction error. The resulting token-to-depth correspondence is aligned implicitly, through learned attention under a single reconstruction loss, rather than through an explicit alignment or distribution-matching objective. Building on this, we present Neural Attention with Implicit Multi-token Alignment (NAIMA), which, to the best of our knowledge, is the first GDSR framework guided by undecoded semantic tokens. NAIMA remains competitive in-distribution while achieving the strongest cross-dataset generalization. --- ### NAIMA: Semantics Aware RGB Guided Depth Super-Resolution Authors: Tayyab Nasir, Daochang Liu, Ajmal Mian Published: 2026-04-06 Categories: eess.IV, cs.CV, cs.LG, cs.MM arXiv: https://arxiv.org/abs/2604.04407v1 Guided depth super-resolution (GDSR) is a multi-modal approach for depth map super-resolution that relies on a low-resolution depth map and a high-resolution RGB image to restore finer structural details. However, the misleading color and texture cues indicating depth discontinuities in RGB images often lead to artifacts and blurred depth boundaries in the generated depth map. We propose a solution that introduces global contextual semantic priors, generated from pretrained vision transformer token embeddings. Our approach to distilling semantic knowledge from pretrained token embeddings is motivated by their demonstrated effectiveness in related monocular depth estimation tasks. We introduce a Guided Token Attention (GTA) module, which iteratively aligns encoded RGB spatial features with depth encodings, using cross-attention for selectively injecting global semantic context extracted from different layers of a pretrained vision transformer. Additionally, we present an architecture called Neural Attention for Implicit Multi-token Alignment (NAIMA), which integrates DINOv2 with GTA blocks for a semantics-aware GDSR. Our proposed architecture, with its ability to distill semantic knowledge, achieves significant improvements over existing methods across multiple scaling factors and datasets. --- ### 3D-Fixer: Coarse-to-Fine In-place Completion for 3D Scenes from a Single Image Authors: Ze-Xin Yin, Liu Liu, Xinjie Wang, Wei Sui, Zhizhong Su, Jian Yang, Jin Xie Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04406v1 Compositional 3D scene generation from a single view requires the simultaneous recovery of scene layout and 3D assets. Existing approaches mainly fall into two categories: feed-forward generation methods and per-instance generation methods. The former directly predict 3D assets with explicit 6DoF poses through efficient network inference, but they generalize poorly to complex scenes. The latter improve generalization through a divide-and-conquer strategy, but suffer from time-consuming pose optimization. To bridge this gap, we introduce 3D-Fixer, a novel in-place completion paradigm. Specifically, 3D-Fixer extends 3D object generative priors to generate complete 3D assets conditioned on the partially visible point cloud at the original locations, which are cropped from the fragmented geometry obtained from the geometry estimation methods. Unlike prior works that require explicit pose alignment, 3D-Fixer uses fragmented geometry as a spatial anchor to preserve layout fidelity. At its core, we propose a coarse-to-fine generation scheme to resolve boundary ambiguity under occlusion, supported by a dual-branch conditioning network and an Occlusion-Robust Feature Alignment (ORFA) strategy for stable training. Furthermore, to address the data scarcity bottleneck, we present ARSG-110K, the largest scene-level dataset to date, comprising over 110K diverse scenes and 3M annotated images with high-fidelity 3D ground truth. Extensive experiments show that 3D-Fixer achieves state-of-the-art geometric accuracy, which significantly outperforms baselines such as MIDI and Gen3DSR, while maintaining the efficiency of the diffusion process. Code and data will be publicly available at https://zx-yin.github.io/3dfixer. --- ### MolDA: Molecular Understanding and Generation via Large Language Diffusion Model Authors: Seohyeon Shin, HanJun Choi, Jun-Hyung Park, Hong Kook Kim, Mansu Kim Published: 2026-04-06 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.04403v2 Large Language Models (LLMs) have significantly advanced molecular discovery, but existing multimodal molecular architectures fundamentally rely on autoregressive (AR) backbones. This strict left-to-right inductive bias is sub-optimal for generating chemically valid molecules, as it struggles to account for non-local global constraints (e.g., ring closures) and often accumulates structural errors during sequential generation. To address these limitations, we propose MolDA (Molecular language model with masked Diffusion with mAsking), a novel multimodal framework that replaces the conventional AR backbone with a discrete Large Language Diffusion Model. MolDA extracts comprehensive structural representations using a hybrid graph encoder, which captures both local and global topologies, and aligns them into the language token space via a Q-Former. Furthermore, we mathematically reformulate Molecular Structure Preference Optimization specifically for the masked diffusion. Through bidirectional iterative denoising, MolDA ensures global structural coherence, chemical validity, and robust reasoning across molecule generation, captioning, and property prediction. --- ### UENR-600K: A Large-Scale Physically Grounded Dataset for Nighttime Video Deraining Authors: Pei Yang, Hai Ci, Beibei Lin, Yiren Song, Mike Zheng Shou Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04402v1 Nighttime video deraining is uniquely challenging because raindrops interact with artificial lighting. Unlike daytime white rain, nighttime rain takes on various colors and appears locally illuminated. Existing small-scale synthetic datasets rely on 2D rain overlays and fail to capture these physical properties, causing models to generalize poorly to real-world night rain. Meanwhile, capturing real paired nighttime videos remains impractical because rain effects cannot be isolated from other degradations like sensor noise. To bridge this gap, we introduce UENR-600K, a large-scale, physically grounded dataset containing 600,000 1080p frame pairs. We utilize Unreal Engine to simulate rain as 3D particles within virtual environments. This approach guarantees photorealism and physically real raindrops, capturing correct details like color refractions, scene occlusions, rain curtains. Leveraging this high-quality data, we establish a new state-of-the-art baseline by adapting the Wan 2.2 video generation model. Our baseline treat deraining as a video-to-video generation task, exploiting strong generative priors to almost entirely bridge the sim-to-real gap. Extensive benchmarking demonstrates that models trained on our dataset generalize significantly better to real-world videos. Project page: https://showlab.github.io/UENR-600K/. --- ### ReinVBC: A Model-based Reinforcement Learning Approach to Vehicle Braking Controller Authors: Haoxin Lin, Junjie Zhou, Daheng Xu, Yang Yu Published: 2026-04-06 Categories: cs.RO, cs.LG, eess.SY arXiv: https://arxiv.org/abs/2604.04401v1 Braking system, the key module to ensure the safety and steer-ability of current vehicles, relies on extensive manual calibration during production. Reducing labor and time consumption while maintaining the Vehicle Braking Controller (VBC) performance greatly benefits the vehicle industry. Model-based methods in offline reinforcement learning, which facilitate policy exploration within a data-driven dynamics model, offer a promising solution for addressing real-world control tasks. This work proposes ReinVBC, which applies an offline model-based reinforcement learning approach to deal with the vehicle braking control problem. We introduce useful engineering designs into the paradigm of model learning and utilization to obtain a reliable vehicle dynamics model and a capable braking policy. Several results demonstrate the capability of our method in real-world vehicle braking and its potential to replace the production-grade anti-lock braking system. --- ### GUIDE: Interpretable GUI Agent Evaluation via Hierarchical Diagnosis Authors: Yuwen Zhai, Runze Li, Liang Wang, Nian Shi, Liwu Xu, Wei Zhang, Ran Lin, Bo Xu, Benlei Cui Published: 2026-04-06 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.04399v1 Evaluating GUI agents presents a distinct challenge: trajectories are long, visually grounded, and open-ended, yet evaluation must be both accurate and interpretable. Existing approaches typically apply a single holistic judgment over the entire action-observation sequence-a strategy that proves unreliable on long-horizon tasks and yields binary verdicts offering no insight into where or why an agent fails. This opacity limits the utility of evaluation as a diagnostic tool for agent development. We introduce GUIDE (GUI Understanding and Interpretable Diagnostic Evaluation), a framework that decomposes trajectory assessment into three sequential stages mirroring the compositional structure of GUI tasks. Trajectory Segmentation partitions the full trace into semantically coherent subtask units. Subtask Diagnosis evaluates each unit in context, assigning a completion verdict and generating a structured error analysis with corrective recommendations. Overall Summary aggregates per-subtask diagnoses into a task-level judgment. By operating on bounded subtask segments rather than full trajectories, GUIDE mitigates the context overload that degrades existing evaluators as task complexity grows. We validate GUIDE on three benchmarks: an industrial e-commerce dataset of 932 trajectories, AGENTREWARDBENCH spanning five web agent tasks with 1302 trajectories, and AndroidBench for mobile device control. Across all settings, GUIDE substantially outperforms existing evaluators-achieving up to 5.35 percentage points higher accuracy than the strongest baseline-while producing structured diagnostic reports that directly inform agent improvement. --- ### BiTDiff: Fine-Grained 3D Conducting Motion Generation via BiMamba-Transformer Diffusion Authors: Tianzhi Jia, Kaixing Yang, Xiaole Yang, Xulong Tang, Ke Qiu, Shikui Wei, Yao Zhao Published: 2026-04-06 Categories: cs.CV, cs.MM arXiv: https://arxiv.org/abs/2604.04395v2 3D conducting motion generation aims to synthesize fine-grained conductor motions from music, with broad potential in music education, virtual performance, digital human animation, and human-AI co-creation. However, this task remains underexplored due to two major challenges: (1) the lack of large-scale fine-grained 3D conducting datasets and (2) the absence of effective methods that can jointly support long-sequence generation with high quality and efficiency. To address the data limitation, we develop a quality-oriented 3D conducting motion collection pipeline and construct CM-Data, a fine-grained SMPL-X dataset with about 10 hours of conducting motion data. To the best of our knowledge, CM-Data is the first and largest public dataset for 3D conducting motion generation. To address the methodological limitation, we propose BiTDiff, a novel framework for 3D conducting motion generation, built upon a BiMamba-Transformer hybrid model architecture for efficient long-sequence modeling and a Diffusion-based generative strategy with human-kinematic decomposition for high-quality motion synthesis. Specifically, BiTDiff introduces auxiliary physical-consistency losses and a hand-/body-specific forward-kinematics design for better fine-grained motion modeling, while leveraging BiMamba for memory-efficient long-sequence temporal modeling and Transformer for cross-modal semantic alignment. In addition, BiTDiff supports training-free joint-level motion editing, enabling downstream human-AI interaction design. Extensive quantitative and qualitative experiments demonstrate that BiTDiff achieves state-of-the-art (SOTA) performance for 3D conducting motion generation on the CM-Data dataset. Code will be available upon acceptance. --- ### Finite-Time Analysis of Q-Value Iteration for General-Sum Stackelberg Games Authors: Narim Jeong, Donghwan Lee Published: 2026-04-06 Categories: cs.LG, eess.SY arXiv: https://arxiv.org/abs/2604.04394v1 Reinforcement learning has been successful both empirically and theoretically in single-agent settings, but extending these results to multi-agent reinforcement learning in general-sum Markov games remains challenging. This paper studies the convergence of Stackelberg Q-value iteration in two-player general-sum Markov games from a control-theoretic perspective. We introduce a relaxed policy condition tailored to the Stackelberg setting and model the learning dynamics as a switching system. By constructing upper and lower comparison systems, we establish finite-time error bounds for the Q-functions and characterize their convergence properties. Our results provide a novel control-theoretic perspective on Stackelberg learning. Moreover, to the best of the authors' knowledge, this paper offers the first finite-time convergence guarantees for Q-value iteration in general-sum Markov games under Stackelberg interactions. --- ### Gradual Cognitive Externalization: From Modeling Cognition to Constituting It Authors: Zhimin Zhao Published: 2026-04-06 Categories: cs.AI, cs.CY, cs.ET, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2604.04387v2 Developers are publishing AI agent skills that replicate a colleague's communication style, encode a supervisor's mentoring heuristics, or preserve a person's behavioral repertoire beyond biological death. To explain why, we propose Gradual Cognitive Externalization (GCE), a framework arguing that ambient AI systems, through sustained causal coupling with users, transition from modeling cognitive functions to constituting part of users' cognitive architectures. GCE adopts an explicit functionalist commitment: cognitive functions are individuated by their causal-functional roles, not by substrate. The framework rests on the behavioral manifold hypothesis and a central falsifiable assumption, the no behaviorally invisible residual (NBIR) hypothesis: for any cognitive function whose behavioral output lies on a learnable manifold, no behaviorally invisible component is necessary for that function's operation. We document evidence from deployed AI systems showing that externalization preconditions are already observable, formalize three criteria separating cognitive integration from tool use (bidirectional adaptation, functional equivalence, causal coupling), and derive five testable predictions with theory-constrained thresholds. --- ### Automatically Generating Hard Math Problems from Hypothesis-Driven Error Analysis Authors: Jiayu Fu, Mourad Heddaya, Chenhao Tan Published: 2026-04-06 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.04386v1 Numerous math benchmarks exist to evaluate LLMs' mathematical capabilities. However, most involve extensive manual effort and are difficult to scale. Consequently, they cannot keep pace with LLM development or easily provide new instances to mitigate overfitting. Some researchers have proposed automatic benchmark generation methods, but few focus on identifying the specific math concepts and skills on which LLMs are error-prone, and most can only generate category-specific benchmarks. To address these limitations, we propose a new math benchmark generation pipeline that uses AI-generated hypotheses to identify the specific math concepts and skills that LLMs struggle with, and then generates new benchmark problems targeting these weaknesses. Experiments show that hypothesis accuracy positively correlates with the difficulty of the generated problems: problems generated from the most accurate hypotheses reduce Llama-3.3-70B-Instruct's accuracy to as low as 45%, compared to 77% on the original MATH benchmark. Furthermore, our pipeline is highly adaptable and can be applied beyond math to explore a wide range of LLM capabilities, making it a valuable tool for investigating how LLMs perform across different domains. --- ### How Alignment Routes: Localizing, Scaling, and Controlling Policy Circuits in Language Models Authors: Gregory N. Frank Published: 2026-04-06 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.04385v5 We localize the policy routing mechanism in alignment-trained language models. An intermediate-layer attention gate reads detected content and triggers deeper amplifier heads that boost the signal toward refusal. In smaller models the gate and amplifier are single heads; at larger scale they become bands of heads across adjacent layers. The gate contributes under 1% of output DLA, yet interchange testing (p < 0.001) and knockout cascade confirm it is causally necessary. Interchange screening at n >= 120 detects the same motif in twelve models from six labs (2B to 72B), though specific heads differ by lab. Per-head ablation weakens up to 58x at 72B and misses gates that interchange identifies; at scale, interchange is the only reliable audit. Modulating the detection-layer signal continuously controls policy from hard refusal through evasion to factual answering. On safety prompts the same intervention turns refusal into harmful guidance, showing that the safety-trained capability is gated by routing, not removed. Thresholds vary by topic and by input language, and the circuit relocates across generations within a family even while behavioral benchmarks register no change. Routing is early-commitment: the gate fires at its own layer before deeper layers finish processing the input. An in-context substitution cipher collapses gate interchange necessity by 70 to 99% across three models, and the model switches to puzzle-solving rather than refusal. Injecting the plaintext gate activation into the cipher forward pass restores 48% of refusals in Phi-4-mini, localizing the bypass to the routing interface. A second method, cipher contrast analysis, uses plain/cipher DLA differences to map the full cipher-sensitive routing circuit in O(3n) forward passes. Any encoding that defeats detection-layer pattern matching bypasses the policy regardless of whether deeper layers reconstruct the content. --- ### Compressible Softmax-Attended Language under Incompressible Attention Authors: Wonsuk Lee Published: 2026-04-06 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.04384v2 Softmax attention defines an interaction through $d_h$ head dimensions, but not all dimensions carry equal weight once real text passes through. We decompose the attention logit field into a learned component and a generated component and measure their spectra separately. For all 5,888 KV heads in five transformer language models (124M--7B parameters, four architecture families), the logit energy field $\tilde{E}$ reaches 90\% of its variance in 2--11 singular components. The learned interaction matrix $W_Q^\mathrm{T} W_K$ needs 38--75 components for the same threshold out of $d_h \in {64, 128}$. The spectral gap is 5--25$\times$ in effective rank. The compressibility of softmax-attended language is a property of the data, not the frame that analyzes it. --- ### Optimizing Service Operations via LLM-Powered Multi-Agent Simulation Authors: Yanyuan Wang, Xiaowei Zhang Published: 2026-04-06 Categories: cs.AI, cs.MA, math.OC arXiv: https://arxiv.org/abs/2604.04383v1 Service system performance depends on how participants respond to design choices, but modeling these responses is hard due to the complexity of human behavior. We introduce an LLM-powered multi-agent simulation (LLM-MAS) framework for optimizing service operations. We pose the problem as stochastic optimization with decision-dependent uncertainty: design choices are embedded in prompts and shape the distribution of outcomes from interacting LLM-powered agents. By embedding key numerical information in prompts and extracting it from LLM-generated text, we model this uncertainty as a controlled Markov chain. We develop an on-trajectory learning algorithm that, on a single simulation run, simultaneously constructs zeroth-order gradient estimates and updates design parameters to optimize steady-state performance. We also incorporate variance reduction techniques. In a sustainable supply chain application, our method outperforms benchmarks, including blackbox optimization and using LLMs as numerical solvers or as role-playing system designers. A case study on optimal contest design with real behavioral data shows that LLM-MAS is both as a cost-effective evaluator of known designs and an exploratory tool that can uncover strong designs overlooked by traditional approaches. --- ### CPT: Controllable and Editable Design Variations with Language Models Authors: Karthik Suresh, Amine Ben Khalifa, Li Zhang, Wei-ting Hsu, Fangzheng Wu, Vinay More, Asim Kadav Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.04380v1 Designing visually diverse and high-quality designs remains a manual, time-consuming process, limiting scalability and personalization in creative workflows. We present a system for generating editable design variations using a decoder-only language model, the Creative Pre-trained Transformer (CPT), trained to predict visual style attributes in design templates. At the core of our approach is a new representation called Creative Markup Language (CML), a compact, machine-learning-friendly format that captures canvas-level structure, page layout, and element-level details (text, images, and vector graphics), including both content and style. We fine-tune CPT on a large corpus of design templates authored by professional designers, enabling it to learn meaningful, context-aware predictions for attributes such as color schemes and font choices. The model produces semantically structured and stylistically coherent outputs, preserving internal consistency across elements. Unlike generative image models, our system yields fully editable design documents rather than pixel-only images, allowing users to iterate and personalize within a design editor. In experiments, our approach generates contextual color and font variations for existing templates and shows promise in adjusting layouts while maintaining design principles. --- ### Reinforce to Learn, Elect to Reason: A Dual Paradigm for Video Reasoning Authors: Songyuan Yang, Weijiang Yu, Jilin Ma, Ziyu Liu, Guijian Tang, Wenjing Yang, Huibin Tan, Nong Xiao Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04379v1 Video reasoning has advanced with large multimodal models (LMMs), yet their inference is often a single pass that returns an answer without verifying whether the reasoning is evidence-aligned. We introduce Reinforce to Learn, Elect to Reason (RLER), a dual paradigm that decouples learning to produce evidence from obtaining a reliable answer. In RLER-Training, we optimize the policy with group-relative reinforcement learning (RL) and 3 novel task-driven rewards: Frame-sensitive reward grounds reasoning on explicit key frames, Think-transparency reward shapes readable and parsable reasoning traces, and Anti-repetition reward boosts information density. These signals teach the model to emit structured, machine-checkable evidence and potentiate reasoning capabilities. In RLER-Inference, we apply a train-free orchestrator that generates a small set of diverse candidates, parses their answers and cited frames, scores them by evidence consistency, confidence, transparency, and non-redundancy, and then performs a robust evidence-weighted election. This closes the loop between producing and using evidence, improving reliability and interpretability without enlarging the model. We comprehensively evaluate RLER against various open-source and RL-based LMMs on 8 representative benchmarks. RLER achieves state of the art across all benchmarks and delivers an average improvement of 6.3\% over base models, while using on average 3.1 candidates per question, indicating a favorable balance between compute and quality. The results support a simple thesis: making evidence explicit during learning and electing by evidence during inference is a robust path to trustworthy video reasoning. --- ### Towards Considerate Human-Robot Coexistence: A Dual-Space Framework of Robot Design and Human Perception in Healthcare Authors: Yuanchen Bai, Zijian Ding, Ruixiang Han, Niti Parikh, Wendy Ju, Angelique Taylor Published: 2026-04-06 Categories: cs.RO, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2604.04374v2 The rapid advancement of robotics is reshaping what it means for humans and robots to coexist -- through expanded capabilities, more intuitive interactions, and deeper integration into real-world workflows. Beyond sharing physical space, this coexistence is increasingly characterized by organizational embeddedness, temporal evolution, social situatedness, and open-ended uncertainty. Because such coexistence extends beyond a single encounter, understanding healthcare robots requires looking beyond initial acceptance to how stakeholders' perceptions evolve through continued engagement. Yet, prior work has largely relied on single-point snapshots of attitudes and acceptance, offering limited insight into coexistence as a long-term, dynamic process. We address these gaps through in-depth follow-up interviews with nine participants from a 14-week co-design study on healthcare robots. We identify the human perception space, which includes four interpretive dimensions (i.e., degree of decomposition, source of evidence, scope of reasoning, and temporal orientation). We enrich the conceptual framework of human-robot coexistence by conceptualizing the mutual relationship between the human perception space and the robot design space as a co-evolving loop, in which human needs, design decisions, situated interpretations, and social mediation continuously reshape one another over time. Building on this, we propose considerate human-robot coexistence, arguing that humans act not only as design contributors but also as interpreters and mediators who actively shape how robots are understood and integrated across deployment stages. Our related prior work and supplementary materials, including the interview protocol, are available at https://byc-sophie.github.io/considerate-human-robot-coexistence/ --- ### Decocted Experience Improves Test-Time Inference in LLM Agents Authors: Maohao Shen, Kaiwen Zha, Zexue He, Zhang-Wei Hong, Siru Ouyang, J. Jon Ryu, Prasanna Sattigeri, Suhas Diggavi, Gregory Wornell Published: 2026-04-06 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.04373v1 There is growing interest in improving LLMs without updating model parameters. One well-established direction is test-time scaling, where increased inference-time computation (e.g., longer reasoning, sampling, or search) is used to improve performance. However, for complex reasoning and agentic tasks, naively scaling test-time compute can substantially increase cost and still lead to wasted budget on suboptimal exploration. In this paper, we explore \emph{context} as a complementary scaling axis for improving LLM performance, and systematically study how to construct better inputs that guide reasoning through \emph{experience}. We show that effective context construction critically depends on \emph{decocted experience}. We present a detailed analysis of experience-augmented agents, studying how to derive context from experience, how performance scales with accumulated experience, what characterizes good context, and which data structures best support context construction. We identify \emph{decocted experience} as a key mechanism for effective context construction: extracting essence from experience, organizing it coherently, and retrieving salient information to build effective context. We validate our findings across reasoning and agentic tasks, including math reasoning, web browsing, and software engineering. --- ### Graph-to-Frame RAG: Visual-Space Knowledge Fusion for Training-Free and Auditable Video Reasoning Authors: Songyuan Yang, Weijiang Yu, Ziyu Liu, Guijian Tang, Wenjing Yang, Huibin Tan, Nong Xiao Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04372v1 When video reasoning requires external knowledge, many systems with large multimodal models (LMMs) adopt retrieval augmentation to supply the missing context. Appending textual or multi-clip evidence, however, forces heterogeneous signals into a single attention space. We observe diluted attention and higher cognitive load even on non-long videos. The bottleneck is not only what to retrieve but how to represent and fuse external knowledge with the video backbone.We present Graph-to-Frame RAG (G2F-RAG), a training free and auditable paradigm that delivers knowledge in the visual space. On the offline stage, an agent builds a problem-agnostic video knowledge graph that integrates entities, events, spatial relations, and linked world knowledge. On the online stage, a hierarchical multi-agent controller decides whether external knowledge is needed, retrieves a minimal sufficient subgraph, and renders it as a single reasoning frame appended to the video. LMMs then perform joint reasoning in a unified visual domain. This design reduces cognitive load and leaves an explicit, inspectable evidence trail.G2F-RAG is plug-and-play across backbones and scales. It yields consistent gains on diverse public benchmarks, with larger improvements in knowledge-intensive settings. Ablations further confirm that knowledge representation and delivery matter. G2F-RAG reframes retrieval as visual space knowledge fusion for robust and interpretable video reasoning. --- ### Attributed Network Alignment: Statistical Limits and Efficient Algorithm Authors: Dong Huang, Chenyang Tian, Pengkun Yang Published: 2026-04-06 Categories: math.ST, stat.ML arXiv: https://arxiv.org/abs/2604.04365v1 This paper studies the problem of recovering a hidden vertex correspondence between two correlated graphs when both edge weights and node features are observed. While most existing work on graph alignment relies primarily on edge information, many real-world applications provide informative node features in addition to graph topology. To capture this setting, we introduce the featured correlated Gaussian Wigner model, where two graphs are coupled through an unknown vertex permutation, and the node features are correlated under the same permutation. We characterize the optimal information-theoretic thresholds for exact recovery and partial recovery of the latent mapping. On the algorithmic side, we propose QPAlign, an algorithm based on a quadratic programming relaxation, and demonstrate its strong empirical performance on both synthetic and real datasets. Moreover, we also derive theoretical guarantees for the proposed procedure, supporting its reliability and providing convergence guarantees. --- ### Context is All You Need Authors: Jean Erik Delanois, Shruti Joshi, Ryan Golden, Teresa Nick, Maxim Bazhenov Published: 2026-04-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.04364v1 Artificial Neural Networks (ANNs) are increasingly deployed across diverse real-world settings, where they must operate under data distributions that differ from those seen during training. This challenge is central to Domain Generalization (DG), which trains models to generalize to unseen domains without target data, and Test-Time Adaptation (TTA), which improves robustness by adapting to unlabeled test data at deployment. Existing approaches to address these challenges are often complex, resource-intensive, and difficult to scale. We introduce CONTXT (Contextual augmentatiOn for Neural feaTure X Transforms), a simple and intuitive method for contextual adaptation. CONTXT modulates internal representations using simple additive and multiplicative feature transforms. Within a TTA setting, it yields consistent gains across discriminative tasks (e.g., ANN/CNN classification) and generative models (e.g., LLMs). The method is lightweight, easy to integrate, and incurs minimal overhead, enabling robust performance under domain shift without added complexity. More broadly, CONTXT provides a compact way to steer information flow and neural processing without retraining. --- ### Integer-Only Operations on Extreme Learning Machine Test Time Classification Authors: Emerson Lopes Machadoa, Cristiano Jacques Miosso, Ricardo Pezzuol Jacobi Published: 2026-04-06 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.04363v1 We present a theoretical analysis and empirical evaluations of a novel set of techniques for computational cost reduction of test time operations of network classifiers based on extreme learning machine (ELM). By exploring some characteristics we derived from these models, we show that the classification at test time can be performed using solely integer operations without compromising the classification accuracy. Our contributions are as follows: (i) We show empirical evidence that the input weights values can be drawn from the ternary set with limited reduction of the classification accuracy. This has the computational advantage of dismissing multiplications; (ii) We prove the classification accuracy of normalized and non-normalized test signals are the same; (iii) We show how to create an integer version of the output weights that results in a limited reduction of the classification accuracy. We tested our techniques on 5 computer vision datasets commonly used in the literature and the results indicate that our techniques can allow the reduction of the computational cost of the operations necessary for the classification at test time in FPGAs. This is important in embedded applications, where power consumption is limited, and crucial in data centers of large corporations, where power consumption is expensive. --- ### GROUNDEDKG-RAG: Grounded Knowledge Graph Index for Long-document Question Answering Authors: Tianyi Zhang, Andreas Marfurt Published: 2026-04-06 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.04359v1 Retrieval-augmented generation (RAG) systems have been widely adopted in contemporary large language models (LLMs) due to their ability to improve generation quality while reducing the required input context length. In this work, we focus on RAG systems for long-document question answering. Current approaches suffer from a heavy reliance on LLM descriptions resulting in high resource consumption and latency, repetitive content across hierarchical levels, and hallucinations due to no or limited grounding in the source text. To improve both efficiency and factual accuracy through grounding, we propose GroundedKG-RAG, a RAG system in which the knowledge graph is explicitly extracted from and grounded in the source document. Specifically, we define nodes in GroundedKG as entities and actions, and edges as temporal or semantic relations, with each node and edge grounded in the original sentences. We construct GroundedKG from semantic role labeling (SRL) and abstract meaning representation (AMR) parses and then embed it for retrieval. During querying, we apply the same transformation to the query and retrieve the most relevant sentences from the grounded source text for question answering. We evaluate GroundedKG-RAG on examples from the NarrativeQA dataset and find that it performs on par with a state-of-the art proprietary long-context model at smaller cost and outperforms a competitive baseline. Additionally, our GroundedKG is interpretable and readable by humans, facilitating auditing of results and error analysis. --- ### Spatially-Weighted CLIP for Street-View Geo-localization Authors: Ting Han, Fengjiao Li, Chunsong Chen, Haoling Huang, Yiping Chen, Meiliu Wu Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.04357v1 This paper proposes Spatially-Weighted CLIP (SW-CLIP), a novel framework for street-view geo-localization that explicitly incorporates spatial autocorrelation into vision-language contrastive learning. Unlike conventional CLIP-based methods that treat all non-matching samples as equally negative, SW-CLIP leverages Tobler's First Law of Geography to model geographic relationships through distance-aware soft supervision. Specifically, we introduce a location-as-text representation to encode geographic positions and replace one-hot InfoNCE targets with spatially weighted soft labels derived from geodesic distance. Additionally, a neighborhood-consistency regularization is employed to preserve local spatial structure in the embedding space. Experiments on a multi-city dataset demonstrate that SW-CLIP significantly improves geo-localization accuracy, reduces long-tail errors, and enhances spatial coherence compared to standard CLIP. The results highlight the importance of shifting from semantic alignment to geographic alignment for robust geo-localization and provide a general paradigm for integrating spatial principles into multimodal representation learning. ---