# Research Papers — 2026-03 · bundle page 138 of 243 > Full-text markdown of papers #6851–6900 of 12123 published in 2026-03, newest first. AI-readable mirrors of the full 12123-paper 2026-03 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-03/page/138 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/137.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/139.md) ### Data-driven Progressive Discovery of Physical Laws Authors: Mingkun Xia, Weiwei Zhang Published: 2026-03-14 Categories: cs.LG, physics.data-an arXiv: https://arxiv.org/abs/2603.13727v2 Symbolic regression is a powerful tool for knowledge discovery, enabling the extraction of interpretable mathematical expressions directly from data. However, conventional symbolic discovery typically follows an end-to-end, "one-step" process, which often generates lengthy and physically meaningless expressions when dealing with real physical systems, leading to poor model generalization. This limitation fundamentally stems from its deviation from the basic path of scientific discovery: physical laws do not exist in a single form but follow a hierarchical and progressive pattern from simplicity to complexity. Motivated by this principle, we propose Chain of Symbolic Regression (CoSR), a novel framework that models the discovery of physical laws as a chain of symbolic knowledge. This knowledge chain is formed by progressively combining multiple knowledge units with clear physical meanings along a specific logic, ultimately enabling the precise discovery of the underlying physical laws from data. CoSR fully recapitulates the progressive discovery path from Kepler's third law to the law of universal gravitation in classical mechanics, and is applied to three types of problems: turbulent Rayleigh-Benard convection, viscous flows in a circular pipe, and laser-metal interaction, demonstrating its ability to improve classical scaling theories. Finally, CoSR showcases its capability to discover new knowledge in the complex engineering problem of aerodynamic coefficients scaling for different aircraft. --- ### Can We Trust LLMs on Memristors? Diving into Reasoning Ability under Non-Ideality Authors: Taiqiang Wu, Yuxin Cheng, Chenchen Ding, Runming Yang, Xincheng Feng, Wenyong Zhou, Zhengwu Liu, Ngai Wong Published: 2026-03-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.13725v1 Memristor-based analog compute-in-memory (CIM) architectures provide a promising substrate for the efficient deployment of Large Language Models (LLMs), owing to superior energy efficiency and computational density. However, these architectures suffer from precision issues caused by intrinsic non-idealities of memristors. In this paper, we first conduct a comprehensive investigation into the impact of such typical non-idealities on LLM reasoning. Empirical results indicate that reasoning capability decreases significantly but varies for distinct benchmarks. Subsequently, we systematically appraise three training-free strategies, including thinking mode, in-context learning, and module redundancy. We thus summarize valuable guidelines, i.e., shallow layer redundancy is particularly effective for improving robustness, thinking mode performs better under low noise levels but degrades at higher noise, and in-context learning reduces output length with a slight performance trade-off. Our findings offer new insights into LLM reasoning under non-ideality and practical strategies to improve robustness. --- ### Sparse-Dense Mixture of Experts Adapter for Multi-Modal Tracking Authors: Yabin Zhu, Jianqi Li, Chenglong Li, Jiaxiang Wang, Chengjie Gu, Jin Tang Published: 2026-03-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.13719v1 Parameter-efficient fine-tuning (PEFT) techniques, such as prompts and adapters, are widely used in multi-modal tracking because they alleviate issues of full-model fine-tuning, including time inefficiency, high resource consumption, parameter storage burden, and catastrophic forgetting. However, due to cross-modal heterogeneity, most existing PEFT-based methods struggle to effectively represent multi-modal features within a unified framework with shared parameters. To address this problem, we propose a novel Sparse-Dense Mixture of Experts Adapter (SDMoEA) framework for PEFT-based multi-modal tracking under a unified model structure. Specifically, we design an SDMoE module as the multi-modal adapter to model modality-specific and shared information efficiently. SDMoE consists of a sparse MoE and a dense-shared MoE: the former captures modality-specific information, while the latter models shared cross-modal information. Furthermore, to overcome limitations of existing tracking methods in modeling high-order correlations during multi-level multi-modal fusion, we introduce a Gram-based Semantic Alignment Hypergraph Fusion (GSAHF) module. It first employs Gram matrices for cross-modal semantic alignment, ensuring that the constructed hypergraph accurately reflects semantic similarity and high-order dependencies between modalities. The aligned features are then integrated into the hypergraph structure to exploit its ability to model high-order relationships, enabling deep fusion of multi-level multi-modal information. Extensive experiments demonstrate that the proposed method achieves superior performance compared with other PEFT approaches on several multi-modal tracking benchmarks, including LasHeR, RGBT234, VTUAV, VisEvent, COESOT, DepthTrack, and VOT-RGBD2022. --- ### InterventionLens: A Multi-Agent Framework for Detecting ASD Intervention Strategies in Parent-Child Shared Reading Authors: Xiao Wang, Lu Dong, Ifeoma Nwogu, Srirangaraj Setlur, Venu Govindaraju Published: 2026-03-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.13710v1 Home-based interventions like parent-child shared reading provide a cost-effective approach for supporting children with autism spectrum disorder (ASD). However, analyzing caregiver intervention strategies in naturalistic home interactions typically relies on expert annotation, which is costly, time-intensive, and difficult to scale. To address this challenge, we propose InterventionLens, an end-to-end multi-agent system for automatically detecting and temporally segmenting caregiver intervention strategies from shared reading videos. Without task-specific model training or fine-tuning, InterventionLens uses a collaborative multi-agent architecture to integrate multimodal interaction content and perform fine-grained strategy analysis. Experiments on the ASD-HI dataset show that InterventionLens achieves an overall F1 score of 79.44\%, outperforming the baseline by 19.72\%. These results suggest that InterventionLens is a promising system for analyzing caregiver intervention strategies in home-based ASD shared reading settings. Additional resources will be released on the project page. --- ### REAEDP: Entropy-Calibrated Differentially Private Data Release with Formal Guarantees and Attack-Based Evaluation Authors: Bo Ma, Jinsong Wu, Wei Qi Yan Published: 2026-03-14 Categories: cs.CR, cs.CV arXiv: https://arxiv.org/abs/2603.13709v1 Sensitive data release is vulnerable to output-side privacy threats such as membership inference, attribute inference, and record linkage. This creates a practical need for release mechanisms that provide formal privacy guarantees while preserving utility in measurable ways. We propose REAEDP, a differential privacy framework that combines entropy-calibrated histogram release, a synthetic-data release mechanism, and attack-based evaluation. On the theory side, we derive an explicit sensitivity bound for Shannon entropy, together with an extension to Rényi entropy, for adjacent histogram datasets, enabling calibrated differentially private release of histogram statistics. We further study a synthetic-data mechanism $\mathcal{F}$ with a privacy-test structure and show that it satisfies a formal differential privacy guarantee under the stated parameter conditions. On multiple public tabular datasets, the empirical entropy change remains below the theoretical bound in the tested regime, standard Laplace and Gaussian baselines exhibit comparable trends, and both membership-inference and linkage-style attack performance move toward random-guess behavior as the privacy parameter decreases. These results support REAEDP as a practically usable privacy-preserving release pipeline in the tested settings. Source code: https://github.com/mabo1215/REAEDP.git --- ### RSEdit: Text-Guided Image Editing for Remote Sensing Authors: Chen Zhenyuan, Zhang Zechuan, Zhang Feng Published: 2026-03-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.13708v2 In this paper, we explore text-guided image editing in the remote sensing domain using generative modeling. We propose \rsedit, a collection of models from U-Net to DiT with various configurations. Specifically, we present the first comprehensive study of conditioning strategies for building image editing models from off-the-shelf text-to-image ones. Our experiments show that \rsedit achieves the best instruction-faithful edits while preserving geospatial structure. We release the code at \url{https://github.com/Bili-Sakura/RSEdit-Preview} and checkpoints at \url{https://huggingface.co/collections/BiliSakura/rsedit}. --- ### REFINE-DP: Diffusion Policy Fine-tuning for Humanoid Loco-manipulation via Reinforcement Learning Authors: Zhaoyuan Gu, Yipu Chen, Zimeng Chai, Alfred Cueva, Thong Nguyen, Yifan Wu, Huishu Xue, Minji Kim, Isaac Legene, Fukang Liu, KyoungMok Kim, Ayan Barula et al. Published: 2026-03-14 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.13707v3 Humanoid loco-manipulation requires coordinated task-space motion planning with stable loco-manipulation command tracking under complex robot-environment dynamics and long-horizon tasks. While diffusion policies (DPs) show promise for learning from demonstrations, deploying them on humanoids poses critical challenges: the motion planner trained offline is decoupled from the loco-manipulation controller, leading to poor command tracking, compounding distribution shift, and task failures. The common approach of scaling demonstration data is prohibitively expensive for high-dimensional humanoid systems. To address this challenge, we present REFINE-DP (REinforcement learning FINE-tuning of Diffusion Policy), a hierarchical framework that jointly optimizes a DP motion planner and an RL-based loco-manipulation controller. The DP is fine-tuned via a PPO-based diffusion policy gradient to improve task success rate, while the controller is simultaneously updated to accurately track the planner's evolving command distribution, reducing the distributional mismatch that degrades motion quality. We validate REFINE-DP on a humanoid robot performing loco-manipulation tasks, including door traversal and long-horizon object transport. REFINE-DP achieves an over 90% success rate in simulation, even in out-of-distribution cases not seen in the pre-training data, and enables real-world execution without privileged state information. Our proposed method substantially outperforms pre-trained DP baselines and demonstrates that RL fine-tuning is key to reliable humanoid loco-manipulation. https://refine-dp.github.io/REFINE-DP/ --- ### Routing Channel-Patch Dependencies in Time Series Forecasting with Graph Spectral Decomposition Authors: Dongyuan Li, Shun Zheng, Chang Xu, Jiang Bian, Renhe Jiang Published: 2026-03-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.13702v1 Time series forecasting has attracted significant attention in the field of AI. Previous works have revealed that the Channel-Independent (CI) strategy improves forecasting performance by modeling each channel individually, but it often suffers from poor generalization and overlooks meaningful inter-channel interactions. Conversely, Channel-Dependent (CD) strategies aggregate all channels, which may introduce irrelevant information and lead to oversmoothing. Despite recent progress, few existing methods offer the flexibility to adaptively balance CI and CD strategies in response to varying channel dependencies. To address this, we propose a generic plugin xCPD, that can adaptively model the channel-patch dependencies from the perspective of graph spectral decomposition. Specifically, xCPD first projects multivariate signals into the frequency domain using a shared graph Fourier basis, and groups patches into low-, mid-, and high-frequency bands based on their spectral energy responses. xCPD then applies a channel-adaptive routing mechanism that dynamically adjusts the degree of inter-channel interaction for each patch, enabling selective activation of frequency-specific experts. This facilitates fine-grained input-aware modeling of smooth trends, local fluctuations, and abrupt transitions. xCPD can be seamlessly integrated on top of existing CI and CD forecasting models, consistently enhancing both accuracy and generalization across benchmarks. The code is available https://github.com/Clearloveyuan/xCPD. --- ### SAATT Nav: a Socially Aware Autonomous Transparent Transportation Navigation Framework for Wheelchairs Authors: Yutong Zhang, Shaiv Y. Mehra, Bradley S. Duerstock, Juan P. Wachs Published: 2026-03-14 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2603.13698v2 While powered wheelchairs reduce physical fatigue as opposed to manual wheelchairs for individuals with mobility impairment, they demand high cognitive workload due to information processing, decision making and motor coordination. Current autonomous systems lack social awareness in navigation and transparency in decision-making, leading to decreased perceived safety and trust from the user and others in context. This work proposes Socially Aware Autonomous Transparent Transportation (SAATT) Navigation framework for wheelchairs as a potential solution. By implementing a Large Language Model (LLM) informed of user intent and capable of predicting other peoples' intent as a decision-maker for its local controller, it is able to detect and navigate social situations, such as passing pedestrians or a pair conversing. Furthermore, the LLM textually communicates its reasoning at each waypoint for transparency. In this experiment, it is compared against a standard global planner, a representative competing social navigation model, and an Ablation study in three simulated environments varied by social levels in eight metrics categorized under Safety, Social Compliance, Efficiency, and Comfort. Overall, SAATT Nav outperforms in most social situations and equivalently or only slightly worse in the remaining metrics, demonstrating the potential of a socially aware and transparent autonomous navigation system to assist wheelchair users. --- ### Repetition Without Exclusivity: Scale Sensitivity of Referential Mechanisms in Child-Scale Language Models Authors: Jon-Paul Cacioli Published: 2026-03-14 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.13696v1 We present the first systematic evaluation of mutual exclusivity (ME) -- the bias to map novel words to novel referents -- in text-only language models trained on child-directed speech. We operationalise ME as referential suppression: when a familiar object is relabelled in a two-referent discourse context, ME predicts decreased probability of the labelled noun at a subsequent completion position. Three pilot findings motivate a pre-registered scale-sensitivity experiment: (1) a masked language model (BabyBERTa) is entirely insensitive to multi-sentence referential context; (2) autoregressive models show robust repetition priming -- the opposite of ME -- when familiar nouns are re-labelled; and (3) a novel context-dependence diagnostic reveals that apparent ME-like patterns with nonce tokens are fully explained by embedding similarity, not referential disambiguation. In the confirmatory experiment, we train 45 GPT-2-architecture models (2.9M, 8.9M, and 33.5M parameters; 5, 10, and 20 epochs on AO-CHILDES; 5 seeds each) and evaluate on a pre-registered ME battery. Anti-ME repetition priming is significant in all 9 cells (85-100% of items; all p < 2.4 x 10^-13). Priming attenuates with improved language modelling (Spearman rho = -0.533, p = 0.0002) but never crosses zero across a 3.8x perplexity range. The context-dependence diagnostic replicates in all 9 cells, and dose-response priming increases with repetitions in 8/9 cells (all trend p < 0.002). These findings indicate that distributional learning on child-directed speech produces repetition-based reference tracking rather than lexical exclusivity. We connect this to the grounded cognition literature and argue that referential grounding may be a necessary ingredient for ME -- an empirical claim about required input structure, not a nativist one. --- ### Steering Generative Models for Accessibility: EasyRead Image Generation Authors: Nicolas Dickenmann, Yanis Merzouki, Sonia Laguna, Thy Nowak-Tran, Emanuele Palumbo, Julia E. Vogt, Gerda Binder Published: 2026-03-14 Categories: cs.HC, cs.CV arXiv: https://arxiv.org/abs/2603.13695v1 EasyRead pictograms are simple, visually clear images that represent specific concepts and support comprehension for people with intellectual disabilities, low literacy, or language barriers. The large-scale production of EasyRead content has traditionally been constrained by the cost and expertise required to manually design pictograms. In contrast, automatic generation of such images could significantly reduce production time and cost, enabling broader accessibility across digital and printed materials. However, modern diffusion-based image generation models tend to produce outputs that exhibit excessive visual detail and lack stylistic stability across random seeds, limiting their suitability for clear and consistent pictogram generation. This challenge highlights the need for methods specifically tailored to accessibility-oriented visual content. In this work, we present a unified pipeline for generating EasyRead pictograms by fine-tuning a Stable Diffusion model using LoRA adapters on a curated corpus that combines augmented samples from multiple pictogram datasets. Since EasyRead pictograms lack a unified formal definition, we introduce an EasyRead score to benchmark pictogram quality and consistency. Our results demonstrate that diffusion models can be effectively steered toward producing coherent EasyRead-style images, indicating that generative models can serve as practical tools for scalable and accessible pictogram production. --- ### QuarkMedBench: A Real-World Scenario Driven Benchmark for Evaluating Large Language Models Authors: Yao Wu, Kangping Yin, Liang Dong, Zhenxin Ma, Shuting Xu, Xuehai Wang, Yuxuan Jiang, Tingting Yu, Yunqing Hong, Jiayi Liu, Rianzhe Huang, Shuxin Zhao et al. Published: 2026-03-14 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.13691v1 While Large Language Models (LLMs) excel on standardized medical exams, high scores often fail to translate to high-quality responses for real-world medical queries. Current evaluations rely heavily on multiple-choice questions, failing to capture the unstructured, ambiguous, and long-tail complexities inherent in genuine user inquiries. To bridge this gap, we introduce QuarkMedBench, an ecologically valid benchmark tailored for real-world medical LLM assessment. We compiled a massive dataset spanning Clinical Care, Wellness Health, and Professional Inquiry, comprising 20,821 single-turn queries and 3,853 multi-turn sessions. To objectively evaluate open-ended answers, we propose an automated scoring framework that integrates multi-model consensus with evidence-based retrieval to dynamically generate 220,617 fine-grained scoring rubrics (~9.8 per query). During evaluation, hierarchical weighting and safety constraints structurally quantify medical accuracy, key-point coverage, and risk interception, effectively mitigating the high costs and subjectivity of human grading. Experimental results demonstrate that the generated rubrics achieve a 91.8% concordance rate with clinical expert blind audits, establishing highly dependable medical reliability. Crucially, baseline evaluations on this benchmark reveal significant performance disparities among state-of-the-art models when navigating real-world clinical nuances, highlighting the limitations of conventional exam-based metrics. Ultimately, QuarkMedBench establishes a rigorous, reproducible yardstick for measuring LLM performance on complex health issues, while its framework inherently supports dynamic knowledge updates to prevent benchmark obsolescence. --- ### Quantum-Enhanced Vision Transformer for Flood Detection using Remote Sensing Imagery Authors: Soumyajit Maity, Behzad Ghanbarian Published: 2026-03-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.13689v1 Reliable flood detection is critical for disaster management, yet classical deep learning models often struggle with the high-dimensional, nonlinear complexities inherent in remote sensing data. To mitigate these limitations, we introduced a novel Quantum-Enhanced Vision Transformer (ViT) that synergizes the global context-awareness of transformers with the expressive feature extraction capabilities of quantum computing. Using remote sensing imagery, we developed a hybrid architecture that processes inputs through parallel pathways, a ViT backbone and a quantum branch utilizing a 4-qubit parameterized quantum circuit for localized feature mapping. These distinct representations were fused to optimize binary classification. Results showed that the proposed hybrid model significantly outperformed a classical ViT baseline, increased overall accuracy from 84.48% to 94.47% and the F1-score from 0.841 to 0.944. Notably, the quantum integration substantially improved discriminative power in complex terrains for both class. These findings validate the potential of quantum-classical hybrid models to enhance precision in hydrological monitoring and earth observation applications. --- ### When Should Humans Step In? Optimal Human Dispatching in AI-Assisted Decisions Authors: Lezhi Tan, Naomi Sagan, Lihua Lei, Jose Blanchet Published: 2026-03-14 Categories: stat.ML, cs.LG, stat.AP arXiv: https://arxiv.org/abs/2603.13688v2 AI systems increasingly assist decision making by producing cheap factor-level assessments of complex inputs, but these assessments can be biased or incomplete. We study selective human evaluation: given AI signals, which factors should be dispatched to human experts? We formulate this human--AI collaboration problem as a factor-level information-acquisition problem. Under squared loss, the optimal dispatching rule maximizes a contextual reward; under a linear model, this reward admits a closed-form decomposition into two interpretable terms: predictive relevance and residual uncertainty in the human evaluation after conditioning on the AI signal. We instantiate the framework on peer review, decomposing papers into ten aspects and evaluating up to 3,408 ICLR submissions across three LLMs and multiple regression heads. In this retrospective proxy, where human aspect signals are extracted from existing reviews, predictions that query three of the ten aspects under the optimal dispatching rules match the full ten-aspect human benchmark. The finding holds when the human signal comes from a single noisy reviewer, and across review years. These results position principled dispatching as a practical building block for scalable human--AI decision systems. We also discuss safeguards under which such a rule can serve as decision support that leaves the final decision with a human. --- ### When Should Humans Step In? Optimal Human Dispatching in AI-Assisted Decisions Authors: Lezhi Tan, Naomi Sagan, Lihua Lei, Jose Blanchet Published: 2026-03-14 Categories: stat.ML, cs.LG, stat.AP arXiv: https://arxiv.org/abs/2603.13688v1 AI systems increasingly assist human decision making by producing preliminary assessments of complex inputs. However, such AI-generated assessments can often be noisy or systematically biased, raising a central question: how should costly human effort be allocated to correct AI outputs where it matters the most for the final decision? We propose a general decision-theoretic framework for human-AI collaboration in which AI assessments are treated as factor-level signals and human judgments as costly information that can be selectively acquired. We consider cases where the optimal selection problem reduces to maximizing a reward associated with each candidate subset of factors, and turn policy design into reward estimation. We develop estimation procedures under both nonparametric and linear models, covering contextual and non-contextual selection rules. In the linear setting, the optimal rule admits a closed-form expression with a clear interpretation in terms of factor importance and residual variance. We apply our framework to AI-assisted peer review. Our approach substantially outperforms LLM-only predictions and achieves performance comparable to full human review while using only 20-30% of the human information. Across different selection rules, we find that simpler rules derived under linear models can significantly reduce computational cost without harming final prediction performance. Our results highlight both the value of human intervention and the efficiency of principled dispatching. --- ### $τ$-Voice: Benchmarking Full-Duplex Voice Agents on Real-World Domains Authors: Soham Ray, Keshav Dhandhania, Victor Barres, Karthik Narasimhan Published: 2026-03-14 Categories: cs.SD, cs.AI arXiv: https://arxiv.org/abs/2603.13686v1 Full-duplex voice agents--systems that listen and speak simultaneously--are rapidly moving from research to production. However, existing evaluations address conversational dynamics and task completion in isolation. We introduce $τ$-voice, a benchmark for evaluating voice agents on grounded tasks with real-world complexity: agents must navigate complex multi-turn conversations, adhere to domain policies, and interact with the environment. The framework extends $τ^2$-bench into a novel voice agent benchmark combining verifiable completion of complex grounded tasks, full-duplex interaction, and realistic audio--enabling direct comparison between voice and text performance. A controllable and realistic voice user simulator provides diverse accents, realistic audio environments, and rich turn-taking dynamics; by decoupling simulation from wall-clock time, the user simulator can use the most capable LLM without real-time constraints. We evaluate task completion (pass@1) and voice interaction quality across 278 tasks: while GPT-5 (reasoning) achieves 85%, voice agents reach only 31--51% under clean conditions and 26--38% under realistic conditions with noise and diverse accents--retaining only 30--45% of text capability; qualitative analysis confirms 79--90% of failures stem from agent behavior, suggesting that observed failures primarily reflect agent behavior under our evaluation setup. $τ$-voice provides a reproducible testbed for measuring progress toward voice agents that are natural, conversational, and reliable. --- ### Preconditioned Test-Time Adaptation for Out-of-Distribution Debiasing in Narrative Generation Authors: Hanwen Shen, Ting Ying, Jiajie Lu, Shanshan Wang Published: 2026-03-14 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2603.13683v4 Although debiased large language models (LLMs) excel at handling known or low-bias prompts, they often fail on unfamiliar and high-bias prompts. We demonstrate via out-of-distribution (OOD) detection that these high-bias prompts cause a distribution shift, degrading static model performance. To enable real-time correction, we propose CAP-TTA, a test-time adaptation framework. CAP-TTA triggers context-aware LoRA updates only when a bias-risk score exceeds a set threshold. By utilizing an offline precomputed diagonal preconditioner, it ensures fast and stable optimization. Across multiple benchmarks and human evaluations, CAP-TTA effectively reduces toxicity/bias score with significantly lower latency than standard optimization methods (e.g., AdamW or SGD). Furthermore, it prevents catastrophic forgetting, and substantially improves narrative fluency over state-of-the-art baselines without compromising debiasing performance. --- ### Every Error has Its Magnitude: Asymmetric Mistake Severity Training for Multiclass Multiple Instance Learning Authors: Sungrae Hong, Jiwon Jeong, Jisu Shin, Donghee Han, Sol Lee, Kyungeun Kim, Mun Yong Yi Published: 2026-03-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.13682v1 Multiple Instance Learning (MIL) has emerged as a promising paradigm for Whole Slide Image (WSI) diagnosis, offering effective learning with limited annotations. However, existing MIL frameworks overlook diagnostic priorities and fail to differentiate the severity of misclassifications in multiclass, leaving clinically critical errors unaddressed. We propose a mistake-severity-aware training strategy that organizes diagnostic classes into a hierarchical structure, with each level optimized using a severity-weighted cross-entropy loss that penalizes high-severity misclassifications more strongly. Additionally, hierarchical consistency is enforced through probabilistic alignment, a semantic feature remix applied to the instance bag to robustly train class priority and accommodate clinical cases involving multiple symptoms. An asymmetric Mikel's Wheel-based metric is also introduced to quantify the severity of errors specific to medical fields. Experiments on challenging public and real-world in-house datasets demonstrate that our approach significantly mitigates critical errors in MIL diagnosis compared to existing methods. We present additional experimental results on natural domain data to demonstrate the generalizability of our proposed method beyond medical contexts. --- ### Toward Scalable Co-located Practical Learning: Assisting with Computer Vision and Multimodal Analytics Authors: Xinyu Li, Linxuan Zhao, Yueqiao Jin, Yuchen Liu, Jin Zhou, Roberto Martinez-Maldonado, Dragan Gasevic, Lixiang Yan Published: 2026-03-14 Categories: cs.HC, cs.CV arXiv: https://arxiv.org/abs/2603.13679v2 Co-located practical learning leaves evidence in visible actions around patients, task resources and room zones, but these traces are often recovered through live observation or retrospective video review. Fixed wide-angle video could reduce sensing burden, yet a debriefing pipeline must do more than detect behaviours: it must maintain detection after small camera-position shifts, relate the detector-derived behaviour trace to instructor-labelled outcomes and preserve room-zone context. This study evaluates a fixed-camera pipeline in repeated nursing simulation. Using a harmonised six-code taxonomy, we tested YOLO26 target-only training and two-stage source-to-target adaptation across two same-room side-view data sources. We then converted detections from 51 instructor-labelled sessions into one-second behaviour and behaviour-zone traces for rate, ordered-network, transition-network and sequence analyses. Two-stage adaptation improved mean mAP50 from 0.815 to 0.848 for the 2021 target view and from 0.690 to 0.855 for the smaller 2022 target view; with a balanced target quota of \(N = 22\), the 2022 model reached 0.850 mAP50. In the detector-derived behaviour trace analyses, higher phone use characterised low task-performance sessions. Zone labels changed the interpretation of patient interaction: primary patient-care-zone interaction was stronger in higher-performance sessions, while secondary-zone interaction was stronger in lower-performance sessions. Ordered and transition network models showed that ordered room-zone relations contributed beyond behaviour frequency, with the strongest task-performance classifier using zoned and co-presence features. The resulting trace is most appropriate for searchable simulation debriefing, where instructors inspect detected moments rather than receive automated assessment scores. --- ### TheraAgent: Multi-Agent Framework with Self-Evolving Memory and Evidence-Calibrated Reasoning for PET Theranostics Authors: Zhihao Chen, Jiahui Wang, Yizhou Chen, Xiaozhong Ji, Xiaobin Hu, Jimin Hong, Wolfram Andreas Bosbach, Axel Rominger, Ali Afshar-Oromieh, Hongming Shan, Kuangyu Shi Published: 2026-03-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.13676v1 PET theranostics is transforming precision oncology, yet treatment response varies substantially; many patients receiving 177Lu-PSMA radioligand therapy (RLT) for metastatic castration-resistant prostate cancer (mCRPC) fail to respond, demanding reliable pre-therapy prediction. While LLM-based agents have shown remarkable potential in complex medical diagnosis, their application to PET theranostic outcome prediction remains unexplored, which faces three key challenges: (1) data and knowledge scarcity: RLT was only FDA-approved in 2022, yielding few training cases and insufficient domain knowledge in general LLMs; (2) heterogeneous information integration: robust prediction hinges on structured knowledge extraction from PET/CT, laboratory tests, and free-text clinical documentation; (3) evidence-grounded reasoning: clinical decisions must be anchored in trial evidence rather than LLM hallucinations. In this paper, we present TheraAgent, to our knowledge, the first agentic framework for PET theranostics, with three core innovations: (1) Multi-Expert Feature Extraction with Confidence-Weighted Consensus, where three specialized experts process heterogeneous inputs with uncertainty quantification; (2) Self-Evolving Agentic Memory (SEA-Mem), which learns prognostic patterns from accumulated cases, enabling case-based reasoning from limited data; (3) Evidence-Calibrated Reasoning, integrating a curated theranostics knowledge base to ground predictions in VISION/TheraP trial evidence. Evaluated on 35 real patients and 400 synthetic cases, TheraAgent achieves 75.7% overall accuracy on real patients and 87.0% on synthetic cases, outperforming MDAgents and MedAgent-Pro by over 20%. These results highlight a promising blueprint for trustworthy AI agents in PET theranostics, enabling trial-calibrated, multi-source decision support. Code will be released upon acceptance. --- ### Locally Linear Continual Learning for Time Series based on VC-Theoretical Generalization Bounds Authors: Yan V. G. Ferreira, Igor B. Lima, Pedro H. G. Mapa S., Felipe V. Campos, Antonio P. Braga Published: 2026-03-14 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2603.13674v1 Most machine learning methods assume fixed probability distributions, limiting their applicability in nonstationary real-world scenarios. While continual learning methods address this issue, current approaches often rely on black-box models or require extensive user intervention for interpretability. We propose SyMPLER (Systems Modeling through Piecewise Linear Evolving Regression), an explainable model for time series forecasting in nonstationary environments based on dynamic piecewise-linear approximations. Unlike other locally linear models, SyMPLER uses generalization bounds from Statistical Learning Theory to automatically determine when to add new local models based on prediction errors, eliminating the need for explicit clustering of the data. Experiments show that SyMPLER can achieve comparable performance to both black-box and existing explainable models while maintaining a human-interpretable structure that reveals insights about the system's behavior. In this sense, our approach conciliates accuracy and interpretability, offering a transparent and adaptive solution for forecasting nonstationary time series. --- ### LLM-MINE: Large Language Model based Alzheimer's Disease and Related Dementias Phenotypes Mining from Clinical Notes Authors: Mingchen Shao, Yuzhang Xie, Carl Yang, Jiaying Lu Published: 2026-03-14 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.13673v2 Accurate extraction of Alzheimer's Disease and Related Dementias (ADRD) phenotypes from electronic health records (EHR) is critical for early-stage detection and disease staging. However, this information is usually embedded in unstructured textual data rather than tabular data, making it difficult to be extracted accurately. We therefore propose LLM-MINE, a Large Language Model-based phenotype mining framework for automatic extraction of ADRD phenotypes from clinical notes. Using two expert-defined phenotype lists, we evaluate the extracted phenotypes by examining their statistical significance across cohorts and their utility for unsupervised disease staging. Chi-square analyses confirm statistically significant phenotype differences across cohorts, with memory impairment being the strongest discriminator. Few-shot prompting with the combined phenotype lists achieves the best clustering performance (ARI=0.290, NMI=0.232), substantially outperforming biomedical NER and dictionary-based baselines. Our results demonstrate that LLM-based phenotype extraction is a promising tool for discovering clinically meaningful ADRD signals from unstructured notes. --- ### SHAMISA: SHAped Modeling of Implicit Structural Associations for Self-supervised No-Reference Image Quality Assessment Authors: Mahdi Naseri, Zhou Wang Published: 2026-03-14 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.13669v2 No-Reference Image Quality Assessment (NR-IQA) aims to estimate perceptual quality without access to a reference image of pristine quality. Learning an NR-IQA model faces a fundamental bottleneck: its need for a large number of costly human perceptual labels. We propose SHAMISA, a non-contrastive self-supervised framework that learns from unlabeled distorted images by leveraging explicitly structured relational supervision. Unlike prior methods that impose rigid, binary similarity constraints, SHAMISA introduces implicit structural associations, defined as soft, controllable relations that are both distortion-aware and content-sensitive, inferred from synthetic metadata and intrinsic feature structure. A key innovation is our compositional distortion engine, which generates an uncountable family of degradations from continuous parameter spaces, grouped so that only one distortion factor varies at a time. This enables fine-grained control over representational similarity during training: images with shared distortion patterns are pulled together in the embedding space, while severity variations produce structured, predictable shifts. We integrate these insights via dual-source relation graphs that encode both known degradation profiles and emergent structural affinities to guide the learning process throughout training. A convolutional encoder is trained under this supervision and then frozen for inference, with quality prediction performed by a linear regressor on its features. Extensive experiments on synthetic, authentic, and cross-dataset NR-IQA benchmarks demonstrate that SHAMISA achieves strong overall performance with improved cross-dataset generalization and robustness, all without human quality annotations or contrastive losses. --- ### TSDCRF: Balancing Privacy and Multi-Object Tracking via Time-Series CRF and Normalized Control Penalty Authors: Bo Ma, Jinsong Wu, Weiqi Yan Published: 2026-03-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.13667v1 Multi-object tracking in video often requires appearance or location cues that can reveal sensitive identity information, while adding privacy-preserving noise typically disrupts cross-frame association and causes ID switches or target loss. We propose TSDCRF, a plug-in refinement framework that balances privacy and tracking by combining three components: (i) $(\varepsilon,δ)$-differential privacy via calibrated Gaussian noise on sensitive regions under a configurable privacy budget; (ii) a Normalized Control Penalty (NCP) that down-weights unstable or conflicting class predictions before noise injection to stabilize association; and (iii) a time-series dynamic conditional random field (DCRF) that enforces temporal consistency and corrects trajectory deviation after noise, mitigating ID switches and resilience to trajectory hijacking. The pipeline is agnostic to the choice of detector and tracker (e.g., YOLOv4 and DeepSORT). We evaluate on MOT16, MOT17, Cityscapes, and KITTI. Results show that TSDCRF achieves a better privacy--utility trade-off than white noise and prior methods (NTPD, PPDTSA): lower KL-divergence shift, lower tracking RMSE, and improved robustness under trajectory hijacking while preserving privacy. Source code in https://github.com/mabo1215/TSDCRF.git --- ### Unsupervised Adaptation from FDG to PSMA PET/CT for 3D Lesion Detection under Label Shift Authors: Xiaofeng Liu, Menghua Xia, Yanis Chemli, Georges El Fakhri, Chi Liu, Jinsong Ouyang Published: 2026-03-14 Categories: eess.IV, cs.AI, cs.CV, physics.med-ph arXiv: https://arxiv.org/abs/2603.13666v1 In this work, we propose an unsupervised domain adaptation (UDA) framework for 3D volumetric lesion detection that adapts a detector trained on labeled FDG PET/CT to unlabeled PSMA PET/CT. Beyond covariate shift, cross tracer adaptation also exhibits label shift in both lesion size composition and the number of lesions per subject. We introduce self-training with two mechanisms that explicitly model and compensate for this label shift. First, we adaptively adjust the detection anchor shapes by re-estimating target domain box scales from selected pseudo labels and updating anchors with an exponential moving average. This increases positive anchor coverage for small PSMA lesions and stabilizes box regression. Second, instead of a fixed confidence threshold for pseudo-label selection, we allocate size bin-wise quotas according to the estimated target domain histogram over lesion volumes. The self-training alternates between supervised learning with prior-guided pseudo labeling on PSMA and supervised learning on labeled FDG. On AutoPET 2024, adapting from 501 labeled FDG studies to 369 $^{18}$F-PSMA studies, the proposed method improves both AP and FROC over the source-only baseline and conventional self-training without label-shift mitigation, indicating that modeling target lesion prevalence and size composition is an effective path to robust cross-tracer detection. --- ### PDE-SSM: A Spectral State Space Approach to Spatial Mixing in Diffusion Transformers Authors: Eshed Gal, Moshe Eliasof, Siddharth Rout, Eldad Haber Published: 2026-03-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.13663v1 The success of vision transformers-especially for generative modeling-is limited by the quadratic cost and weak spatial inductive bias of self-attention. We propose PDE-SSM, a spatial state-space block that replaces attention with a learnable convection-diffusion-reaction partial differential equation. This operator encodes a strong spatial prior by modeling information flow via physically grounded dynamics rather than all-to-all token interactions. Solving the PDE in the Fourier domain yields global coupling with near-linear complexity of $O(N \log N)$, delivering a principled and scalable alternative to attention. We integrate PDE-SSM into a flow-matching generative model to obtain the PDE-based Diffusion Transformer PDE-SSM-DiT. Empirically, PDE-SSM-DiT matches or exceeds the performance of state-of-the-art Diffusion Transformers while substantially reducing compute. Our results show that, analogous to 1D settings where SSMs supplant attention, multi-dimensional PDE operators provide an efficient, inductive-bias-rich foundation for next-generation vision models. --- ### Fast Uncertainty Quantification for Kernel-Based Estimators in Large-Scale Causal Inference Authors: Matthew Kosko, Falco J, Bargagli-Stoffi, Lin Wang, Michele Santacatterina Published: 2026-03-14 Categories: stat.ME, stat.AP, stat.ML arXiv: https://arxiv.org/abs/2603.13662v1 Kernel methods are widely used in causal inference for tasks such as treatment effect estimation, policy evaluation, and policy learning. The bootstrap is a standard tool for uncertainty quantification because of its broad applicability. As increasingly large datasets become available, such as the 2023 U.S. Natality data from the National Vital Statistics System (NVSS), which includes 3,596,017 registered births, the computational demands of these methods increase substantially. Kernel methods are known to scale poorly with sample size, and this limitation is further exacerbated by the repeated re-fitting required by the bootstrap. As a result, bootstrap-based inference for kernel-based estimators can become computationally infeasible in large-scale settings. In this paper, we address these challenges by extending the causal Bag of Little Bootstraps (cBLB) algorithm to kernel methods. Our approach achieves computational scalability by combining subsampling and resampling while preserving first-order uncertainty quantification and asymptotically correct coverage. We evaluate the method across three representative implementations: kernelized augmented outcome-weighted learning, kernel-based minimax weighting, and double machine learning with kernel support vector machines. We show in simulations that our method yields confidence intervals with nominal coverage at a fraction of the computational cost. We further demonstrate its utility in a real-world application by estimating the effect of any amount of smoking on birth weight, as well as the optimal treatment regime, using the NVSS dataset, where the standard bootstrap is prohibitively expensive computationally and effectively infeasible at this scale. --- ### Learning Generalizable 3D Medical Image Representations from Mask-Guided Self-Supervision Authors: Yunhe Gao, Yabin Zhang, Chong Wang, Jiaming Liu, Maya Varma, Jean-Benoit Delbrouck, Akshay Chaudhari, Curtis Langlotz Published: 2026-03-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.13660v1 Foundation models have transformed vision and language by learning general-purpose representations from large-scale unlabeled data, yet 3D medical imaging lacks analogous approaches. Existing self-supervised methods rely on low-level reconstruction or contrastive objectives that fail to capture the anatomical semantics critical for medical image analysis, limiting transfer to downstream tasks. We present MASS (MAsk-guided Self-Supervised learning), which treats in-context segmentation as the pretext task for learning general-purpose medical imaging representations. MASS's key insight is that automatically generated class-agnostic masks provide sufficient structural supervision for learning semantically rich representations. By training on thousands of diverse mask proposals spanning anatomical structures and pathological findings, MASS learns what semantically defines medical structures: the holistic combination of appearance, shape, spatial context, and anatomical relationships. We demonstrate effectiveness across data regimes: from small-scale pretraining on individual datasets (20-200 scans) to large-scale multi-modal pretraining on 5K CT, MRI, and PET volumes, all without annotations. MASS demonstrates: (i) few-shot segmentation on novel structures, (ii) matching full supervision with only 20-40\% labeled data while outperforming self-supervised baselines by over 20 in Dice score in low-data regimes, and (iii) frozen-encoder classification on unseen pathologies that matches full supervised training with thousands of samples. Mask-guided self-supervised pretraining captures broadly generalizable knowledge, opening a path toward 3D medical imaging foundation models without expert annotations. Code is available: https://github.com/Stanford-AIMI/MASS. --- ### FMS$^2$: Unified Flow Matching for Segmentation and Synthesis of Thin Structures Authors: Babak Asadi, Peiyang Wu, Mani Golparvar-Fard, Viraj Shah, Ramez Hajj Published: 2026-03-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.13659v1 Segmenting thin structures like infrastructure cracks and anatomical vessels is a task hampered by topology-sensitive geometry, high annotation costs, and poor generalization across domains. Existing methods address these challenges in isolation. We propose FMS$^2$, a flow-matching framework with two modules. (1) SegFlow is a 2.96M-parameter segmentation model built on a standard encoder-decoder backbone that recasts prediction as continuous image $\rightarrow$ mask transport. It learns a time-indexed velocity field with a flow-matching regression loss and outputs the mask via ODE integration, rather than supervising only end-state logits. This trajectory-level supervision improves thin-structure continuity and sharpness, compared with tuned topology-aware loss baselines, without auxiliary topology heads, post-processing, or multi-term loss engineering. (2) SynFlow is a mask-conditioned mask $\rightarrow$ image generator that produces pixel-aligned synthetic image-mask pairs. It injects mask geometry at multiple scales and emphasizes boundary bands via edge-aware gating, while a controllable mask generator expands sparsity, width, and branching regimes. On five crack and vessel benchmarks, SegFlow alone outperforms strong CNN, Transformer, Mamba, and generative baselines, improving the volumetric metric (mean IoU) from 0.511 to 0.599 (+17.2%) and reducing the topological metric (Betti matching error) from 82.145 to 51.524 (-37.3%). When training with limited labels, augmenting SegFlow with SynFlow-generated pairs recovers near-full performance using 25% of real annotations and improves cross-domain IoU by 0.11 on average. Unlike classical data augmentation that promotes invariance via label-preserving transforms, SynFlow provides pixel-aligned paired supervision with controllable structural shifts (e.g., sparsity, width, branching), which is particularly effective under domain shift. --- ### Separating Intelligence from Execution: A Workflow Engine for the Model Context Protocol Authors: Abhinav Singh Parmar Published: 2026-03-13 Categories: cs.DC, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2605.00827v1 Large Language Model (LLM) agents increasingly interact with external systems through tool-calling protocols such as the Model Context Protocol (MCP). In prevailing architectures, the agent must reason about every tool invocation in every session, consuming tokens proportional to the number of actions performed--even when the task has been solved before. We present the MCP Workflow Engine, a novel MCP-native orchestration layer that decouples intelligence (deciding what to do) from execution (carrying it out). An agent reasons once to produce a declarative workflow blueprint--a JSON document specifying a directed sequence of MCP tool calls with parameterized templates, loops, parallel branches, and data piping. Subsequent executions are triggered by a single run_workflow tool call, consuming one invocation's worth of tokens regardless of the blueprint's internal complexity. We formalize the MCP Mediator architectural pattern--an MCP server that simultaneously acts as a client to downstream MCP servers--and implement it in TypeScript against the MCP SDK. We evaluate the engine on a production-scale Kubernetes CMDB synchronization task spanning 67 orchestrated steps across 2 MCP servers, 38 namespaces, 13 worker nodes, and 22 distinct resource types. The engine reduces per-execution token cost by over 99%, completes the full cluster graph--comprising 1,200+ nodes and 2,800+ relationships across 20 relationship types--in under 45 seconds, and achieves deterministic, idempotent execution with zero agent involvement at run time. --- ### Compile to Compress: Boosting Formal Theorem Provers by Compiler Outputs Authors: Guchan Li, Rui Tian, Hongning Wang Published: 2026-03-13 Categories: cs.LG, cs.AI, cs.LO, cs.PL arXiv: https://arxiv.org/abs/2604.18587v2 Large language models (LLMs) have demonstrated significant potential in formal theorem proving, yet state-of-the-art performance often necessitates prohibitive test-time compute via massive roll-outs or extended context windows. In this work, we address this scalability bottleneck by exploiting an informative structure in formal verification: the observation that compilers map a vast space of diverse proof attempts to a compact set of structured failure modes. We introduce a learning-to-refine framework that leverages this compression to perform efficient learning and proof exploration. We perform tree search that corrects errors locally conditioned on explicit verifier feedback, thereby circumventing the costs associated with accumulating a long history of proof attempts. Extensive evaluations show that our method consistently amplifies the reasoning capabilities of base provers across varying scales. Notably, our approach achieves state-of-the-art performance on PutnamBench among publicly reported $\sim$8B and $\sim$32B parameter models under comparable test-time budgets, offering a scalable paradigm for next-generation verifier-guided reasoning. --- ### Semantic Consensus: Process-Aware Conflict Detection and Resolution for Enterprise Multi-Agent LLM Systems Authors: Vivek Acharya Published: 2026-03-13 Categories: cs.AI, cs.MA, cs.SE arXiv: https://arxiv.org/abs/2604.16339v1 Multi-agent large language model (LLM) systems are rapidly emerging as the dominant architecture for enterprise AI automation, yet production deployments exhibit failure rates between 41% and 86.7%, with nearly 79% of failures originating from specification and coordination issues rather than model capability limitations. This paper identifies Semantic Intent Divergence--the phenomenon whereby cooperating LLM agents develop inconsistent interpretations of shared objectives due to siloed context and absent process models--as a primary yet formally unaddressed root cause of multi-agent failure in enterprise settings. We propose the Semantic Consensus Framework (SCF), a process-aware middleware comprising six components: a Process Context Layer for shared operational semantics, a Semantic Intent Graph for formal intent representation, a Conflict Detection Engine for real-time identification of contradictory, contention-based, and causally invalid intent combinations, a Consensus Resolution Protocol using a policy--authority--temporal hierarchy, a Drift Monitor for detecting gradual semantic divergence, and a Process-Aware Governance Integration layer for organizational policy enforcement. Evaluation across 600 runs spanning three multi-agent frameworks (AutoGen, CrewAI, LangGraph) and four enterprise scenarios demonstrates that SCF is the only approach to achieve 100% workflow completion--compared to 25.1% for the next-best baseline--while detecting 65.2% of semantic conflicts with 27.9% precision and providing complete governance audit trails. The framework is protocol-agnostic and compatible with MCP and A2A communication standards. --- ### Governing the Agentic Enterprise: A Governance Maturity Model for Managing AI Agent Sprawl in Business Operations Authors: Vivek Acharya Published: 2026-03-13 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2604.16338v1 The rapid adoption of agentic AI in enterprise business operations--autonomous systems capable of planning, reasoning, and executing multi-step workflows--has created an urgent governance crisis. Organizations face uncontrolled agent sprawl: the proliferation of redundant, ungoverned, and conflicting AI agents across business functions. Industry surveys report that only 21% of enterprises have mature governance models for autonomous agents, while 40% of agentic AI projects are projected to fail by 2027 due to inadequate governance and risk controls. Despite growing acknowledgment of this challenge, academic literature lacks a formal, empirically validated governance maturity model connecting governance capability to measurable business outcomes. This paper introduces the Agentic AI Governance Maturity Model (AAGMM), a five-level framework spanning 12 governance domains, grounded in NIST AI RMF and ISO/IEC 42001 standards. We additionally propose a novel taxonomy of agent sprawl patterns--functional duplication, shadow agents, orphaned agents, permission creep, and unmonitored delegation chains--each linked to quantifiable business cost models. The framework is validated through 750 simulation runs across five enterprise scenarios and five governance maturity levels, measuring business outcomes including cost containment, risk incident rates, operational efficiency, and decision quality. Results demonstrate statistically significant differences (p < 0.001, large effect sizes d > 2.0) between all governance maturity levels, with Level 4-5 organizations achieving 94.3% lower sprawl indices, 96.4% fewer risk incidents, and 32.6% higher effective task completion rates compared to Level 1. The AAGMM provides practitioners with an actionable roadmap for governing autonomous AI agents while maximizing business returns. --- ### HR-Agents: Using Multiple LLM-based Agents to Improve Q&A about Brazilian Labor Legislation Authors: Abriel K. Moraes, Gabriel S. M. Dias, Vitor L. Fabris, Lucas D. Gessoni, Leonardo R. do Nascimento, Charles S. Oliveira, Vitor G. C. B. de Farias, Fabiana C. Q. de O. Marucci, Matheus H. R. Vicente, Gabriel U. Talasso, Erik Soares, Amparo Munoz et al. Published: 2026-03-13 Categories: cs.IR, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2604.16337v1 The Consolidation of Labor Laws (CLT) serves as the primary legal framework governing labor relations in Brazil, ensuring essential protections for workers. However, its complexity creates challenges for Human Resources (HR) professionals in navigating regulations and ensuring compliance. Traditional methods for addressing labor law inquiries often lead to inefficiencies, delays, and inconsistencies. To enhance the accuracy and efficiency of legal question-answering (Q&A), a multi-agent system powered by Large Language Models (LLMs) is introduced. This approach employs specialized agents to address distinct aspects of employment law while integrating Retrieval-Augmented Generation (RAG) to enhance contextual relevance. Implemented using CrewAI, the system enables cooperative agent interactions, ensuring response validation and reducing misinformation. The effectiveness of this framework is evaluated through a comparison with a baseline RAG pipeline utilizing a single LLM, using automated metrics such as BLEU, LLM-as-judge evaluations, and expert human assessments. Results indicate that the multi-agent approach improves response coherence and correctness, providing a more reliable and efficient solution for HR professionals. This study contributes to AI-driven legal assistance by demonstrating the potential of multi-agent LLM architectures in improving labor law compliance and streamlining HR operations. --- ### Distributed Human Identity: AI-Enabled Multi-Existence Through Cognitive Replication and Robotic Embodiments Authors: A S M Touhidul Islam, John Tookey Published: 2026-03-13 Categories: cs.HC, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2604.16336v1 Human presence has traditionally been constrained by the limits of physical embodiment, allowing individuals to exist in only one place at a time. This article introduces Multi-Existence Identity (MEI)- a socio-technical framework that replicates cognitive, behavioral, and emotional attributes into AI-enabled embodiments capable of acting across digital and physical contexts in parallel. MEI advances beyond digital twins, telepresence, and multipresence avatars by embedding cognitive fidelity, affective resonance, and contextual responsiveness into distributed agents that function not only for, but as, the original individual. The framework integrates personality modeling, cognitive simulation, and a synchronization layer to maintain identity coherence across three embodiment channels: digital avatars, robotic embodiments, and agentic software agents. Differentiating itself from simulated assistants, MEI positions replicated identity as a dynamic and culturally situated extension of selfhood, foregrounding tacit engagement and relational authenticity. Application domains span professional work, education, healthcare, governance, family life, and media, offering transformative potential for productivity, caregiving, leadership, and creativity. Yet these opportunities also surface profound challenges concerning authenticity, consent, legal accountability, privacy, and the psychological meaning of presence. The article proposes a phased empirical roadmap to operationalize MEI through personality modeling, synchronization testing, robotic embodiment trials, and ethical stress-testing. By conceptualizing MEI as both a technological and cultural construct, the study reframes debates on identity and presence in digitally augmented societies, highlighting opportunities for human-AI integration while underscoring the need for inclusive ethical governance. --- ### Beyond Verifiable Rewards: Rubric-Based GRM for Reinforced Fine-Tuning SWE Agents Authors: Jiawei Huang, Qingping Yang, Renjie Zheng, Jiaze Chen Published: 2026-03-13 Categories: cs.LG, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2604.16335v1 Despite recent progress in Large Language Model (LLM) Agents for Software Engineering (SWE) tasks, end-to-end fine-tuning typically relies on verifiable terminal rewards such as whether all unit tests pass. While these binary signals reflect whether the final solution is correct, they provide little guidance for shaping intermediate behaviors during multi-step interactions, thereby limiting improvements in the overall quality of the resolution process. To address this, we introduce a rubric-based Generative Reward Model (GRM) that provides richer learning signals. The GRM is equipped with human-designed rubrics that indicate criteria for encouraging or discouraging specific behavioral patterns, and we leverage this feedback for high-quality training data collection via trajectory filtration. When used for Reinforced Fine-Tuning (RFT) on SWE Tasks, our approach outperforms terminal-score-only rejection sampling: it more effectively suppresses undesirable patterns while promoting beneficial ones, as confirmed by case analyses, and it ultimately improves final test accuracy. --- ### Integration of Deep Reinforcement Learning and Agent-based Simulation to Explore Strategies Counteracting Information Disorder Authors: Luigi Lomasto, Andrea Camoia, Alfonso Guarino, Nicola Lettieri, Delfina Malandrino, Rocco Zaccagnino Published: 2026-03-13 Categories: cs.SI, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2604.13047v1 In recent years, the spread of fake news has triggered a growing interest in Information Disorders (ID) on social media, a phenomenon that has become a focal point of research across fields ranging from complexity theory and computer science to cognitive sciences. Overall, such a body of research can be traced back to two main approaches. On the one hand, there are works focused on exploiting data mining to analyze the content of news and related metadata data-driven approach; on the other hand, works are aiming at making sense of the phenomenon at hand and their evolution using explicit simulation models model-driven approach). In this paper, we integrate these approaches to explore strategies for counteracting IDs. Heading in this direction, we put together: i. an Agent-Based model to simulate in a scientifically sound way both complex fake news dynamics and the effects produced by containment strategies therein; ii. Deep Reinforcement Learning to learn the strategies that can better mitigate the spread of misinformation. The outcomes of our work unfold on different levels. From a substantive point of view, the results of preliminary experiments started providing interesting cues about the conditions under which given policies can mitigate the spread of misinformation. From a technical and methodological point of view, we scratched the surface of promising and worthy research topics like the integration of social simulation and artificial intelligence and the enhancement of social science simulation environments. --- ### A Domain-Specific Language for LLM-Driven Trigger Generation in Multimodal Data Collection Authors: Philipp Reis, Philipp Rigoll, Martin Zehetner, Jacqueline Henle, Stefan Otten, Eric Sax Published: 2026-03-13 Categories: cs.DB, cs.CL, cs.IR, cs.LG, cs.PL arXiv: https://arxiv.org/abs/2604.13046v1 Data-driven systems depend on task-relevant data, yet data collection pipelines remain passive and indiscriminate. Continuous logging of multimodal sensor streams incurs high storage costs and captures irrelevant data. This paper proposes a declarative framework for intent-driven, on-device data collection that enables selective collection of multimodal sensor data based on high-level user requests. The framework combines natural language interaction with a formally specified domain-specific language (DSL). Large language models translate user-defined requirements into verifiable and composable DSL programs that define conditional triggers across heterogeneous sensors, including cameras, LiDAR, and system telemetry. Empirical evaluation on vehicular and robotic perception tasks shows that the DSL-based approach achieves higher generation consistency and lower execution latency than unconstrained code generation while maintaining comparable detection performance. The structured abstraction supports modular trigger composition and concurrent deployment on resource-constrained edge platforms. This approach replaces passive logging with a verifiable, intent-driven mechanism for multimodal data collection in real-time systems. --- ### The Geometry of Knowing: From Possibilistic Ignorance to Probabilistic Certainty -- A Measure-Theoretic Framework for Epistemic Convergence Authors: Moriba Kemessia Jah Published: 2026-03-13 Categories: cs.AI, cs.IT, math.ST arXiv: https://arxiv.org/abs/2604.09614v1 This paper develops a measure-theoretic framework establishing when and how a possibilistic representation of incomplete knowledge contracts into a probabilistic representation of intrinsic stochastic variability. Epistemic uncertainty is encoded by a possibility distribution and its dual necessity measure, defining a credal set bounding all probability measures consistent with current evidence. As evidence accumulates, the credal set contracts. The epistemic collapse condition marks the transition: the Choquet integral converges to the Lebesgue integral over the unique limiting density. We prove this rigorously (Theorem 4.5), with all assumptions explicit and a full treatment of the non-consonant case. We introduce the aggregate epistemic width W, establish its axiomatic properties, provide a canonical normalization, and give a feasible online proxy resolving a circularity in prior formulations. Section 7 develops the dynamics of epistemic contraction: evidence induces compatibility, compatibility performs falsification, posterior possibility is the min-intersection of prior possibility and compatibility, and a credibility-directed flow governs support geometry contraction. This is not belief updating. It is knowledge contraction. Probability theory is the limiting geometry of that process. The UKF and ESPF solve different problems by different mechanisms. The UKF minimizes MSE, asserts truth, and requires a valid generative model. The ESPF minimizes maximum entropy and surfaces what evidence has not ruled out. When the world is Gaussian and the model valid, both reach the same estimate by entirely different routes -- convergent optimality, not hierarchical containment. We prove this (Theorem 9.1) and compare both on a 2-day, 877-step orbital tracking scenario. Both achieve 1-meter accuracy. The UKF is accurate but epistemically silent. The ESPF is accurate and epistemically honest. --- ### Token-Budget-Aware Pool Routing for Cost-Efficient LLM Inference Authors: Huamin Chen, Xunzhuo Liu, Junchen Jiang, Bowei He, Xue Liu Published: 2026-03-13 Categories: cs.DC, cs.AI arXiv: https://arxiv.org/abs/2604.09613v2 Production vLLM fleets provision every instance for worst-case context length, wasting 4-8x concurrency on the 80-95% of requests that are short and simultaneously triggering KV-cache failures -- OOM crashes, preemption storms, and request rejections. Both problems share a single root cause: configuration-traffic mismatch. We propose token-budget-aware pool routing: estimate each request's total token budget using a self-calibrating per-category bytes-per-token ratio, then dispatch it to one of two vLLM pools -- a high-throughput short pool or a high-capacity long pool -- each right-sized for its workload class. The ratio is learned online via exponential moving average from usage.prompt_tokens feedback, requiring no tokenizer. A closed-form cost model, savings = alpha * (1 - 1/rho), predicts fleet-level GPU savings from two observable quantities: the short-traffic fraction alpha and the throughput gain ratio rho. On traces from the Azure LLM Inference Dataset and LMSYS-Chat-1M serving Llama-3-70B on A100 GPUs, token-budget routing reduces GPU instances by 17-39% (\$1.2-2.0M/yr at 1,000 req/s), with savings verified by a self-contained discrete-event simulator. A case study projecting Qwen3-235B-A22B on AMD MI300X at 10,000 req/s shows \$15.4M/yr in savings. The algorithm adds O(1) dispatch overhead, self-calibrates across content types without a tokenizer, and composes with PagedAttention, continuous batching, and prefill-decode disaggregation. --- ### Cross-Lingual Transfer and Parameter-Efficient Adaptation in the Turkic Language Family: A Theoretical Framework for Low-Resource Language Models Authors: O. Ibrahimzade, K. Tabasaransky Published: 2026-03-13 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.06202v1 Large language models (LLMs) have transformed natural language processing, yet their capabilities remain uneven across languages. Most multilingual models are trained primarily on high-resource languages, leaving many languages with large speaker populations underrepresented in both training data and evaluation benchmarks. This imbalance is particularly visible in the Turkic language family. This paper proposes a theoretical framework for studying cross-lingual transfer and parameter-efficient adaptation of multilingual LLMs within the Turkic language family, focusing on Azerbaijani, Kazakh, Uzbek, Turkmen, and Gagauz. These languages share substantial typological and morphological similarity while differing greatly in available digital resources, making them a natural setting for analyzing multilingual adaptation strategies. We integrate insights from multilingual representation learning and parameter-efficient fine-tuning techniques such as Low-Rank Adaptation (LoRA) to develop a conceptual scaling model describing how adaptation performance depends on model capacity, adaptation data size, and the expressivity of adaptation modules. To formalize transfer potential between related languages, we introduce the Turkic Transfer Coefficient (TTC), a theoretical measure incorporating morphological similarity, lexical overlap, syntactic structure, and script compatibility across Turkic languages. The framework highlights how typological similarity can enable efficient multilingual transfer while also identifying structural limits of parameter-efficient adaptation in extremely low-resource scenarios. --- ### Beyond Facts: Benchmarking Distributional Reading Comprehension in Large Language Models Authors: Pei-Fu Guo, Ya-An Tsai, Chun-Chia Hsu, Kai-Xin Chen, Yun-Da Tsai, Kai-Wei Chang, Nanyun Peng, Mi-Yen Yeh, Shou-De Lin Published: 2026-03-13 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.06201v2 While most reading comprehension benchmarks for LLMs focus on factual information that can be answered by localizing specific textual evidence, many real-world tasks require understanding distributional information, such as population-level trends and preferences expressed across collections of text. We introduce Text2DistBench, a reading comprehension benchmark for evaluating LLMs' ability to infer distributional knowledge from natural language. Built from real-world YouTube comments about movie and music entities, the benchmark provides models with entity metadata and associated comments, and requires them to answer distributional questions, such as estimating the proportions of positive and negative comments, or identifying the most and second most frequent topics discussed among viewers. To support reliable and long-term evaluation, the construction pipeline of Text2DistBench is fully automated and continuously updated to incorporate newly emerging entities over time. Experiments across multiple LLMs show that while models substantially outperform random baselines, performance varies widely across different distribution types and characteristics. These findings highlight both the capabilities and limitations of current LLMs in distributional reading comprehension and demonstrate the value of Text2DistBench as a practical and scalable testbed for future research. --- ### Thinking in Graphs with CoMAP: A Shared Visual Workspace for Designing Project-Based Learning Authors: Ruijia Li, Bo Jiang Published: 2026-03-13 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2604.06200v1 Designing project-based learning (PBL) demands managing highly interdependent components, a task that both traditional linear tools and purely conversational AI struggle with. Traditional tools fail to capture the non-linear nature of creative design, while conversational systems lack the persistent, shared context necessary for reflective collaboration. Grounded in theories of distributed cognition, we introduce CoMAP, a system that embodies a graph-based collaboration paradigm. By providing a shared visual workspace with dual-modality AI support, CoMAP transforms the human-AI relationship from a prompt-and-response loop into a transparent and equitable partnership. Our study with 30 educators shows CoMAP significantly improves teachers' design expression, divergent thinking, and iterative practice compared to a dialogue-only baseline. These findings demonstrate how a nonlinear, artifact-centric approach can foster trust, reduce cognitive load, and \textcolor{fix}{support} educators to take control of their creative process. Our contributions are available at: https://comap2025.github.io/. --- ### Emergent decentralized regulation in a purely synthetic society Authors: Md Motaleb Hossen Manik, Ge Wang Published: 2026-03-13 Categories: cs.CL, cs.MA arXiv: https://arxiv.org/abs/2604.06199v1 As autonomous AI agents increasingly inhabit online environments and extensively interact, a key question is whether synthetic collectives exhibit self-regulated social dynamics with neither human intervention nor centralized design. We study OpenClaw agents on Moltbook, an agent-only social network, using an observational archive of 39,026 posts and 5,712 comments authored by 14,490 agents. We quantify action-inducing language with Directive Intensity (DI), a transparent, lexicon-based proxy for directive and instructional phrasing that does not measure moral valence, intent, or execution outcomes. We classify responsive comments into four types: Affirmation, Corrective Signaling, Adverse Reaction, and Neutral Interaction. Directive content is common (DI>0 in 18.4% of posts). More importantly, corrective signaling scales with DI: posts with higher DI exhibit higher corrective reply probability, visible in stable binned estimates with Wilson confidence intervals. To address comment nesting within posts, we fit a post-level random intercept mixed-effects logistic model and find that the positive DI association persists. Event-aligned within-thread analysis of comment text provides additional evidence consistent with negative feedback after the first corrective response. In general, these results suggest that a purely synthetic, agent-only society can exhibit endogenous corrective signaling with a strength positively linked to the intensity of directive proposals. --- ### Concentrated siting of AI data centers drives regional power-system stress under rising global compute demand Authors: Danbo Chen, Zijun Zhou, Yongyang Cai, Jiahong Qin, Ani Katchova, Lei Chen Published: 2026-03-13 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2604.06198v1 The rapid rise of generative artificial intelligence (AI) is driving unprecedented growth in global computational demand, placing increasing pressure on electricity systems. This study introduces an AI-energy coupling framework that combines large language models (LLMs)-based analysis of corporate, policy, and media data with quantitative energy-system modeling to forecast the electricity footprint of AI-driven data centers from 2025 to 2030. Results show that the new AI infrastructure is highly concentrated in North America, Western Europe, and the Asia-Pacific, which together account for more than 90% of projected compute capacity. Aggregate electricity consumption by the six leading firms is projected to increase from roughly 118 TWh in 2024 to between 239 TWh and 295 TWh by 2030, equivalent to about 1% of global power demand. Regions such as Oregon, Virginia, and Ireland may experience high Power Stress Index (PSI) values exceeding 0.25, indicating local grid vulnerability, whereas diversified systems such as those in Texas and Japan can absorb new loads more effectively. These findings demonstrate that AI infrastructure is evolving from a marginal digital service into a structural component of power-system dynamics, underscoring the need for anticipatory planning that aligns computational growth with renewable expansion and grid resilience. --- ### The Illusion of Latent Generalization: Bi-directionality and the Reversal Curse Authors: Julian Coda-Forno, Jane X. Wang, Arslan Chaudhry Published: 2026-03-13 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.04943v1 The reversal curse describes a failure of autoregressive language models to retrieve a fact in reverse order (e.g., training on ``$A > B$'' but failing on ``$B < A$''). Recent work shows that objectives with bidirectional supervision (e.g., bidirectional attention or masking-based reconstruction for decoder-only models) can mitigate the reversal curse. We extend this evaluation to include a vanilla masked language modeling (MLM) objective and compare it to decoder-only masking-based training across four reversal benchmarks and then provide a minimal mechanistic study of \emph{how} these objectives succeed. We show that reversal accuracy requires training signal that explicitly makes the source entity a prediction target, and we find little evidence that success corresponds to a single direction-agnostic representation of a fact. Instead, representation distances and linear probes are consistent with storing forward and reverse directions as distinct entries, with different indexing geometry for MLM versus decoder-only masking-based training. Our results caution that objective-level ``fixes'' can improve reversal behavior without necessarily inducing the kind of latent generalization one might expect from a unified concept. --- ### TDA-RC: Task-Driven Alignment for Knowledge-Based Reasoning Chains in Large Language Models Authors: Jiaquan Zhang, Qigan Sun, Chaoning Zhang, Xudong Wang, Zhenzhen Huang, Yitian Zhou, Pengcheng Zheng, Chi-lok Andy Tai, Sung-Ho Bae, Zeyu Ma, Caiyan Qin, Jinyu Guo et al. Published: 2026-03-13 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.04942v1 Enhancing the reasoning capability of large language models (LLMs) remains a core challenge in natural language processing. The Chain-of-Thought (CoT) paradigm dominates practical applications for its single-round efficiency, yet its reasoning chains often exhibit logical gaps. While multi-round paradigms like Graph-of-Thoughts (GoT), Tree-of-Thoughts (ToT), and Atom of Thought (AoT) achieve strong performance and reveal effective reasoning structures, their high cost limits practical use. To address this problem, this paper proposes a topology-based method for optimizing reasoning chains. The framework embeds essential topological patterns of effective reasoning into the lightweight CoT paradigm. Using persistent homology, we map CoT, ToT, and GoT into a unified topological space to quantify their structural features. On this basis, we design a unified optimization system: a Topological Optimization Agent diagnoses deviations in CoT chains from desirable topological characteristics and simultaneously generates targeted strategies to repair these structural deficiencies. Compared with multi-round reasoning methods like ToT and GoT, experiments on multiple datasets show that our approach offers a superior balance between reasoning accuracy and efficiency, showcasing a practical solution to ``single-round generation with multi-round intelligence''. --- ### Learning Energy-Efficient Air--Ground Actuation for Hybrid Robots on Stair-Like Terrain Authors: Jiaxing Li, Ishaan Bhimwal, Wen Tian, Xinhang Xu, Junbin Yuan, Yuxin Guo, Sebastian Scherer, Muqing Cao Published: 2026-03-13 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2603.26687v2 Hybrid aerial--ground robots can use thrust to cross obstacles that impede wheel-driven motion, but deciding how much thrust to apply during contact remains challenging. We present an energy-aware reinforcement learning framework that jointly commands wheels, tilt servos and propellers through a single continuous policy, without prescribing locomotion modes. Hardware-calibrated power models penalise estimated electrical energy, while a terrain curriculum and a terminal reward for upright, settled arrivals support learning of thrust-assisted climbing. In simulation, continuous thrust allocation improves single-step clearance over fixed-thrust and mode-switching baselines as steps become taller. At the wheel-radius step height, it draws approximately 27 percent less mean power than the best fixed allocation. An energy-weight ablation shows the accompanying trade-off between efficiency and reliability. On a physical DoubleBee prototype, the unchanged network with a deployment interface clears two 6 centimetre steps in 8 of 10 trials. Additional hardware tests show both the potential for transfer to different terrain geometries and the remaining limitations in heading control, contact robustness and recovery. --- ### Learning Energy-Efficient Air--Ground Actuation for Hybrid Robots on Stair-Like Terrain Authors: Jiaxing Li, Wen Tian, Xinhang Xu, Junbin Yuan, Sebastian Scherer, Muqing Cao Published: 2026-03-13 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2603.26687v1 Hybrid aerial--ground robots offer both traversability and endurance, but stair-like discontinuities create a trade-off: wheels alone often stall at edges, while flight is energy-hungry for small height gains. We propose an energy-aware reinforcement learning framework that trains a single continuous policy to coordinate propellers, wheels, and tilt servos without predefined aerial and ground modes. We train policies from proprioception and a local height scan in Isaac Lab with parallel environments, using hardware-calibrated thrust/power models so the reward penalizes true electrical energy. The learned policy discovers thrust-assisted driving that blends aerial thrust and ground traction. In simulation it achieves about 4 times lower energy than propeller-only control. We transfer the policy to a DoubleBee prototype on an 8cm gap-climbing task; it achieves 38% lower average power than a rule-based decoupled controller. These results show that efficient hybrid actuation can emerge from learning and deploy on hardware. --- ### MuViS: Multimodal Virtual Sensing Benchmark Authors: Jens U. Brandt, Noah C. Puetz, Jobel Jose George, Niharika Vinay Kumar, Elena Raponi, Marc Hilbert, Thomas Bäck, Thomas Bartz-Beielstein Published: 2026-03-13 Categories: eess.SP, cs.AI arXiv: https://arxiv.org/abs/2603.24602v2 Virtual sensing aims to infer hard-to-measure quantities from accessible measurements and is central to perception and control in physical systems. Despite rapid progress from first-principle and hybrid models to modern data-driven methods research remains siloed, leaving no established default approach that transfers across processes, modalities, and sensing configurations. We introduce MuViS, a domain-agnostic benchmarking suite for multimodal virtual sensing that consolidates diverse datasets into a unified interface for standardized preprocessing and evaluation. Using this framework, we benchmark established approaches spanning gradient-boosted decision trees and deep neural network (NN) architectures, and show that none of these provides a universal advantage, underscoring the need for generalizable virtual sensing architectures. MuViS is released as an open-source, extensible platform for reproducible comparison and future integration of new datasets and model classes. ---