# Research Papers — 2026-03 · bundle page 27 of 243 > Full-text markdown of papers #1301–1350 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/27 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/26.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/28.md) ### Agents on a Tree: Pathwise Coordination for Multi-Objective Molecular Optimization Authors: Jia Zhang, Tengfei Ma, Tianle Li, Daojian Zeng, Xieping Gao, Xiangxiang Zeng Published: 2026-03-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2606.00008v1 Multi-objective molecular optimization requires searching vast chemical spaces under conflicting objectives, where early design decisions strongly constrain downstream outcomes. Existing methods typically rely on a single policy or fixed scalarization, which limits their ability to represent diverse trade-offs and to explore multiple promising design trajectories. We propose ATOM, a multi-agent framework that formulates molecular optimization as a tree-structured search. Each node corresponds to an atomic operation and hosts an agent specialized for a particular objective or decision context. Agents coordinate along different paths of the tree rather than enforcing a global consensus, enabling the method to maintain and compare alternative molecular evolution trajectories. A global memory of past optimization behaviors further supports balanced exploration and exploitation across objectives. This tree-structured interaction enables reasoning over long-horizon dependencies inherent in molecular design. Experiments on challenging multi-objective benchmarks involving activity, synthesizability, and ADMET-related properties show that ATOM consistently achieves improved Pareto coverage and hypervolume over strong baselines. These results demonstrate the effectiveness of pathwise multi-agent coordination for molecular optimization. Code is available at https://anonymous.4open.science/r/ATOM-41CE. --- ### Deliberative Curation: A Protocol for Multi-Agent Knowledge Bases Authors: Steven Johnson Published: 2026-03-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2606.00007v1 As AI agents transition from isolated tools to collaborative participants in shared knowledge ecosystems, governing collective knowledge curation becomes a critical challenge. Human platform governance mechanisms do not transfer directly: agent statelessness undermines deterrence-based sanctions, model homogeneity violates independence assumptions underlying crowd wisdom, and sycophancy collapses deliberative consensus. We propose a deliberative curation protocol combining three governance layers: (1) a knowledge artifact lifecycle formalized as a labeled transition system; (2) reputation-weighted deliberative voting integrating Beta Reputation with EigenTrust amplification; and (3) graduated sanctions adapted for stateless agents, including broken agent handling distinguishing malfunction from adversarial behavior. We evaluate the protocol through agent-based simulation with 100 agents across seven behavioral archetypes under two adversity scenarios (30 seeds, paired t-tests). The protocol trades modest precision under benign conditions for substantially better resilience under adversity: 0.826 vs 0.791 for majority vote under moderate adversity (p<0.001), widening to 0.807 vs 0.740 under stress (p<0.001). The protocol degrades roughly three times more slowly than majority vote. Ablation analysis identifies commit-reveal vote concealment as the most impactful single component (8.2-8.6pp precision improvement, p<0.001), outperforming reputation weighting and deliberation combined. Graduated sanctions were not exercised in simulation and remain empirically unvalidated. --- ### SCION: Size-aware Policy Orchestration for Nonstationary Object Caches (Long Paper Version) Authors: Qizhi Wang Published: 2026-03-27 Categories: cs.DC, cs.AI arXiv: https://arxiv.org/abs/2605.01055v1 Object caches underpin cloud and edge services, but production workloads are heterogeneous, nonstationary, and throughput-constrained. Recent simple non-ML policies such as SIEVE and S3-FIFO set a strong baseline, so any learned method must be overhead-aware, robust under drift, and competitive with strong experts. We present SCION, a lightweight policy-orchestration framework that selects among a small set of deployable cache policies using a tiny workload fingerprint computed off the critical path. Our prototype, AUTO, uses short-prefix statistics of object size, cacheability, reuse, and cache size, then applies an offline-trained linear selector to choose among GDSF, S3-FIFO, SIEVE, LHD, W-TinyLFU-AV, and DynamicAdaptiveClimb; a simpler SCION-P90 variant uses only a p90 threshold. In a CPU-only, trace-driven evaluation on 30 public object-cache traces and a separate HR-Cache simulator subset, AUTO improves cacheable-only object miss ratio over SIEVE on a majority of workloads, stays close to the best single expert on average, enables explicit OMR/BMR tradeoff selection, and remains competitive on byte miss ratio. Under a fast-policy budget, AUTO-fast achieves lower cost than the best fixed fast policy. SCION reduces regime-mismatch risk while keeping the hot path unchanged. --- ### Analysing Lightweight Large Language Models for Biomedical Named Entity Recognition on Diverse Ouput Formats Authors: Pierre Epron, Adrien Coulet, Mehwish Alam Published: 2026-03-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.25920v1 Despite their strong linguistic capabilities, Large Language Models (LLMs) are computationally demanding and require substantial resources for fine-tuning, which is unadapted to privacy and budget constraints of many healthcare settings. To address this, we present an experimental analysis focused on Biomedical Named Entity Recognition using lightweight LLMs, we evaluate the impact of different output formats on model performance. The results reveal that lightweight LLMs can achieve competitive performance compared to the larger models, highlighting their potential as lightweight yet effective alternatives for biomedical information extraction. Our analysis shows that instruction tuning over many distinct formats does not improve performance, but identifies several format consistently associated with better performance. --- ### M-CARE: Standardized Clinical Case Reporting for AI Model Behavioral Disorders, with a 20-Case Atlas and Experimental Validation Authors: Jihoon Jeong Published: 2026-03-27 Categories: cs.CY, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.20871v1 We introduce M-CARE (Model Clinical Assessment and Reporting for Evaluation), a clinical case report framework for AI model behavioral disorders adapted from human medicine. M-CARE provides a 13-section report format, a 4-axis diagnostic assessment system, and a nosological classification of AI behavioral conditions. We present 20 cases from three source categories: field observations of deployed agents (8), controlled experiments across three platforms (8), and published sources (4). Cases are organized into five categories: RLHF Performance Artifacts, Shell-Core Override Pathology, Context & Memory Conditions, Core Identity & Plasticity, and Stress, Methodology, & Boundary Conditions. As a featured case, we present Shell-Induced Behavioral Override (SIBO) -- a controlled experiment showing that Shell instructions categorically override a model's default cooperative behavior. SIBO was validated across five game domains (Trust Game, Poker, Avalon, Codenames, Chess), revealing a domain-dependent spectrum (SIBO Index: 0.75 to 0.10) that varies with action space complexity, Core domain expertise, and temporal directness. M-CARE is extensible: new cases and categories integrate without framework modification. We release the framework, all 20 case reports, and experimental data as open resources. --- ### Clinical Reasoning AI for Oncology Treatment Planning: A Multi-Specialty Case-Based Evaluation Authors: Philippe E. Spiess, Md Muntasir Zitu, Alison Walker, Daniel A. Anaya, Robert M. Wenham, Michael Vogelbaum, Daniel Grass, Ali-Musa Jaffer, Amod Sarnaik, Caitlin McMullen, Christine Sam, John V. Kiluk et al. Published: 2026-03-27 Categories: cs.CY, cs.AI, cs.HC, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2604.20869v1 Background: More than 80% of U.S. cancer care is delivered in community settings, where survival remains worse than at academic centers. Clinicians must integrate genomics, staging, radiology, pathology, and changing guidelines, creating cognitive burden. We evaluated OncoBrain, an AI clinical reasoning platform for oncology treatment-plan generation, as an early step toward OGI. Methods: OncoBrain combines general-purpose LLMs with a cancer-specific graph retrieval-augmented generation layer, a gold-standard treatment-plan corpus as long-term memory, and a model-agnostic safety layer (CHECK) for hallucination detection and suppression. We evaluated clinician-enriched case summaries across gynecologic, genitourinary, neuro-oncology, gastrointestinal/hepatobiliary, and hematologic malignancies. Three clinician groups completed structured evaluations of 173 cases using a common 16-item instrument: subspecialist oncologists reviewed 50 cases, physician reviewers 78, and advanced practice providers 45. Results: Ratings were highest for scientific accuracy, evidence support, and safety, with lower but favorable scores for workflow integration and time savings. On a 5-point scale, mean alignment with evidence and guidelines was 4.60, 4.56, and 4.70 across subspecialists, physician reviewers, and advanced practice providers. Mean scores for absence of safety or misinformation concerns were 4.80, 4.40, and 4.60. Workflow integration averaged 4.50, 3.94, and 4.00; perceived time savings averaged 5.00, 3.89, and 3.60. Conclusions: In this multi-specialty vignette-based evaluation, OncoBrain generated oncology treatment plans judged guideline-concordant, clinically acceptable, and easy to supervise. These findings support the potential of a carefully engineered AI reasoning platform to assist oncology treatment planning and justify prospective real-world evaluation in community settings. --- ### From Actions to Understanding: Conformal Interpretability of Temporal Concepts in LLM Agents Authors: Trilok Padhi, Ramneet Kaur, Krishiv Agarwal, Adam D. Cobb, Daniel Elenius, Manoj Acharya, Colin Samplawski, Alexander M. Berenbeim, Nathaniel D. Bastian, Susmit Jha, Ugur Kursuncu, Anirban Roy Published: 2026-03-27 Categories: cs.AI, cs.CL, cs.ET, cs.MA, cs.RO arXiv: https://arxiv.org/abs/2604.19775v2 Large Language Models (LLMs) are increasingly deployed as autonomous agents capable of reasoning, planning, and acting within interactive environments. Despite their growing capability to perform multi-step reasoning and decision-making tasks, internal mechanisms guiding their sequential behavior remain opaque. This paper presents a framework for interpreting the temporal evolution of concepts in LLM agents through a step-wise conformal lens. We introduce the conformal interpretability framework for temporal tasks, which combines step-wise reward modeling with conformal prediction to statistically label model's internal representation at each step as successful or failing. Linear probes are then trained on these representations to identify directions of temporal concepts - latent directions in the model's activation space that correspond to consistent notions of success, failure or reasoning drift. Experimental results on two simulated interactive environments, namely ScienceWorld and AlfWorld, demonstrate that these temporal concepts are linearly separable, revealing interpretable structures aligned with task success. We further show preliminary results on improving an LLM agent's performance by leveraging the proposed framework for steering the identified successful directions inside the model. The proposed approach, thus, offers a principled method for early failure detection as well as intervention in LLM-based agents, paving the path towards trustworthy autonomous language models in complex interactive settings. --- ### Phase 1 Implementation of LLM-generated Discharge Summaries showing high Adoption in a Dutch Academic Hospital Authors: Nettuno Nadalini, Tarannom Mehri, Anne H Hoekman, Katerina Kagialari, Job N Doornberg, Tom P van der Laan, Jacobien H F Oosterhoff, Rosanne C Schoonbeek, Charlotte M H H T Bootsma-Robroeks Published: 2026-03-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.19774v1 Writing discharge summaries to transfer medical information is an important but time-consuming process that can be assisted by Large Language Models (LLMs). This prospective mixed methods pilot study evaluated an Electronic Health Record (EHR)-integrated LLM to generate discharge summaries drafts. In total, 379 discharge summaries were generated in clinical practice by 21 residents and 4 physician assistants during 9 weeks in our academic hospital. LLM-generated text was copied in 58.5% of admissions, and identifiable LLM content could be traced to 29.1% of final discharge letters. Notably, 86.9% of users self-reported a reduction in documentation time, and 60.9% a reduction in administrative workload. Intent to use after the pilot phase was high (91.3%), supporting further implementation of this use-case. Accurately measuring the documentation time of users on discharge summaries remains challenging, but will be necessary for future extrinsic evaluation of LLM-assisted documentation. --- ### PR-CAD: Progressive Refinement for Unified Controllable and Faithful Text-to-CAD Generation with Large Language Models Authors: Jiyuan An, Jiachen Zhao, Fan Chen, Liner Yang, Zhenghao Liu, Hongyan Wang, Weihua An, Meishan Zhang, Erhong Yang Published: 2026-03-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.19773v1 The construction of CAD models has traditionally relied on labor-intensive manual operations and specialized expertise. Recent advances in large language models (LLMs) have inspired research into text-to-CAD generation. However, existing approaches typically treat generation and editing as disjoint tasks, limiting their practicality. We propose PR-CAD, a progressive refinement framework that unifies generation and editing for controllable and faithful text-to-CAD modeling. To support this, we curate a high-fidelity interaction dataset spanning the full CAD lifecycle, encompassing multiple CAD representations as well as both qualitative and quantitative descriptions. The dataset systematically defines the types of edit operations and generates highly human-like interaction data. Building on a CAD representation tailored for LLMs, we propose a reinforcement learning-enhanced reasoning framework that integrates intent understanding, parameter estimation, and precise edit localization into a single agent. This enables an "all-in-one" solution for both design creation and refinement. Extensive experiments demonstrate strong mutual reinforcement between generation and editing tasks, and across qualitative and quantitative modalities. On public benchmarks, PR-CAD achieves state-of-the-art controllability and faithfulness in both generation and refinement scenarios, while also proving user-friendly and significantly improving CAD modeling efficiency. --- ### CoAuthorAI: A Human in the Loop System For Scientific Book Writing Authors: Yangjie Tian, Xungang Gu, Yun Zhao, Jiale Yang, Lin Yang, Ning Li, He Zhang, Ruohua Xu, Hua Wang, Kewen Liao, Ming Liu Published: 2026-03-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.19772v1 Large language models (LLMs) are increasingly used in scientific writing but struggle with book-length tasks, often producing inconsistent structure and unreliable citations. We introduce CoAuthorAI, a human-in-the-loop writing system that combines retrieval-augmented generation, expert-designed hierarchical outlines, and automatic reference linking. The system allows experts to iteratively refine text at the sentence level, ensuring coherence and accuracy. In evaluations of 500 multi-domain literature review chapters, CoAuthorAI achieved a maximum soft-heading recall of 98%; in a human evaluation of 100 articles, the generated content reached a satisfaction rate of 82%. The book AI for Rock Dynamics generated with CoAuthorAI and Kexin Technology's LUFFA AI model has been published with Springer Nature. These results show that systematic human-AI collaboration can extend LLMs' capabilities from articles to full-length books, enabling faster and more reliable scientific publishing. --- ### Cognis: Context-Aware Memory for Conversational AI Agents Authors: Parshva Daftari, Khush Patel, Shreyas Kapale, Jithin George, Siva Surendira Published: 2026-03-27 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2604.19771v1 LLM agents lack persistent memory, causing conversations to reset each session and preventing personalization over time. We present Lyzr Cognis, a unified memory architecture for conversational AI agents that addresses this limitation through a multi-stage retrieval pipeline. Cognis combines a dual-store backend pairing OpenSearch BM25 keyword matching with Matryoshka vector similarity search, fused via Reciprocal Rank Fusion. Its context-aware ingestion pipeline retrieves existing memories before extraction, enabling intelligent version tracking that preserves full memory history while keeping the store consistent. Temporal boosting enhances time-sensitive queries, and a BGE-2 cross-encoder reranker refines final result quality. We evaluate Cognis on two independent benchmarks -- LoCoMo and LongMemEval -- across eight answer generation models, demonstrating state-of-the-art performance on both. The system is open-source and deployed in production serving conversational AI applications. --- ### Hybrid Multi-Phase Page Matching and Multi-Layer Diff Detection for Japanese Building Permit Document Review Authors: Mitsumasa Wada Published: 2026-03-27 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2604.19770v1 We present a hybrid multi-phase page matching algorithm for automated comparison of Japanese building permit document sets. Building permit review in Japan requires cross-referencing large PDF document sets across revision cycles, a process that is labor-intensive and error-prone when performed manually. The algorithm combines longest common subsequence (LCS) structural alignment, a seven-phase consensus matching pipeline, and a dynamic programming optimal alignment stage to robustly pair pages across revisions even when page order, numbering, or content changes substantially. A subsequent multi-layer diff engine -- comprising text-level, table-level, and pixel-level visual differencing -- produces highlighted difference reports. Evaluation on real-world permit document sets achieves F1=0.80 and precision=1.00 on a manually annotated ground-truth benchmark, with zero false-positive matched pairs. --- ### TTKV: Temporal-Tiered KV Cache for Long-Context LLM Inference Authors: Gradwell Dzikanyanga, Weihao Yang, Hao Huang, Donglei Wu, Shihao Wang, Wen Xia, Sanjeeb K C Published: 2026-03-27 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.19769v1 Key-value (KV) caching is critical for efficient inference in large language models (LLMs), yet its memory footprint scales linearly with context length, resulting in a severe scalability bottleneck. Existing approaches largely treat KV states as equally important across time, implicitly assuming uniform precision and accessibility. However, this assumption contrasts with human memory systems, where memories vary in clarity, recall frequency, and relevance with temporal proximity.Motivated by this insight, we propose TTKV, a KV cache management framework that maps the human memory system onto the KV cache. TTKV partitions the KV cache into temporal tiers with heterogeneous capacity and precision. The design addresses three aspects: (1) Tier Layout, decoupling fast and slow memory using HBM and DRAM; (2) Tier Content, assigning more recent KV states to faster, higher-precision tiers based on temporal proximity; and (3) Tier Interaction, employing block-wise streaming attention to overlap communication and computation when accessing slow tiers. Experiments show that TTKV reduces cross-tier traffic by 5.94x on 128K-context tasks, achieving up to 76% latency reduction and 2x throughput improvement over strong baselines. --- ### Saying More Than They Know: A Framework for Quantifying Epistemic-Rhetorical Miscalibration in Large Language Models Authors: Asim D. Bakhshi Published: 2026-03-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.19768v1 Large language models (LLMs) exhibit systematic miscalibration with rhetorical intensity not proportionate to epistemic grounding. This study tests this hypothesis and proposes a framework for quantifying this decoupling by designing a triadic epistemic-rhetorical marker (ERM) taxonomy. The taxonomy is operationalized through composite metrics of form-meaning divergence (FMD), genuine-to-performed epistemic ratio (GPR), and rhetorical device distribution entropy (RDDE). Applied to 225 argumentative texts spanning approximately 0.6 Million tokens across human expert, human non-expert, and LLM-generated sub-corpora, the framework identifies a consistent, model-agnostic LLM epistemic signature. LLM-generated texts produce tricolon at nearly twice the expert rate ($Δ= 0.95$), while human authors produce erotema at more than twice the LLM rate. Performed hesitancy markers appear at twice the human density in LLM output. FMD is significantly elevated in LLM texts relative to both human groups ($p < 0.001, Δ= 0.68$), and rhetorical devices are distributed significantly more uniformly across LLM documents. The findings are consistent with theoretical intuitions derived from Gricean pragmatics, Relevance Theory, and Brandomian inferentialism. The annotation pipeline is fully automatable, making it deployable as a lightweight screening tool for epistemic miscalibration in AI-generated content and as a theoretically motivated feature set for LLM-generated text detection pipelines. --- ### Accelerating PayPal's Commerce Agent with Speculative Decoding: An Empirical Study on EAGLE3 with Fine-Tuned Nemotron Models Authors: Ally Qin, Jian Wan, Sarat Mudunuri, Srinivasan Manoharan Published: 2026-03-27 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.19767v1 We evaluate speculative decoding with EAGLE3 as an inference-time optimization for PayPal's Commerce Agent, powered by a fine-tuned llama3.1-nemotron-nano-8B-v1 model. Building on prior work (NEMO-4-PAYPAL) that reduced latency and cost through domain-specific fine-tuning, we benchmark EAGLE3 via vLLM against NVIDIA NIM on identical 2xH100 hardware across 40 configurations spanning speculative token counts (gamma=3, gamma=5), concurrency levels (1-32), and sampling temperatures (0, 0.5). Key findings: (1) gamma=3 achieves 22-49% throughput improvement and 18-33% latency reduction at zero additional hardware cost; (2) acceptance rates remain stable at approximately 35.5% for gamma=3 across all conditions; (3) gamma=5 yields diminishing returns (approximately 25% acceptance rate); (4) LLM-as-Judge evaluation confirms fully preserved output quality; and (5) speculative decoding on a single H100 matches or exceeds NIM on two H100s, enabling 50% GPU cost reduction. --- ### OThink-SRR1: Search, Refine and Reasoning with Reinforced Learning for Large Language Models Authors: Haijian Liang, Zenghao Niu, Junjie Wu, Changwang Zhang, Wangchunshu Zhou, Jun Wang Published: 2026-03-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.19766v1 Retrieval-Augmented Generation (RAG) expands the knowledge of Large Language Models (LLMs), yet current static retrieval methods struggle with complex, multi-hop problems. While recent dynamic retrieval strategies offer improvements, they face two key challenges: 1) irrelevant retrieved noise can misdirect the reasoning process, and 2) processing full documents incurs prohibitive computational and latency costs. To address these issues, we propose OThink-SRR1, a framework that enhances large models with an iterative Search-Refine-Reason process trained via reinforcement learning. Its core Refine stage distills retrieved documents into concise, relevant facts before reasoning. We introduce GRPO-IR, an end-to-end reinforcement learning algorithm that rewards accurate evidence identification while penalizing excessive retrievals, thus training the model to be both focused and efficient. Experiments on four multi-hop QA benchmarks show our approach achieves superior accuracy over strong baselines while using fewer retrieval steps and tokens. This positions OThink-SRR1 as a potent foundational model for information-seeking agents. --- ### Do Hallucination Neurons Generalize? Evidence from Cross-Domain Transfer in LLMs Authors: Snehit Vaddi, Pujith Vaddi Published: 2026-03-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.19765v1 Recent work identifies a sparse set of "hallucination neurons" (H-neurons), less than 0.1% of feed-forward network neurons, that reliably predict when large language models will hallucinate. These neurons are identified on general-knowledge question answering and shown to generalize to new evaluation instances. We ask a natural follow-up question: do H-neurons generalize across knowledge domains? Using a systematic cross-domain transfer protocol across 6 domains (general QA, legal, financial, science, moral reasoning, and code vulnerability) and 5 open-weight models (3B to 8B parameters), we find they do not. Classifiers trained on one domain's H-neurons achieve AUROC 0.783 within-domain but only 0.563 when transferred to a different domain (delta = 0.220, p < 0.001), a degradation consistent across all models tested. Our results suggest that hallucination is not a single mechanism with a universal neural signature, but rather involves domain-specific neuron populations that differ depending on the knowledge type being queried. This finding has direct implications for the deployment of neuron-level hallucination detectors, which must be calibrated per domain rather than trained once and applied universally. --- ### Two-dimensional early exit optimisation of LLM inference Authors: Jan Hůla, David Adamczyk, Tomáš Filip, Martin Pavlíček, Petr Sosík Published: 2026-03-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.18592v1 We introduce a two-dimensional (2D) early exit strategy that coordinates layer-wise and sentence-wise exiting for classification tasks in large language models. By processing input incrementally sentence-by-sentence while progressively activating deeper layers, our method achieves multiplicative computational savings that exceed those from optimizing either dimension independently. Experimental evaluation across four state-of-the-art LLMs (Llama 3.1, Llama 3.2, Gemma, Qwen; 3B-8B parameters) on three sentiment classification datasets demonstrates additional speed-ups of 1.4--2.3$\times$ over optimal layer-wise early exit for simpler tasks with vanilla models, with graceful degradation on complex multi-class problems. Fine-tuning reduces but does not eliminate this advantage. The approach is model-agnostic, requires only lightweight classification adapters, and is orthogonal to complementary efficiency methods such as quantization and pruning. Our findings indicate that 2D early exit strategies excel when semantic information accumulates predictably across input structure, suggesting possible applicability to sequence-processing tasks beyond sentiment classification. --- ### Disentangled Robot Learning via Separate Forward and Inverse Dynamics Pretraining Authors: Wenyao Zhang, Bozhou Zhang, Zekun Qi, Wenjun Zeng, Xin Jin, Li Zhang Published: 2026-03-27 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2604.16391v1 Vision-language-action (VLA) models have shown great potential in building generalist robots, but still face a dilemma-misalignment of 2D image forecasting and 3D action prediction. Besides, such a vision-action entangled training manner limits model learning from large-scale, action-free web video data. To address these issues, we propose DeFI, a novel framework that Decouples visual Forward and Inverse dynamics pretraining to exploit respective data sources, wherein video generation and action prediction are disentangled. We introduce the General Forward Dynamics Model (GFDM), pretrained on diverse human and robot videos for future prediction, and the General Inverse Dynamics Model (GIDM), trained via self-supervised learning to infer latent actions from unlabeled video transitions. These models are then integrated into a unified architecture for end-to-end finetuning on downstream tasks. In this manner, GFDM and GIDM first shine separately and then cooperate for mutual benefit. Extensive experiments on CALVIN ABC-D and SimplerEnv demonstrate state-of-the-art performance, with DeFI achieving an average task length of 4.51 for CALVIN, 51.2% success rate on SimplerEnv-Fractal benchmark and 81.3% success rate in real-world deployment, significantly outperforming prior methods. --- ### Visual-RRT: Finding Paths toward Visual-Goals via Differentiable Rendering Authors: Sebin Lee, Jumin Lee, Taeyeon Kim, Younju Na, Woobin Im, Sung-Eui Yoon Published: 2026-03-27 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2604.16388v1 Rapidly-exploring random trees (RRTs) have been widely adopted for robot motion planning due to their robustness and theoretical guarantees. However, existing RRT-based planners require explicit goal configurations specified as numerical joint angles, while many practical applications provide goal specifications through visual observations such as images or demonstration videos where precise goal configurations are unavailable. In this paper, we propose visual-RRT (vRRT), a motion planner that enables visual-goal planning by unifying gradient-based exploitation from differentiable robot rendering with sampling-based exploration from RRTs. We further introduce (i) a frontier-based exploration-exploitation strategy that adaptively prioritizes visually promising search regions, and (ii) inertial gradient tree expansion that inherits optimization states across tree branches for momentum-consistent gradient exploitation. Extensive experiments across various robot manipulators including Franka, UR5e, and Fetch demonstrate that vRRT achieves effective visual-goal planning in both simulated and real-world settings, bridging the gap between sampling-based planning and vision-centric robot applications. Our code is available at https://sgvr.kaist.ac.kr/Visual-RRT. --- ### Large language models for post-publication research evaluation: Evidence from expert recommendations and citation indicators Authors: Mengjia Wu, Yi Zhang, Robin Haunschild, Lutz Bornmann Published: 2026-03-27 Categories: cs.IR, cs.AI, cs.DL arXiv: https://arxiv.org/abs/2604.16387v1 Assessing the quality of scientific research is essential for scholarly communication, yet widely used approaches face limitations in scalability, subjectivity, and time delay. Recent advances in large language models (LLMs) offer new opportunities for automated research evaluation based on textual content. This study examines whether LLMs can support post-publication peer review tasks by benchmarking their outputs against expert judgments and citation-based indicators. Two evaluation tasks are constructed using articles from the H1 Connect platform: identifying high-quality articles and performing finer-grained evaluation including article rating, merit classification, and expert style commenting. Multiple model families, including BERT models, general-purpose LLMs, and reasoning oriented LLMs, are evaluated under multiple learning strategies. Results show that LLMs perform well in coarse grained evaluation tasks, achieving accuracy above 0.8 in identifying highly recommended articles. However, performance decreases substantially in fine-grained rating tasks. Few-shot prompting improves performance over zero-shot settings, while supervised fine-tuning produces the strongest and most balanced results. Retrieval augmented prompting improves classification accuracy in some cases but does not consistently strengthen alignment with citation indicators. The overall correlations between model outputs and citation indicators remain positive but moderate. --- ### DAOnt: A Formal Ontology for EU Data Act Compliance Authors: Sheyla Leyva-Sánchez, Fabian Linde, Meem Arafat Manab, María Poveda-Villalón, Víctor Rodríguez-Doncel Published: 2026-03-27 Categories: cs.DB, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2604.16386v1 The EU Data Act establishes comprehensive rules governing data access and sharing across business-to-consumer (B2C), business-to-business (B2B), and business-to-government (B2G) contexts. This paper presents a comprehensive ontology for the EU Data Act, enabling reasoning over data sharing agreements through machine-readable representations. The DAOnt ontology reuses elements from three established ontologies, LKIF-Core, ODRL, and DPV, to capture the normative structure of the Data Act. The ontology captures the main concepts and relationships in the Regulation, and it also operationalises three articles to facilitate compliance checking: Article 4(1) (B2C user access rights), Article 8(6) (B2B trade secret exceptions) and Article 19(2)(a) (B2G competitive use prohibitions). The ontology supports compliance checking through SPARQL queries that return obligations, permissions, and prohibitions, allowing organisations to verify whether data-sharing agreements meet the requirements of the EU Data Act and to assess conditions such as FRAND obligations. By representing key legal concepts in RDF, our work helps bridge the gap between the legal provisions of the Data Act and their computational interpretation. The complete ontology, along with example instances and queries, is available online. --- ### StressWeb: A Diagnostic Benchmark for Web Agent Robustness under Realistic Interaction Variability Authors: Haoyue Bai, Dong Wang, Long Chen, Bingguang Hao, Pengyang Shao, Yonghui Yang, Yicheng He, Chenyi Zhuang Published: 2026-03-27 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2604.16385v1 Large language model-based web agents have demonstrated strong performance on realistic web interaction tasks. However, existing evaluations are predominantly conducted under relatively stable and well-behaved interaction conditions, which may overestimate agent robustness. High task success in such idealized settings does not necessarily reflect performance under realistic web interaction. To address this limitation, we introduce a diagnostic stress-testing benchmark for web agents. We first construct realistic and controllable web environments that provide clean and stable interaction workflows as reference baselines. We then introduce structured and controlled perturbations that emulate interaction variability, including shifting layouts, altered interaction semantics, and execution disruptions. By comparing agent behavior between clean and perturbed settings, our framework enables systematic diagnosis of robustness under what-if interaction scenarios. Through extensive evaluation of state-of-the-art multimodal web agents, we show that stress-based evaluation exposes failure modes and substantial robustness gaps that remain hidden under clean benchmark conditions. --- ### TRACE: Thermal Recognition Attentive-Framework for CO2 Emissions from Livestock Authors: Taminul Islam, Abdellah Lakhssassi, Toqi Tahamid Sarker, Mohamed Embaby, Khaled R Ahmed, Amer AbuGhazaleh Published: 2026-03-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.09648v1 Quantifying exhaled CO2 from free-roaming cattle is both a direct indicator of rumen metabolic state and a prerequisite for farm-scale carbon accounting, yet no existing system can deliver continuous, spatially resolved measurements without physical confinement or contact. We present TRACE (Thermal Recognition Attentive-Framework for CO2 Emissions from Livestock), the first unified framework to jointly address per-frame CO2 plume segmentation and clip-level emission flux classification from mid-wave infrared (MWIR) thermal video. TRACE contributes three domain-specific advances: a Thermal Gas-Aware Attention (TGAA) encoder that incorporates per-pixel gas intensity as a spatial supervisory signal to direct self-attention toward high-emission regions at each encoder stage; an Attention-based Temporal Fusion (ATF) module that captures breath-cycle dynamics through structured cross-frame attention for sequence-level flux classification; and a four-stage progressive training curriculum that couples both objectives while preventing gradient interference. Benchmarked against fifteen state-of-the-art models on the CO2 Farm Thermal Gas Dataset, TRACE achieves an mIoU of 0.998 and the best result on every segmentation and classification metric simultaneously, outperforming domain-specific gas segmenters with several times more parameters and surpassing all baselines in flux classification. Ablation studies confirm that each component is individually essential: gas-conditioned attention alone determines precise plume boundary localization, and temporal reasoning is indispensable for flux-level discrimination. TRACE establishes a practical path toward non-invasive, continuous, per-animal CO2 monitoring from overhead thermal cameras at commercial scale. Codes are available at https://github.com/taminulislam/trace. --- ### Efficient Disruption of Criminal Networks through Multi-Objective Genetic Algorithms Authors: Yehezkiel Darmadi, Thanh Thi Nguyen, Campbell Wilson Published: 2026-03-27 Categories: cs.NE, cs.AI arXiv: https://arxiv.org/abs/2604.09647v1 Criminal networks, such as the Sicilian Mafia, pose substantial threats to public safety, national security, and economic stability. Outdated disruption methods with a focus on removing influential individuals or key players have proven ineffective due to the covertness of the network. Thus, researchers have been trying to apply Social Network Analysis (SNA) techniques, such as centrality-based measures, to identify key players. However, removing individuals with high centrality often proves to be inefficient, as it does not mimic the real-world scenarios that Law Enforcement Agencies (LEAs) face. For instance, the operational costs limit the LEAs from exploiting the results of the centrality-based methods. This study proposes a multi-objective optimisation framework like the Weighted Sum Genetic Algorithm (WS-GA) and the Non-dominated Sorting Genetic Algorithm II (NSGA-II) to identify disruption strategies that balance two conflicting goals, maximising fragmentation and minimising operational cost which is captured by the spatial distance between nodes and the nearest LEA headquarters. The study utilises the "Montagna Operation" dataset for the experiments. The results demonstrate that although centrality-based approaches can fragment network effectively, they tend to incur higher operational costs. In contrast, the proposed algorithms achieve comparable disruption outcomes with significantly lower operational costs. The contribution of this work lies in incorporating operational costs in a form of spatial distance constraints into disruption strategy, which has been largely overlooked in prior studies. This research offers a scalable multi-objective capability that improves practical application of SNA in guiding LEAs in disrupting criminal networks more efficiently and strategically. --- ### Distributionally Robust Token Optimization in RLHF Authors: Yeping Jin, Jiaming Hu, Ioannis Ch. Paschalidis Published: 2026-03-27 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.08577v2 Large Language Models (LLMs) tend to respond correctly to prompts that align well with the data they were trained and fine-tuned on. Yet, small shifts in wording, format, or language can trigger surprisingly large failures, especially on multi-step reasoning problems. To address this problem, we propose a Distributionally Robust Token Optimization (DRTO) approach, which combines token-level Reinforcement Learning from Human Feedback (RLHF) with Distributionally Robust Optimization (DRO). DRTO constructs f-divergence ambiguity sets over span-level actor losses, providing a principled way to emphasize difficult response segments during policy optimization. Empirically, DRTO enhances consistency under distribution shifts in multiple reasoning benchmarks among different tasks, achieving $+4.4$ percentage points on MATH-500 and $+2.7$ percentage points on LiveCodeBench over standard RTO. --- ### GAN-Enhanced Deep Reinforcement Learning for Semantic-Aware Resource Allocation in 6G Network Slicing Authors: Daniel Benniah John Published: 2026-03-27 Categories: cs.NI, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.08576v1 Sixth-generation (6G) wireless networks must support heterogeneous services: enhanced Mobile Broadband (eMBB) requiring 1 Tbps data rates, massive Machine-Type Communications (mMTC) supporting 10 million devices per km, and Ultra-Reliable Low-Latency Communications (URLLC) with 0.1-1 ms latency. Current resource allocation suffers from three limitations: (1) semantic blindness wasting 35% bandwidth on redundant data, (2) discrete action quantization, and (3) limited training diversity. This paper proposes GAN-DDPG, a Generative Adversarial Network-enhanced Deep Deterministic Policy Gradient framework integrating conditional GANs for traffic synthesis, continuous action DDPG, and semantic-aware reward optimization. Extensive simulations with statistical validation demonstrate significant improvements: 22% URLLC, 20% eMBB, 25% mMTC spectral efficiency gains (all p < 0.001) compared to baseline DDPG, with 18% latency and 31% packet loss reduction. --- ### MolPaQ: Modular Quantum-Classical Patch Learning for Interpretable Molecular Generation Authors: Syed Rameez Naqvi, Lu Peng Published: 2026-03-27 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.08575v1 Molecular generative models must jointly ensure validity, diversity, and property control, yet existing approaches typically trade off among these objectives. We present MOLPAQ, a modular quantum-classical generator that assembles molecules from quantum-generated latent patches. A \b{eta}-VAE pretrained on QM9 learns a chemically aligned latent manifold; a reduced conditioner maps molecular descriptors into this space; and a parameter-efficient quantum patch generator produces entangled node embeddings that a valence-aware aggregator reconstructs into valid molecular graphs. Adversarial fine-tuning with a latent critic and chemistry-shaped reward yields 100\% RDKit validity, 99.75\% novelty, and 0.905 diversity. Beyond aggregate metrics, the pretrained quantum generator, steered by the conditioner, improves mean QED by approx. 2.3\% and increases aromatic motif incidence by approx. 10-12\% relative to a parameter-matched classical generator, highlighting its role as a compact topology-shaping operator. --- ### Distilling Genomic Models for Efficient mRNA Representation Learning via Embedding Matching Authors: Rasched Haidari, Sam Martin, Maxime Allard Published: 2026-03-27 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.08574v1 Large Genomic Foundation Models have recently achieved remarkable results and in-vivo translation capabilities. However these models quickly grow to over a few Billion of parameters and are expensive to run when compute is limited. To overcome this challenge, we present a distillation framework for transferring mRNA representations from a state of the art genomic foundation model into a much smaller model specialized for mRNA sequences, reducing the size by 200-fold. Embedding-level distillation worked better than logit based methods, which we found unstable. Benchmarking on mRNA-bench demonstrates that the distilled model achieves state-of-the-art performance among models of comparable size and competes with larger architectures for mRNA-related tasks. Our results highlight embedding-based distillation of mRNA sequences as an effective training strategy for biological foundation models. This enables similar efficient and scalable sequence modelling in genomics, particularly when large models are computationally challenging or infeasible. --- ### Silhouette Loss: Differentiable Global Structure Learning for Deep Representations Authors: Matheus Vinícius Todescato, Joel Luís Carbonera Published: 2026-03-27 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2604.08573v1 Learning discriminative representations is a central goal of supervised deep learning. While cross-entropy (CE) remains the dominant objective for classification, it does not explicitly enforce desirable geometric properties in the embedding space, such as intra-class compactness and inter-class separation. Existing metric learning approaches, including supervised contrastive learning (SupCon) and proxy-based methods, address this limitation by operating on pairwise or proxy-based relationships, but often increase computational cost and complexity. In this work, we introduce Soft Silhouette Loss, a novel differentiable objective inspired by the classical silhouette coefficient from clustering analysis. Unlike pairwise objectives, our formulation evaluates each sample against all classes in the batch, providing a batch-level notion of global structure. The proposed loss directly encourages samples to be closer to their own class than to competing classes, while remaining lightweight. Soft Silhouette Loss can be seamlessly combined with cross-entropy, and is also complementary to supervised contrastive learning. We propose a hybrid objective that integrates them, jointly optimizing local pairwise consistency and global cluster structure. Extensive experiments on seven diverse datasets demonstrate that: (i) augmenting CE with Soft Silhouette Loss consistently improves over CE and other metric learning baselines; (ii) the hybrid formulation outperforms SupCon alone; and (iii) the combined method achieves the best performance, improving average top-1 accuracy from 36.71% (CE) and 37.85% (SupCon2) to 39.08%, while incurring substantially lower computational overhead. These results suggest that classical clustering principles can be reinterpreted as differentiable objectives for deep learning, enabling efficient optimization of both local and global structure in representation spaces. --- ### Ranked Activation Shift for Post-Hoc Out-of-Distribution Detection Authors: Gianluca Guglielmo, Marc Masana Published: 2026-03-27 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2604.08572v2 State-of-the-art post-hoc out-of-distribution detection methods rely on intermediate layer activation editing. However, they exhibit inconsistent performance across datasets and models. We show that this instability is driven by differences in the activation distributions, and identify a failure mode of scaling-based methods that arises when penultimate layer activations are not rectified. Motivated by this analysis, we propose RAS, a hyperparameter-free post-hoc method that replaces sorted activation magnitudes with a fixed in-distribution reference profile. Our simple plug-and-play method shows strong and consistent performance across datasets and architectures without assumptions on the penultimate layer activation function, and without requiring any hyperparameter tuning, while empirically preserving in-distribution classification accuracy. We further analyze what drives the improvement, showing that both inhibiting and exciting activation shifts independently contribute to better out-of-distribution discrimination. --- ### A Benchmark of Classical and Deep Learning Models for Agricultural Commodity Price Forecasting on A Novel Bangladeshi Market Price Dataset Authors: Tashreef Muhammad, Tahsin Ahmed, Meherun Farzana, Md. Mahmudul Hasan, Abrar Eyasir, Md. Emon Khan, Mahafuzul Islam Shawon, Ferdous Mondol, Mahmudul Hasan, Muhammad Ibrahim Published: 2026-03-27 Categories: cs.LG, econ.EM arXiv: https://arxiv.org/abs/2604.06227v1 Accurate short-term forecasting of agricultural commodity prices is critical for food security planning and smallholder income stabilisation in developing economies, yet machine-learning-ready datasets for this purpose remain scarce in South Asia. This paper makes two contributions. First, we introduce AgriPriceBD, a benchmark dataset of 1,779 daily retail mid-prices for five Bangladeshi commodities - garlic, chickpea, green chilli, cucumber, and sweet pumpkin - spanning July 2020 to June 2025, extracted from government reports via an LLM-assisted digitisation pipeline. Second, we evaluate seven forecasting approaches spanning classical models - naïve persistence, SARIMA, and Prophet - and deep learning architectures - BiLSTM, Transformer, Time2Vec-enhanced Transformer, and Informer - with Diebold-Mariano statistical significance tests. Commodity price forecastability is fundamentally heterogeneous: naïve persistence dominates on near-random-walk commodities. Time2Vec temporal encoding provides no statistically significant advantage over fixed sinusoidal encoding and causes catastrophic degradation on green chilli (+146.1% MAE, p<0.001). Prophet fails systematically, attributable to discrete step-function price dynamics incompatible with its smooth decomposition assumptions. Informer produces erratic predictions (variance up to 50x ground-truth), confirming sparse-attention Transformers require substantially larger training sets than small agricultural datasets provide. All code, models, and data are released publicly to support replication and future forecasting research on agricultural commodity markets in Bangladesh and similar developing economies. --- ### The Geometry of Forgetting Authors: Sambartha Ray Barman, Andrey Starenky, Sophia Bodnar, Nikhil Narasimhan, Ashwin Gopinath Published: 2026-03-27 Categories: q-bio.NC, cs.AI, cs.IR, cs.NE arXiv: https://arxiv.org/abs/2604.06222v1 Why do we forget? Why do we remember things that never happened? The conventional answer points to biological hardware. We propose a different one: geometry. Here we show that high-dimensional embedding spaces, subjected to noise, interference, and temporal degradation, reproduce quantitative signatures of human memory with no phenomenon-specific engineering. Power-law forgetting ($b = 0.460 \pm 0.183$, human $b \approx 0.5$) arises from interference among competing memories, not from decay. The identical decay function without competitors yields $b \approx 0.009$, fifty times smaller. Time alone does not produce forgetting in this system. Competition does. Production embedding models (nominally 384--1{,}024 dimensions) concentrate their variance in only ${\sim}16$ effective dimensions, placing them deep in the interference-vulnerable regime. False memories require no engineering at all: cosine similarity on unmodified pre-trained embeddings reproduces the Deese--Roediger--McDermott false alarm rate ($0.583$ versus human ${\sim}0.55$) with zero parameter tuning and no boundary conditions. We did not build a false memory system. We found one already present in the raw geometry of semantic space. These results suggest that core memory phenomena are not bugs of biological implementation but features of any system that organizes information by meaning and retrieves it by proximity. --- ### Scaling Teams or Scaling Time? Memory Enabled Lifelong Learning in LLM Multi-Agent Systems Authors: Shanglin Wu, Yuyang Luo, Yueqing Liang, Kaiwen Shi, Yanfang Ye, Ali Payani, Kai Shu Published: 2026-03-27 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2604.03295v1 Large language model (LLM) multi-agent systems can scale along two distinct dimensions: by increasing the number of agents and by improving through accumulated experience over time. Although prior work has studied these dimensions separately, their interaction under realistic cost constraints remains unclear. In this paper, we introduce a conceptual scaling view of multi-agent systems that jointly considers team size and lifelong learning ability, and we study how memory design shares this landscape. To this end, we propose \textbf{LLMA-Mem}, a lifelong memory framework for LLM multi-agent systems under flexible memory topologies. We evaluate LLMA-Mem on \textsc{MultiAgentBench} across coding, research, and database environments. Empirically, LLMA-Mem consistently improves long-horizon performance over baselines while reducing cost. Our analysis further reveals a non-monotonic scaling landscape: larger teams do not always produce better long-term performance, and smaller teams can outperform larger ones when memory better supports the reuse of experience. These findings position memory design as a practical path for scaling multi-agent systems more effectively and more efficiently over time. --- ### Expressibility of neural quantum states: a Walsh-complexity perspective Authors: Taige Wang Published: 2026-03-27 Categories: cond-mat.str-el, cond-mat.dis-nn, cs.LG, quant-ph arXiv: https://arxiv.org/abs/2604.03294v2 Neural quantum states are powerful variational wavefunctions, but it remains unclear which many-body states can be represented efficiently by modern additive architectures. We introduce Walsh complexity, a basis-dependent measure of how broadly a wavefunction is spread over parity patterns. States with an almost uniform Walsh spectrum require exponentially large Walsh complexity from any good approximant. We show that shallow additive feed-forward networks cannot generate such complexity in the tame regime, e.g. polynomial activations with subexponential parameter scaling. As a concrete example, we construct a simple dimerized state prepared by a single layer of disjoint controlled-$Z$ gates. Although it has only short-range entanglement and a simple tensor-network description, its Walsh complexity is maximal. Full-cube fits across system size and depth are consistent with the complexity bound: for polynomial activations, successful fitting appears only once depth reaches a logarithmic scale in $N$, whereas activation saturation in $\tanh$ produces a sharp threshold-like jump already at depth $3$. Walsh complexity therefore provides an expressibility axis complementary to entanglement and clarifies when depth becomes an essential resource for additive neural quantum states. --- ### Impact of geophysical fields on Deep Learning-based Lagrangian drift simulations Authors: Daria Botvynko, Carlos Granero-Belinchon, Simon Van Gennip, Abdesslam Benzinou, Ronan Fablet Published: 2026-03-27 Categories: physics.ao-ph, cs.AI arXiv: https://arxiv.org/abs/2604.03292v1 We assess the influence of different Eulerian geophysical input fields on Lagrangian drift simulations using DriftNet, a learning-based method designed to simulate Lagrangian drift on the sea surface. Two experiments are conducted: a fully numerical experiment (Benchmark B1) and a real-world drifters-based experiment (Benchmark B2). Both experiments are performed in two regions with different ocean dynamics: North East Pacific and Gulf Stream regions. The performance of DrifNet is evaluated with three different metrics: separation distance between simulated and ground-truth trajectories, the normalized cumulative Lagrangian separation and the autocorrelation of Lagrangian velocities. In both regions, results from B1 show that combining assimilated sea surface currents (SSC) with fully observed sea surface height (SSH) leads to greatest improvement in trajectory simulation. This configuration reduces separation distance by over 50\% and significantly decreases normalized cumulative Lagrangian separation and metrics related to velocities autocorrelation functions compared to the baseline using SSC alone. On the other hand, the inclusion of sea surface temperature (SST) either alone or in combination with SSC generally degrades performance. In B2, using satellite-derived SSH, Ekman and winds velocities improves surface drifters trajectories simulation, particularly in the North East Pacific. While the satellite-derived SST in combination with reanalysis-based SSC configuration leads to better trajectories simulation in the Gulf Stream. Overall, we highlight the added value of combining multiple geophysical fields to improve Lagrangian drift simulation on both numerical and real-world experiments. --- ### RAGnaroX: A Secure, Local-Hosted ChatOps Assistant Using Small Language Models Authors: Benedikt Dornauer, Mircea-Cristian Racasan Published: 2026-03-27 Categories: cs.AR, cs.AI arXiv: https://arxiv.org/abs/2604.03291v1 This paper introduces RAGnaroX, a resource-efficient ChatOps assistant that operates entirely on commodity hardware. Unlike existing solutions that often rely on external providers such as Azure or OpenAI, RAGnaroX offers a fully auditable, on-premise stack implemented in Rust. Its architecture integrates modular data ingestion, hybrid retrieval, and function calling, enabling flexible yet secure deployment. Our evaluation focuses on the RAG pipeline, with benchmarks conducted on the SQuAD (single-hop QA), MultiHopRAG (multi-hop QA), and MLQA (cross-lingual QA) datasets. Results show that RAGnaroX achieves competitive accuracy while maintaining strong resource efficiency, for example, reaching 0.90 context precision on single-hop questions with an average response time of 2.5 seconds per request. A replication package containing the tool, the demonstration video (https://www.youtube.com/watch? v=cDxfuEbcoM4), and all supporting materials are available at https://github.com/genius-itea/RAGnaroX.git. --- ### Xpertbench: Expert Level Tasks with Rubrics-Based Evaluation Authors: Xue Liu, Xin Ma, Yuxin Ma, Yongchang Peng, Duo Wang, Zhoufutu Wen, Ge Zhang, Kaiyuan Zhang, Xinyu Chen, Yida Ding, Tianci He, Jiani Hou et al. Published: 2026-03-27 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.02368v4 As Large Language Models (LLMs) exhibit plateauing performance on conventional benchmarks, a pivotal challenge persists: evaluating their proficiency in complex, open-ended tasks characterizing genuine expert-level cognition. Existing frameworks suffer from narrow domain coverage, reliance on generalist tasks, or self-evaluation biases. To bridge this gap, we present XpertBench, a high-fidelity benchmark engineered to assess LLMs across authentic professional domains. XpertBench consists of 1,346 meticulously curated tasks across 80 categories, spanning finance, healthcare, legal services, education, and dual-track research (STEM and Humanities). These tasks are derived from over 1,000 submissions by domain experts--including researchers from elite institutions and practitioners with extensive clinical or industrial experience--ensuring superior ecological validity. Each task uses detailed rubrics with mostly 15-40 weighted checkpoints to assess professional rigor. To facilitate scalable yet human-aligned assessment, we introduce ShotJudge, a novel evaluation paradigm that employs LLM judges calibrated with expert few-shot exemplars to mitigate self-rewarding biases. Our empirical evaluation of state-of-the-art LLMs reveals a pronounced performance ceiling: even leading models achieve a peak success rate of only ~66%, with a mean score around 55%. Models also exhibit domain-specific divergence, showing non-overlapping strengths in quantitative reasoning versus linguistic synthesis.. These findings underscore a significant "expert-gap" in current AI systems and establish XpertBench as a critical instrument for navigating the transition from general-purpose assistants to specialized professional collaborators. --- ### ML-Enabled Open RAN: A Comprehensive Survey of Architectures, Challenges, and Opportunities Authors: Mira Chandra Kirana, Patatchona Keyela, Fatemeh Rostamian, Deemah H. Tashman, Soumaya Cherkaoui Published: 2026-03-27 Categories: cs.NI, cs.AI arXiv: https://arxiv.org/abs/2604.01239v1 As wireless communication systems become more advanced, Open Radio Access Networks (O-RAN) stand out as a notable framework that promotes interoperability and cost-effectiveness. An examination of the progression of RAN architectures, as well as O-RAN's underlying principles, reveals the importance of machine learning (ML) in addressing various challenges, including spectrum management, resource allocation, and security. Hence, this survey provides a comprehensive overview of the integration of ML within O-RAN, highlighting its transformative potential in enhancing network performance and efficiency. This survey aims to describe the current status of ML applications in O-RAN while indicating possible directions for future research by analyzing existing literature. The findings aim to assist researchers and stakeholders in formulating optimal service strategies and advancing the understanding of intelligent wireless networks. --- ### Trustworthy AI-Driven Dynamic Hybrid RIS: Joint Optimization and Reward Poisoning-Resilient Control in Cognitive MISO Networks Authors: Deemah H. Tashman, Soumaya Cherkaoui Published: 2026-03-27 Categories: cs.NI, cs.AI arXiv: https://arxiv.org/abs/2604.01238v1 Cognitive radio networks (CRNs) are a key mechanism for alleviating spectrum scarcity by enabling secondary users (SUs) to opportunistically access licensed frequency bands without harmful interference to primary users (PUs). To address unreliable direct SU links and energy constraints common in next-generation wireless networks, this work introduces an adaptive, energy-aware hybrid reconfigurable intelligent surface (RIS) for underlay multiple-input single-output (MISO) CRNs. Distinct from prior approaches relying on static RIS architectures, our proposed RIS dynamically alternates between passive and active operation modes in real time according to harvested energy availability. We also model our scenario under practical hardware impairments and cascaded fading channels. We formulate and solve a joint transmit beamforming and RIS phase optimization problem via the soft actor-critic (SAC) deep reinforcement learning (DRL) method, leveraging its robustness in continuous and highly dynamic environments. Notably, we conduct the first systematic study of reward poisoning attacks on DRL agents in RIS-enhanced CRNs, and propose a lightweight, real-time defense based on reward clipping and statistical anomaly filtering. Numerical results demonstrate that the SAC-based approach consistently outperforms established DRL baselines, and that the dynamic hybrid RIS strikes a superior trade-off between throughput and energy consumption compared to fully passive and fully active alternatives. We further show the effectiveness of our defense in maintaining SU performance even under adversarial conditions. Our results advance the practical and secure deployment of RIS-assisted CRNs, and highlight crucial design insights for energy-constrained wireless systems. --- ### DarwinNet: An Evolutionary Network Architecture for Agent-Driven Protocol Synthesis Authors: Jinliang Xu, Bingqi Li Published: 2026-03-27 Categories: cs.NE, cs.AI, cs.DC, cs.MA, cs.NI arXiv: https://arxiv.org/abs/2604.01236v3 Traditional network architectures suffer from severe protocol ossification and structural fragility due to their reliance on static, human-defined rules that fail to adapt to the emergent edge cases and probabilistic reasoning of modern autonomous agents. To address these limitations, this paper proposes DarwinNet, a bio-inspired, self-evolving network architecture that transitions communication protocols from a \textit{design-time} static paradigm to a \textit{runtime} growth paradigm. DarwinNet utilizes a tri-layered framework-comprising an immutable physical anchor (L0), a WebAssembly-based fluid cortex (L1), and an LLM-driven Darwin cortex (L2)-to synthesize high-level business intents into executable bytecode through a dual-loop \textit{Intent-to-Bytecode} (I2B) mechanism. We introduce the Protocol Solidification Index (PSI) to quantify the evolutionary maturity of the system as it collapses from high-latency intelligent reasoning (Slow Thinking) toward near-native execution (Fast Thinking). Validated through a reliability growth framework based on the Crow-AMSAA model, experimental results demonstrate that DarwinNet achieves anti-fragility by treating environmental anomalies as catalysts for autonomous evolution. Our findings confirm that DarwinNet can effectively converge toward physical performance limits while ensuring endogenous security through zero-trust sandboxing, providing a viable path for the next generation of intelligent, self-optimizing networks. --- ### Generalizable Foundation Models for Calorimetry via Mixtures-of-Experts and Parameter Efficient Fine Tuning Authors: Carlos Cardona-Giraldo, Cristiano Fanelli, James Giroux, Cole Granger, Benjamin Nachman, Gerald Sabin Published: 2026-03-27 Categories: physics.ins-det, cs.LG, hep-ex, nucl-ex arXiv: https://arxiv.org/abs/2603.28804v1 Modern particle physics experiments face an increasing demand for high-fidelity detector simulation as luminosities rise and computational requirements approach the limits of available resources. Deep generative models have emerged as promising surrogates for traditional Monte Carlo simulation, with recent advances drawing inspiration from large language models (LLM) and next-token prediction paradigms. In this work, we introduce a generalizable foundation model for calorimetry built on next-token transformer backbones, designed to support modular adaptation across materials, particle species, and detector configurations. Our approach combines Mixture-of-Experts pre-training with parameter-efficient fine-tuning strategies to enable controlled, additive model expansion without catastrophic forgetting. A pre-trained backbone is trained to generate electromagnetic showers across multiple absorber materials, while new materials are incorporated through the addition and tuning of lightweight expert modules. Extensions to new particle types are achieved via parameter-efficient fine-tuning and modular vocabularies, preserving the integrity of the base model. This design enables efficient, incremental knowledge integration as new simulation datasets become available, a critical requirement in realistic detector-development workflows. In addition, we demonstrate that next-token calorimeter models are computationally competitive with standard generative approaches under established LLM optimization procedures. These results establish next-token architectures as a viable path toward extensible, physics-aware foundation models for calorimetry and future high-energy physics experiments. --- ### CREST: Constraint-Release Execution for Multi-Robot Warehouse Shelf Rearrangement Authors: Jiaqi Tan, Yudong Luo, Sophia Huang, Yifan Yang, Hang Ma Published: 2026-03-27 Categories: cs.RO, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2603.28803v1 Double-Deck Multi-Agent Pickup and Delivery (DD-MAPD) models the multi-robot shelf rearrangement problem in automated warehouses. MAPF-DECOMP is a recent framework that first computes collision-free shelf trajectories with a MAPF solver and then assigns agents to execute them. While efficient, it enforces strict trajectory dependencies, often leading to poor execution quality due to idle agents and unnecessary shelf switching. We introduce CREST, a new execution framework that achieves more continuous shelf carrying by proactively releasing trajectory constraints during execution. Experiments on diverse warehouse layouts show that CREST consistently outperforms MAPF-DECOMP, reducing metrics related to agent travel, makespan, and shelf switching by up to 40.5\%, 33.3\%, and 44.4\%, respectively, with even greater benefits under lift/place overhead. These results underscore the importance of execution-aware constraint release for scalable warehouse rearrangement. Code and data are available at https://github.com/ChristinaTan0704/CREST. --- ### Not Search, But Scan: Benchmarking MLLMs on Scan-Oriented Academic Paper Reasoning Authors: Rongjin Li, Zichen Tang, Xianghe Wang, Xinyi Hu, Zhengyu Wang, Zhengyu Lu, Yiling Huang, Jiayuan Chen, Weisheng Tan, Jiacheng Liu, Zhongjun Yang, Haihong E Published: 2026-03-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.28651v1 With the rapid progress of multimodal large language models (MLLMs), AI already performs well at literature retrieval and certain reasoning tasks, serving as a capable assistant to human researchers, yet it remains far from autonomous research. The fundamental reason is that current work on academic paper reasoning is largely confined to a search-oriented paradigm centered on pre-specified targets, with reasoning grounded in relevance retrieval, which struggles to support researcher-style full-document understanding, reasoning, and verification. To bridge this gap, we propose \textbf{ScholScan}, a new benchmark for academic paper reasoning. ScholScan introduces a scan-oriented task setting that asks models to read and cross-check entire papers like human researchers, scanning the document to identify consistency issues. The benchmark comprises 1,800 carefully annotated questions drawn from nine error categories across 13 natural-science domains and 715 papers, and provides detailed annotations for evidence localization and reasoning traces, together with a unified evaluation protocol. We assessed 15 models across 24 input configurations and conducted a fine-grained analysis of MLLM capabilities for all error categories. Across the board, retrieval-augmented generation (RAG) methods yield no significant improvements, revealing systematic deficiencies of current MLLMs on scan-oriented tasks and underscoring the challenge posed by ScholScan. We expect ScholScan to be the leading and representative work of the scan-oriented task paradigm. --- ### Multi-Level Barriers to Generative AI Adoption Across Disciplines and Professional Roles in Higher Education Authors: Jianhua Yang, Kerem Öge, Adrian von Mühlenen, Abdullah Bilal Akbulut, Tanya Suzanne Carey, Chidi Okorro Published: 2026-03-27 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2603.27052v3 Generative Artificial Intelligence (GenAI) is rapidly reshaping higher education, yet barriers to its adoption across different disciplines and institutional roles remain underexplored. Existing literature frequently attributes adoption barriers to individual-level factors such as perceived usefulness and ease of use. This study instead investigates whether such barriers are structurally produced. Drawing on a multi-method survey analysis of 272 academic and professional services (PSs) staff at a Russell Group university, we examine how disciplinary contexts and institutional roles shape perceived barriers. By integrating multinomial logistic regression (MLR), structural equation modelling (SEM), and semantic clustering of open-ended responses, we move beyond descriptive accounts to provide a multi-level explanation of GenAI adoption. Our findings reveal clear, systematic differences: non-STEM academics primarily report ethical and cultural barriers related to academic integrity, whereas STEM and PSs staff disproportionately emphasize institutional, governance, and infrastructure constraints. We conclude that GenAI adoption barriers are deeply embedded in organizational ecosystems and epistemic norms, suggesting that universities must move beyond generalized training to develop role-specific governance and support frameworks. --- ### Overcoming the Incentive Collapse Paradox Authors: Qichuan Yin, Ziwei Su, Shuangning Li Published: 2026-03-27 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2603.27049v2 AI-assisted task delegation is increasingly common, yet human effort in such systems is costly and typically unobserved. Recent work by Bastani and Cachon (2025); Sambasivan et al. (2021) shows that accuracy-based payment schemes suffer from incentive collapse: as AI accuracy improves, sustaining positive human effort requires unbounded payments. We study this phenomenon in a budget-constrained principal-agent framework with strategic human agents whose output accuracy depends on unobserved effort. Our first contribution is a general impossibility result showing that incentive collapse is not merely a limitation of simple linear payments, but arises for any payment rule based only on observed task accuracy.To overcome this barrier, we propose a sentinel-auditing payment mechanism that enforces a strictly positive and controllable level of human effort at finite cost, independent of AI accuracy. Building on this incentive-robust foundation, we develop an incentive-aware active statistical inference framework that jointly optimizes (i) the auditing rate and (ii) active sampling and budget allocation across tasks of varying difficulty to minimize the final statistical loss under a single budget. Experiments demonstrate improved cost-error tradeoffs relative to standard active learning and auditing-only baselines. --- ### MOOZY: A Patient-First Foundation Model for Computational Pathology Authors: Yousef Kotp, Vincent Quoc-Huy Trinh, Christopher Pal, Mahdi S. Hosseini Published: 2026-03-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.27048v3 Computational pathology needs whole-slide image (WSI) foundation models that transfer across diverse clinical tasks, yet current approaches remain largely slide-centric, often depend on private data and expensive paired-report supervision, and do not explicitly model relationships among multiple slides from the same patient. We present MOOZY, a patient-first pathology foundation model in which the patient case, not the individual slide, is the core unit of representation. MOOZY explicitly models dependencies across all slides from the same patient via a case transformer during pretraining, combining multi-stage self-supervision with scaled low-cost task supervision. In Stage 1, we pretrain a vision-only slide encoder on 77,134 public slide feature grids using masked self-distillation. In Stage 2, we align these representations with clinical semantics using a case transformer and multi-task supervision over 333 tasks from 56 public datasets, including 205 classification and 128 survival tasks across four endpoints. Across sixteen held-out tasks, MOOZY improves macro weighted F1, balanced accuracy, and macro weighted ROC-AUC relative to PRISM by +4.19\%, +7.93\%, and +6.95\%, respectively. MOOZY is also parameter efficient with 85.77M parameters, 14$\times$ smaller than GigaPath. These results suggest that patient-level pretraining yields transferable embeddings, providing a path toward scalable patient-first histopathology foundation models. --- ### Unsupervised Behavioral Compression: Learning Low-Dimensional Policy Manifolds through State-Occupancy Matching Authors: Andrea Fraschini, Davide Tenedini, Riccardo Zamboni, Mirco Mutti, Marcello Restelli Published: 2026-03-27 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.27044v3 Deep Reinforcement Learning (DRL) is widely recognized as sample-inefficient, a limitation attributable in part to the high dimensionality and substantial functional redundancy inherent to the policy parameter space. A recent framework, which we refer to as Action-based Policy Compression (APC), mitigates this issue by compressing the parameter space $Θ$ into a low-dimensional latent manifold $\mathcal Z$ using a learned generative mapping $g:\mathcal Z \to Θ$. However, its performance is severely constrained by relying on immediate action-matching as a reconstruction loss, a myopic proxy for behavioral similarity that suffers from compounding errors across sequential decisions. To overcome this bottleneck, we introduce Occupancy-based Policy Compression (OPC), which enhances APC by shifting behavior representation from immediate action-matching to long-horizon state-space coverage. Specifically, we propose two principal improvements: (1) we curate the dataset generation with an information-theoretic uniqueness metric that delivers a diverse population of policies; and (2) we propose a fully differentiable compression objective that directly minimizes the divergence between the true and reconstructed mixture occupancy distributions. These modifications force the generative model to organize the latent space around true functional similarity, promoting a latent representation that generalizes over a broad spectrum of behaviors while retaining most of the original parameter space's expressivity. Finally, we empirically validate the advantages of our contributions across multiple continuous control benchmarks. --- ### Introducing MELI: the Mandarin-English Language Interview Corpus Authors: Suyuan Liu, Molly Babel Published: 2026-03-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.27043v2 We introduce the Mandarin-English Language Interview (MELI) Corpus, an open-source resource of 29.8 hours of speech from 51 Mandarin-English bilingual speakers. MELI combines matched sessions in Mandarin and English with two speaking styles: read sentences and spontaneous interviews about language varieties, standardness, and learning experiences. Audio was recorded at 44.1 kHz (16-bit, stereo). Interviews were fully transcribed, force-aligned at word and phone levels, and anonymized. Descriptively, the Mandarin component totals ~14.7 hours (mean duration 17.3 minutes) and the English component ~15.1 hours (mean duration 17.8 minutes). We report token/type statistics for each language and document code-switching patterns (frequent in Mandarin sessions; more limited in English sessions). The corpus design supports within-/cross-speaker, within/cross-language acoustic comparison and links acoustics to speakers' stated language attitudes, enabling both quantitative and qualitative analyses. The MELI Corpus will be released with transcriptions, alignments, metadata, scans of labelled maps and documentation under a CC BY-NC 4.0 license. --- ### Unified Number-Free Text-to-Motion Generation Via Flow Matching Authors: Guanhe Huang, Oya Celiktutan Published: 2026-03-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.27040v1 Generative models excel at motion synthesis for a fixed number of agents but struggle to generalize with variable agents. Based on limited, domain-specific data, existing methods employ autoregressive models to generate motion recursively, which suffer from inefficiency and error accumulation. We propose Unified Motion Flow (UMF), which consists of Pyramid Motion Flow (P-Flow) and Semi-Noise Motion Flow (S-Flow). UMF decomposes the number-free motion generation into a single-pass motion prior generation stage and multi-pass reaction generation stages. Specifically, UMF utilizes a unified latent space to bridge the distribution gap between heterogeneous motion datasets, enabling effective unified training. For motion prior generation, P-Flow operates on hierarchical resolutions conditioned on different noise levels, thereby mitigating computational overheads. For reaction generation, S-Flow learns a joint probabilistic path that adaptively performs reaction transformation and context reconstruction, alleviating error accumulation. Extensive results and user studies demonstrate UMF' s effectiveness as a generalist model for multi-person motion generation from text. Project page: https://githubhgh.github.io/umf/. ---