# Research Papers — 2026-03 · bundle page 79 of 243 > Full-text markdown of papers #3901–3950 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/79 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/78.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/80.md) ### Improving Diffusion Generalization with Weak-to-Strong Segmented Guidance Authors: Liangyu Yuan, Yufei Huang, Mingkun Lei, Tong Zhao, Ruoyu Wang, Changxi Chi, Yiwei Wang, Chi Zhang Published: 2026-03-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.20584v1 Diffusion models generate synthetic images through an iterative refinement process. However, the misalignment between the simulation-free objective and the iterative process often causes accumulated gradient error along the sampling trajectory, which leads to unsatisfactory results and a failure to generalize. Guidance techniques like Classifier Free Guidance (CFG) and AutoGuidance (AG) alleviate this by extrapolating between the main and inferior signal for stronger generalization. Despite empirical success, the effective operational regimes of prevalent guidance methods are still under-explored, leading to ambiguity when selecting the appropriate guidance method given a precondition. In this work, we first conduct synthetic comparisons to isolate and demonstrate the effective regime of guidance methods represented by CFG and AG from the perspective of weak-to-strong principle. Based on this, we propose a hybrid instantiation called SGG under the principle, taking the benefits of both. Furthermore, we demonstrate that the W2S principle along with SGG can be migrated into the training objective, improving the generalization ability of unguided diffusion models. We validate our approach with comprehensive experiments. At inference time, evaluations on SD3 and SD3.5 confirm that SGG outperforms existing training-free guidance variants. Training-time experiments on transformer architectures demonstrate the effective migration and performance gains in both conditional and unconditional settings. Code is available at https://github.com/851695e35/SGG. --- ### GHOST: Ground-projected Hypotheses from Observed Structure-from-Motion Trajectories Authors: Tomasz Frelek, Rohan Patil, Akshar Tumu, Henrik I. Christensen Published: 2026-03-21 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2603.20583v2 We present a scalable self-supervised approach for segmenting feasible vehicle trajectories from monocular images for autonomous driving in complex urban environments. Leveraging large-scale dashcam videos, we treat recorded ego-vehicle motion as implicit supervision and recover camera trajectories via monocular structure-from-motion, projecting them onto the ground plane to generate spatial masks of traversed regions without manual annotation. These automatically generated labels are used to train a deep segmentation network that predicts motion-conditioned path proposals from a single RGB image at run time, without explicit modeling of road or lane markings. Trained on diverse, unconstrained internet data, the model implicitly captures scene layout, lane topology, and intersection structure, and generalizes across varying camera configurations. We evaluate our approach on NuScenes, demonstrating reliable trajectory prediction, and further show transfer to an electric scooter platform through light fine-tuning. Our results indicate that large-scale ego-motion distillation yields structured and generalizable path proposals beyond the demonstrated trajectory, enabling trajectory hypothesis estimation via image segmentation. --- ### Generative Diffusion Model for Risk-Neutral Derivative Pricing Authors: Nilay Tiwari Published: 2026-03-21 Categories: q-fin.MF, stat.ML arXiv: https://arxiv.org/abs/2603.20582v1 Denoising diffusion probabilistic models (DDPMs) have emerged as powerful generative models for complex distributions, yet their use in arbitrage-free derivative pricing remains largely unexplored. Financial asset prices are naturally modeled by stochastic differential equations (SDEs), whose forward and reverse density evolution closely parallels the forward noising and reverse denoising structure of diffusion models. In this paper, we develop a framework for using DDPMs to generate risk-neutral asset price dynamics for derivative valuation. Starting from log-return dynamics under the physical measure, we analyze the associated forward diffusion and derive the reverse-time SDE. We show that the change of measure from the physical to the risk-neutral measure induces an additive shift in the score function, which translates into a closed-form risk-neutral epsilon shift in the DDPM reverse dynamics. This correction enforces the risk-neutral drift while preserving the learned variance and higher-order structure, yielding an explicit bridge between diffusion-based generative modeling and classical risk-neutral SDE-based pricing. We show that the resulting discounted price paths satisfy the martingale condition under the risk-neutral measure. Empirically, the method reproduces the risk-neutral terminal distribution and accurately prices both European and path-dependent derivatives, including arithmetic Asian options, under a GBM benchmark. These results demonstrate that diffusion-based generative models provide a flexible and principled approach to simulation-based derivative pricing. --- ### JUBAKU: An Adversarial Benchmark for Exposing Culturally Grounded Stereotypes in Japanese LLMs Authors: Taihei Shiotani, Masahiro Kaneko, Ayana Niwa, Yuki Maruyama, Daisuke Oba, Masanari Ohi, Naoaki Okazaki Published: 2026-03-21 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.20581v2 Social biases reflected in language are inherently shaped by cultural norms, which vary significantly across regions and lead to diverse manifestations of stereotypes. Existing evaluations of social bias in large language models (LLMs) for non-English contexts, however, often rely on translations of English benchmarks. Such benchmarks fail to reflect local cultural norms, including those found in Japanese. For instance, Western benchmarks may overlook Japan-specific stereotypes related to hierarchical relationships, regional dialects, or traditional gender roles. To address this limitation, we introduce Japanese cUlture adversarial BiAs benchmarK Under handcrafted creation (JUBAKU), a benchmark tailored to Japanese cultural contexts. JUBAKU uses adversarial construction to expose latent biases across ten distinct cultural categories. Unlike existing benchmarks, JUBAKU features dialogue scenarios hand-crafted by native Japanese annotators, specifically designed to trigger and reveal latent social biases in Japanese LLMs. We evaluated nine Japanese LLMs on JUBAKU and three others adapted from English benchmarks. All models clearly exhibited biases on JUBAKU, performing below the random baseline of 50% with an average accuracy of 23% (ranging from 13% to 33%), despite higher accuracy on the other benchmarks. Human annotators achieved 91% accuracy in identifying unbiased responses, confirming JUBAKU's reliability and its adversarial nature to LLMs. --- ### Context Cartography: Toward Structured Governance of Contextual Space in Large Language Model Systems Authors: Zihua Wu, Georg Gartner Published: 2026-03-21 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.20578v1 The prevailing approach to improving large language model (LLM) reasoning has centered on expanding context windows, implicitly assuming that more tokens yield better performance. However, empirical evidence - including the "lost in the middle" effect and long-distance relational degradation - demonstrates that contextual space exhibits structural gradients, salience asymmetries, and entropy accumulation under transformer architectures. We introduce Context Cartography, a formal framework for the deliberate governance of contextual space. We define a tripartite zonal model partitioning the informational universe into black fog (unobserved), gray fog (stored memory), and the visible field (active reasoning surface), and formalize seven cartographic operators - reconnaissance, selection, simplification, aggregation, projection, displacement, and layering - as transformations governing information transitions between and within zones. The operators are derived from a systematic coverage analysis of all non-trivial zone transformations and are organized by transformation type (what the operator does) and zone scope (where it applies). We ground the framework in the salience geometry of transformer attention, characterizing cartographic operators as necessary compensations for linear prefix memory, append-only state, and entropy accumulation under expanding context. An analysis of four contemporary systems (Claude Code, Letta, MemOS, and OpenViking) provides interpretive evidence that these operators are converging independently across the industry. We derive testable predictions from the framework - including operator-specific ablation hypotheses - and propose a diagnostic benchmark for empirical validation. --- ### LJ-Bench: Ontology-Based Benchmark for U.S. Crime Authors: Hung Yun Tseng, Wuzhen Li, Blerina Gkotse, Grigorios Chrysos Published: 2026-03-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.20572v1 The potential of Large Language Models (LLMs) to provide harmful information remains a significant concern due to the vast breadth of illegal queries they may encounter. Unfortunately, existing benchmarks only focus on a handful types of illegal activities, and are not grounded in legal works. In this work, we introduce an ontology of crime-related concepts grounded in the legal frameworks of Model Panel Code, which serves as an influential reference for criminal law and has been adopted by many U.S. states, and instantiated using Californian Law. This structured knowledge forms the foundation for LJ-Bench, the first comprehensive benchmark designed to evaluate LLM robustness against a wide range of illegal activities. Spanning 76 distinct crime types organized taxonomically, LJ-Bench enables systematic assessment of diverse attacks, revealing valuable insights into LLM vulnerabilities across various crime categories: LLMs exhibit heightened susceptibility to attacks targeting societal harm rather than those directly impacting individuals. Our benchmark aims to facilitate the development of more robust and trustworthy LLMs. The LJ-Bench benchmark and LJ-Ontology, along with experiments implementation for reproducibility are publicly available at https://github.com/AndreaTseng/LJ-Bench. --- ### Can LLM Agents Simulate Dynamic Networks? A Case Study on Email Networks with Phishing Synthesis Authors: Siqi Miao, Ziyang Chen, Yuhong Luo, Hans Hao-Hsun Hsu, Mufei Li, Kaiqing Zhang, Pan Li Published: 2026-03-20 Categories: cs.SI, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2605.12507v1 While Large Language Model (LLM) multi-agent systems (MAS) offer a transformative approach to simulating human behavior in complex systems, it remains largely unexplored whether these simulations can replicate realistic structural and temporal dynamics from a dynamic network perspective. Our evaluation indicates that existing frameworks excel at generating plausible micro-level interactions but fail to capture the emergent, macroscopic topologies necessary for domains that rely on realistic network dynamics, such as modeling information propagation and cybersecurity threats. To bridge this gap, we introduce two easily integrable extensions to simulation frameworks to ensure they preserve macroscopic network fidelity: 1) augmenting LLM agents with data-driven event triggers to organically sustain long-horizon interactions, and 2) integrating Hawkes processes to accurately model temporal activation dynamics. Our approach allows LLM MAS to capture both plausible micro-level patterns and macroscopic topologies. We further demonstrate the utility of this framework in synthesizing realistic phishing campaigns within evolving communication networks. The study reveals how threats exploit structural vulnerabilities, highlighting the potential of our framework for developing next-generation defenses. Our code is available at https://github.com/Graph-COM/NSL. --- ### Same Voice, Different Lab: On the Homogenization of Frontier LLM Personalities Authors: Avinash Krishna, Kalyana Chadalavada, Unso Eun Seo Jo Published: 2026-03-20 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2605.02897v1 LLM assistant personalities play a critical role in user experience and perceived response quality. We present a large-scale experiment of frontier LLM personalities using external ELO-based traits scoring across 144 traits. We find that all models tested converge on a form of trait expression that is systematic, methodical, and analytical and suppress traits such as remorseful and sycophantic. Moreover, models tend to diverge more in their expression of ``middle-of-distribution traits`` such as poetic or playful, but even these so-called ``creative`` models tend to have more neutral identities. These similarities suggest an implicit emergence of a standard of optimal assistant behavior. In a landscape of varied training methods, character training, therefore, stands out for its uniformity, offering insight into a tacit consensus between model developers. --- ### Architecture of an AI-Based Automated Course of Action Generation System for Military Operations Authors: Ji-il Park, Inwook Shim, Chong Hui Kim Published: 2026-03-20 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2604.20862v1 The automation system for Course of Action (CoA) planning is an essential element in future warfare. As maneuver speeds increase, surveillance ranges extend, and weapon ranges grow, the operational area expands, making traditional manned-based CoA planning increasingly challenging. Consequently, the development of an AI-based automated CoA planning system is becoming increasingly necessary. Accordingly, several countries and defense organizations are actively developing AI-based CoA planning systems. However, due to security restrictions and limited public disclosure, the technical maturity of such systems remains difficult to assess. Furthermore, as these systems are military-related, their details are not publicly disclosed, making it difficult to accurately assess the current level of development. In response to this, this study aims to introduce relevant doctrines within the scope of publicly available information and present applicable AI technologies for each stage of the CoA planning process. Ultimately, it proposes an architecture for the development of an automated CoA planning system. --- ### Algorithm Selection with Zero Domain Knowledge via Text Embeddings Authors: Stefan Szeider Published: 2026-03-20 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.19753v2 We propose a feature-free approach to algorithm selection: instead of hand-crafted instance features, we use pretrained text embeddings. Our method, ZeroFolio, proceeds in three steps. First, it reads the raw instance file as plain text. Second, it embeds it with a pretrained embedding model. Third, it selects an algorithm via weighted k-nearest neighbors. The key to our approach is the fact that pretrained embeddings can distinguish problem instances without any domain knowledge or task-specific training. Hence, we can apply the same three-step pipeline (serialize, embed, select) across any problem domain with text-based instance formats. We evaluate our approach on 11 ASlib scenarios spanning 7 domains (SAT, MaxSAT, QBF, ASP, CSP, MIP, and graph problems). Our experiments show that this approach outperforms a random forest trained on hand-crafted features in 9 of 11 scenarios, robustly across serialization seeds (every seed, not just a favorable one) and often by a substantial margin; it still wins 8 of 11 against a per-scenario-tuned random forest. On the three scenarios with published AutoFolio results from the 2015 ASlib competition, ZeroFolio comes within a small margin of AutoFolio without any per-scenario configuration tuning. Our ablation study shows that inverse-distance weighting, line shuffling, and Manhattan distance are the key design choices, and we analyze the sensitivity of the selector to the serialization seed. On SAT12-ALL, where both selectors are competitive, combining embeddings with hand-crafted features via soft voting yields a further improvement. --- ### CSF: Black-box Fingerprinting via Compositional Semantics for Text-to-Image Models Authors: Junhoo Lee, Mijin Koo, Nojun Kwak Published: 2026-03-20 Categories: cs.CR, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2604.16363v1 Text-to-image models are commercially valuable assets often distributed under restrictive licenses, but such licenses are enforceable only when violations can be detected. Existing methods require pre-deployment watermarking or internal model access, which are unavailable in commercial API deployments. We present Compositional Semantic Fingerprinting (CSF), the first black-box method for attributing fine-tuned text-to-image models to protected lineages using only query access. CSF treats models as semantic category generators and probes them with compositional underspecified prompts that remain rare under fine-tuning. This gives IP owners an asymmetric advantage: new prompt compositions can be generated after deployment, while attackers must anticipate and suppress a much broader space of fingerprints. Across 6 model families (FLUX, Kandinsky, SD1.5/2.1/3.0/XL) and 13 fine-tuned variants, our Bayesian attribution framework enables controlled-risk lineage decisions, with all variants satisfying the dominance criterion. --- ### SetFlow: Generating Structured Sets of Representations for Multiple Instance Learning Authors: Nikola Jovišić, Milica Škipina, Vanja Švenda Published: 2026-03-20 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2604.16362v1 Data scarcity and weak supervision continue to limit the performance of machine learning models in many real-world applications, such as mammography, where Multiple Instance Learning (MIL) often offers the best formulation. While recent foundation models provide strong semantic representations out of the box, effective augmentation of such representations of MIL data remains limited, as existing methods operate at the instance level and fail to capture intra-bag dependencies. In this work, we introduce SetFlow, a generative architecture that models entire MIL bags (i.e., sets) directly in the representation space. Our approach leverages the flow matching paradigm combined with a Set Transformer-inspired design, enabling it to handle permutation-invariant inputs while capturing interactions between instances within each bag. The model is conditioned on both class labels and input scale, allowing it to generate coherent and semantically consistent sets of representations. We evaluate SetFlow on a large-scale mammography benchmark using a state-of-the-art MIL-PF classification pipeline. The generated samples are shown to closely match the original data distribution and even improve downstream performance when used for augmentation. Furthermore, training on synthetic data alone shows competitive results, demonstrating the effectiveness of representation-space generative modeling for data-scarce and privacy-sensitive tasks. --- ### An Edge-Cloud Collaborative Architecture for Proactive Elderly Care: Real-Time Risk Assessment and Three-Level Emergency Response Authors: Lijie Zhou, Luran Wang Published: 2026-03-20 Categories: eess.SP, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2604.14154v1 The rapid aging of global populations has created an urgent need for intelligent healthcare monitoring systems to ensure the safety of elderly individuals living independently. Existing cloud-centric platforms face critical limitations, including high latency unsuitable for emergency response, privacy risks from continuous transmission of sensitive data, and limited, single-channel alert mechanisms lacking scalability and context awareness. This paper proposes an edge-cloud collaborative architecture that addresses these challenges through real-time multi-modal sensor fusion, a four-dimensional risk assessment model, and a three-level emergency response system. The framework adopts a five-layer design - device, edge, service, data, and application layers - enabling real-time risk evaluation with end-to-end alert latency under three seconds. At the edge, a weighted multi-modal fusion algorithm integrates data from five sensor types with confidence propagation. A unified risk score is generated by combining fall probability, physiological indicators, behavioral patterns, and sensor anomaly metrics. Based on dynamic thresholds, a three-tier notification system coordinates responses among family members, community doctors, and nearby volunteers. Experiments on CASAS, MIMIC-III, and SisFall datasets show that the approach achieves 91% activity recognition accuracy and an 84% anomaly detection F1-score, outperforming single-sensor methods. Deployment on Raspberry Pi 4 gateways demonstrates sub-100 ms inference latency while preserving privacy by keeping raw data local. This architecture advances practical, privacy-preserving, and responsive elderly care systems. --- ### DeEscalWild: A Real-World Benchmark for Automated De-Escalation Training with SLMs Authors: Md Hasebul Hasan, Krity Haque Charu, Eshwara Prasad Sridhar, Shuchisnigdha Deb, Mohammad A. Islam Published: 2026-03-20 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.13075v2 Effective de-escalation is critical for law enforcement safety and community trust, yet traditional training methods lack scalability and realism. While Large Language Models (LLMs) enable dynamic, open-ended simulations, their substantial computational footprint renders them impractical for deployment on the lightweight, portable hardware required for immersive field training. Small Language Models (SLMs) offer a viable real-time alternative but suffer from a critical scarcity of high-quality, domain-specific training data. To bridge this gap, we present DeEscalWild, a novel benchmark dataset curated from a multi-stage pipeline of in-the-wild police-civilian interactions extracted from publicly available video repositories. Starting with 5,000 raw inputs, we employed a rigorous hybrid filtering process combining human-in-the-loop verification with LLM-as-a-Judge evaluation to distill 1,500 high-fidelity scenarios. The resulting corpus comprises 285,887 dialogue turns, totaling approximately 4.7 million tokens. Extensive experiments demonstrate that SLMs fine-tuned on this data significantly outperform their base counterparts across ROUGE-L, BLEU-4, METEOR, BERTScore, Realism Score, and human evaluation metrics. Notably, our fine-tuned Qwen 2.5 (3B-Instruct) surpasses the general-purpose Gemini 2.5 Flash model when evaluated under equivalent conditions, demonstrating that domain-optimized SLMs can achieve superior performance with a fraction of the computational cost. This work establishes the foundational infrastructure for accessible, low-latency, and privacy-preserving officer training systems at the edge. We publicly release our code(https://github.com/Hasebul/DeEscalWild-Benchmark-Framework) and dataset(https://doi.org/10.7910/DVN/CWMCZI). --- ### PersonaVLM: Long-Term Personalized Multimodal LLMs Authors: Chang Nie, Chaoyou Fu, Yifan Zhang, Haihua Yang, Caifeng Shan Published: 2026-03-20 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2604.13074v1 Multimodal Large Language Models (MLLMs) serve as daily assistants for millions. However, their ability to generate responses aligned with individual preferences remains limited. Prior approaches enable only static, single-turn personalization through input augmentation or output alignment, and thus fail to capture users' evolving preferences and personality over time (see Fig.1). In this paper, we introduce PersonaVLM, an innovative personalized multimodal agent framework designed for long-term personalization. It transforms a general-purpose MLLM into a personalized assistant by integrating three key capabilities: (a) Remembering: It proactively extracts and summarizes chronological multimodal memories from interactions, consolidating them into a personalized database. (b) Reasoning: It conducts multi-turn reasoning by retrieving and integrating relevant memories from the database. (c) Response Alignment: It infers the user's evolving personality throughout long-term interactions to ensure outputs remain aligned with their unique characteristics. For evaluation, we establish Persona-MME, a comprehensive benchmark comprising over 2,000 curated interaction cases, designed to assess long-term MLLM personalization across seven key aspects and 14 fine-grained tasks. Extensive experiments validate our method's effectiveness, improving the baseline by 22.4% (Persona-MME) and 9.8% (PERSONAMEM) under a 128k context, while outperforming GPT-4o by 5.2% and 2.0%, respectively. Project page: https://PersonaVLM.github.io. --- ### OmniTrace: A Unified Framework for Generation-Time Attribution in Omni-Modal LLMs Authors: Qianqi Yan, Yichen Guo, Ching-Chen Kuo, Shan Jiang, Hang Yin, Yang Zhao, Xin Eric Wang Published: 2026-03-20 Categories: cs.CL, cs.AI, cs.MM arXiv: https://arxiv.org/abs/2604.13073v1 Modern multimodal large language models (MLLMs) generate fluent responses from interleaved text, image, audio, and video inputs. However, identifying which input sources support each generated statement remains an open challenge. Existing attribution methods are primarily designed for classification settings, fixed prediction targets, or single-modality architectures, and do not naturally extend to autoregressive, decoder-only models performing open-ended multimodal generation. We introduce OmniTrace, a lightweight and model-agnostic framework that formalizes attribution as a generation-time tracing problem over the causal decoding process. OmniTrace provides a unified protocol that converts arbitrary token-level signals such as attention weights or gradient-based scores into coherent span-level, cross-modal explanations during decoding. It traces each generated token to multimodal inputs, aggregates signals into semantically meaningful spans, and selects concise supporting sources through confidence-weighted and temporally coherent aggregation, without retraining or supervision. Evaluations on Qwen2.5-Omni and MiniCPM-o-4.5 across visual, audio, and video tasks demonstrate that generation-aware span-level attribution produces more stable and interpretable explanations than naive self-attribution and embedding-based baselines, while remaining robust across multiple underlying attribution signals. Our results suggest that treating attribution as a structured generation-time tracing problem provides a scalable foundation for transparency in omni-modal language models. --- ### LiveClawBench: Benchmarking LLM Agents on Complex, Real-World Assistant Tasks Authors: Xiang Long, Li Du, Yilong Xu, RongJian Xu, Qiyanhui Lu, Ying Gao, Qinhua Xie, Fangcheng Liu, Ning Ding, Haoqing Wang, Ziheng Li, Changjiang Zhou et al. Published: 2026-03-20 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.13072v2 OpenClaw-style personal assistants extend LLM agents from isolated tool use to open-ended, stateful, and personalized software environments. Evaluating these assistants is fundamentally a fidelity problem: benchmarks must be faithful both to the distribution of real assistant tasks and to the execution semantics of the environments in which those tasks unfold. Existing benchmarks often lose fidelity in one dimension or the other. Their task distributions are shaped by what is easy to isolate, mock, and verify, underrepresenting real-world difficulties such as cross-service dependency, contaminated state, implicit intent, and runtime change. Their environments are either live but hard to reproduce, or reproducible but reduced to endpoint-level stubs that remove sessions, artifacts, state transitions, and downstream side effects. We introduce LiveClawBench, a benchmark designed around this dual-fidelity requirement. LiveClawBench combines a Triple-Axis Complexity Framework for difficulty-driven task construction with reproducible full-stack mock applications that preserve stateful execution semantics. With 134 executable cases across 10 domains with 22 mocked services, LiveClawBench supports controlled, extensible, and factor-level diagnostic evaluation of realistic agentic tasks. We release the benchmark resources: (1) Benchmark: https://github.com/Mosi-AI/LiveClawBench; (2) Leaderboard: https://mosi-ai.github.io/LiveClawBench/leaderboard/; (3) Trajectories: https://huggingface.co/datasets/Mosi-AI/LiveClawbench-trajectories --- ### EVE: A Domain-Specific LLM Framework for Earth Intelligence Authors: Àlex R. Atrio, Antonio Lopez, Jino Rohit, Yassine El Ouahidi, Marcello Politi, Vijayasri Iyer, Umar Jamil, Sébastien Bratières, Nicolas Longépé Published: 2026-03-20 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.13071v2 We introduce Earth Virtual Expert (EVE), the first open-source, end-to-end initiative for developing and deploying domain-specialized LLMs for Earth Intelligence. At its core is EVE-Instruct, a domain-adapted 24B model built on Mistral Small 3.2 and optimized for reasoning and question answering. On newly constructed Earth Observation and Earth Sciences benchmarks, it outperforms comparable models while preserving general capabilities. We release curated training corpora and the first systematic domain-specific evaluation benchmarks, covering MCQA, open-ended QA, and factuality. EVE further integrates RAG and a hallucination-detection pipeline into a production system deployed via API and GUI, supporting 350 pilot users so far. All models, datasets, and code are ready to be released under open licenses as contributions to our field at huggingface.co/eve-esa and github.com/eve-esa. --- ### Curation of a Palaeohispanic Dataset for Machine Learning Authors: Gonzalo Martínez-Fernández, Jose F Quesada, Agustín Riscos-Núñez, Francisco José Salguero-Lamillar Published: 2026-03-20 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.13070v1 Palaeohispanic languages are those spoken in the Iberian Peninsula before the arrival of the Romans in the 3rd Century B.C. Their study was really put on motion after Gómez Moreno deciphered the Iberian Levantine script, one of the several semi-sillabaries used by these languages. Still, the Palaeohispanic languages have varying degrees of decipherment, and none is fully known to this day. Most of the studies have been performed from a purely linguistic point of view, and a computational approach may benefit this research area greatly. However, the resources are limited and presented in an unsuitable format for techniques such as Machine Learning. Therefore, a structured dataset is constructed, which will hopefully allow more progress in the field. --- ### Detection Without Correction: A Robust Asymmetry in Activation-Based Hallucination Probing Authors: Dip Roy, Rajiv Misra, Sanjay Kumar Singh, Anisha Roy Published: 2026-03-20 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.13068v2 Activation-based linear probing is widely proposed as a method for both detecting and correcting hallucinations in autoregressive language models. We present an empirical study across seven models spanning 117M to 7B parameters and three architecture families (GPT-2, Pythia, Qwen-2.5) that documents a robust asymmetry: linear probes can detect hallucination signals with above-chance accuracy in larger models, but activation steering along the probe-derived direction fails to correct hallucinations in 7 of 7 models tested. We further find that output-confidence baselines outperform activation probes on raw detection AUC at every model above 410M parameters, with the gap reaching 0.157 AUC for Pythia-6.9B. The probe's distinguishing value is therefore not detection accuracy but temporal positioning: probe signals are accessible at position zero (before any output tokens are produced), enabling pre-generation flagging that output-based methods structurally cannot provide. The temporal signal is statistically significant in two of seven models (Pythia-1.4B, p = 0.012; Qwen2.5-7B, p = 0.038) and absent in models below 400M parameters and in the base-only Pythia-6.9B. We position these findings as a clean negative result for the dominant probing-as-detection-and-control research direction and as initial evidence that probe-based methods occupy a complementary deployment niche, namely pre-generation flagging, rather than competing with output-based detectors on raw accuracy. --- ### Leveraging Machine Learning Techniques to Investigate Media and Information Literacy Competence in Tackling Disinformation Authors: José Manuel Alcalde-Llergo, Mariana Buenestado Fernández, Carlos Enrique George-Reyes, Andrea Zingoni, Enrique Yeguas-Bolívar Published: 2026-03-20 Categories: cs.CY, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.09635v1 This study develops machine learning models to assess Media and Information Literacy (MIL) skills specifically in the context of disinformation among students, particularly future educators and communicators. While the digital revolution has expanded access to information, it has also amplified the spread of false and misleading content, making MIL essential for fostering critical thinking and responsible media engagement. Despite its relevance, predictive modeling of MIL in relation to disinformation remains underexplored. To address this gap, a quantitative study was conducted with 723 students in education and communication programs using a validated survey. Classification and regression algorithms were applied to predict MIL competencies and identify key influencing factors. Results show that complex models outperform simpler approaches, with variables such as academic year and prior training significantly improving prediction accuracy. These findings can inform the design of targeted educational interventions and personalized strategies to enhance students' ability to critically navigate and respond to disinformation in digital environments. --- ### Can We Still Trace L1 Signals? Investigating the Resilience of Native Language Signals in the LLM Era Authors: Nabelanita Utami, Ryohei Sasano Published: 2026-03-20 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.08568v4 The widespread use of LLM-based writing assistance has raised an interesting question about the homogenization of English. As LLMs tend to revise texts toward mainstream English conventions reflected in their training data, the subtle fingerprints that reflect an author's native language (L1) may be gradually disappearing. This study investigates this phenomenon by analyzing native language identification (NLI) performance on academic abstracts. To this end, we construct two NLI datasets of academic abstracts extracted from arXiv and the ACL Anthology that covers eight native language groups across three time periods: pre-neural network (NN), pre-LLM, and post-LLM. We then evaluate NLI performance for each era using NLI classifiers obtained by fine-tuning LLMs. The results reveal a consistent decline in NLI performance over time. Notably, however, the decline is more pronounced from the pre-NN era to the pre-LLM era than from the pre-LLM era to the post-LLM era. This suggests that although academic English appears to have become increasingly homogenized in the LLM era, this homogenization did not suddenly emerge with the advent of LLMs; rather, it has progressed gradually since the emergence of neural approaches to language processing. Furthermore, a rewriting experiment using recent LLMs shows a larger NLI performance drop than the progression across eras alone, suggesting that increased LLM use may lead to further homogenization in the future. --- ### Can We Still Hear the Accent? Investigating the Resilience of Native Language Signals in the LLM Era Authors: Nabelanita Utami, Ryohei Sasano Published: 2026-03-20 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.08568v2 The evolution of writing assistance tools from machine translation to large language models (LLMs) has changed how researchers write. This study investigates whether this shift is homogenizing research papers by analyzing native language identification (NLI) trends in ACL Anthology papers across three eras: pre-neural network (NN), pre-LLM, and post-LLM. We construct a labeled dataset using a semi-automated framework and fine-tune a classifier to detect linguistic fingerprints of author backgrounds. Our analysis shows a consistent decline in NLI performance over time. Interestingly, the post-LLM era reveals anomalies: while Chinese and French show unexpected resistance or divergent trends, Japanese and Korean exhibit sharper-than-expected declines. --- ### Fighting AI with AI: AI-Agent Augmented DNS Blocking of LLM Services during Student Evaluations Authors: Yonas Kassa, James Bonacci, Ping Wang Published: 2026-03-20 Categories: cs.NI, cs.AI, cs.CY, cs.ET, cs.LG arXiv: https://arxiv.org/abs/2604.02360v1 The transformative potential of large language models (LLMs) in education, such as improving accessibility and personalized learning, is being eclipsed by significant challenges. These challenges stem from concerns that LLMs undermine academic assessment by enabling bypassing of critical thinking, leading to increased cognitive offloading. This emerging trend stresses the dual imperative of harnessing AI's educational benefits while safeguarding critical thinking and academic rigor in the evolving AI ecosystem. To this end, we introduce AI-Sinkhole, an AI-agent augmented DNS-based framework that dynamically discovers, semantically classifies, and temporarily network-wide blocks emerging LLM chatbot services during proctored exams. AI-Sinkhole offers explainable classification via quantized LLMs (LLama 3, DeepSeek-R1, Qwen-3) and dynamic DNS blocking with Pi-Hole. We also share our observations in using LLMs as explainable classifiers which achieved robust cross-lingual performance (F1-score > 0.83). To support future research and development in this domain initial codes with a readily deployable 'AI-Sinkhole' blockist is available on https://github.com/AIMLEdu/ai-sinkhole. --- ### Using LLM-as-a-Judge/Jury to Advance Scalable, Clinically-Validated Safety Evaluations of Model Responses to Users Demonstrating Psychosis Authors: May Lynn Reese, Markela Zeneli, Mindy Ng, Jacob Haimes, Andreea Damien, Elizabeth Stade Published: 2026-03-20 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.02359v1 General-purpose Large Language Models (LLMs) are becoming widely adopted by people for mental health support. Yet emerging evidence suggests there are significant risks associated with high-frequency use, particularly for individuals suffering from psychosis, as LLMs may reinforce delusions and hallucinations. Existing evaluations of LLMs in mental health contexts are limited by a lack of clinical validation and scalability of assessment. To address these issues, this research focuses on psychosis as a critical condition for LLM safety evaluation by (1) developing and validating seven clinician-informed safety criteria, (2) constructing a human-consensus dataset, and (3) testing automated assessment using an LLM as an evaluator (LLM-as-a-Judge) or taking the majority vote of several LLM judges (LLM-as-a-Jury). Results indicate that LLM-as-a-Judge aligns closely with the human consensus (Cohen's $κ_{\text{human} \times \text{gemini}} = 0.75$, $κ_{\text{human} \times \text{qwen}} = 0.68$, $κ_{\text{human} \times \text{kimi}} = 0.56$) and that the best judge slightly outperforms LLM-as-a-Jury (Cohen's $κ_{\text{human} \times \text{jury}} = 0.74$). Overall, these findings have promising implications for clinically grounded, scalable methods in LLM safety evaluations for mental health contexts. --- ### Logic-Gated Time-Shared Feedforward Networks for Alternating Finite Automata: Exact Simulation and Learnability Authors: Sahil Rajesh Dhayalkar Published: 2026-03-20 Categories: cs.FL, cs.AI arXiv: https://arxiv.org/abs/2604.01228v1 We present a formal and constructive framework for simulating Alternating Finite Automata (AFAs) using Logic-Gated Time-Shared Feedforward Networks (LG-TS-FFNs). Unlike prior neural automata models limited to Nondeterministic Finite Automata (NFAs) and existential reachability, our architecture integrates learnable, state-dependent biases that function as differentiable logic gates, enabling the representation of both Existential \textsc{\textsc{OR}} and Universal \textsc{\textsc{AND}} aggregation within a shared-parameter linear recurrence. We prove that this architectural modification upgrades the network's computational class to be structurally isomorphic to AFAs, thereby inheriting their exponential succinctness: the network can represent regular languages requiring $2^n$ states in an NFA with only $n$ neurons. We rigorously establish that the forward pass of an LG-TS-FFN exactly simulates the reachability dynamics of an AFA, including instantaneous $\varepsilon$-closures. Furthermore, we demonstrate empirical learnability: a continuous relaxation of the logic gates allows the network to simultaneously recover the automaton's topology and logical semantics from binary labels via standard gradient descent. Extensive experiments confirm that our model achieves perfect recovery of ground-truth automata, bridging the gap between statistical learning and succinct, universal logical reasoning. --- ### Decomposable Reward Modeling and Realistic Environment Design for Reinforcement Learning-Based Forex Trading Authors: Nabeel Ahmad Saidd Published: 2026-03-20 Categories: q-fin.GN, cs.LG arXiv: https://arxiv.org/abs/2604.00031v1 Applying reinforcement learning (RL) to foreign exchange (Forex) trading remains challenging because realistic environments, well-defined reward functions, and expressive action spaces must be satisfied simultaneously, yet many prior studies rely on simplified simulators, single scalar rewards, and restricted action representations, limiting both interpretability and practical relevance. This paper presents a modular RL framework designed to address these limitations through three tightly integrated components: a friction-aware execution engine that enforces strict anti-lookahead semantics, with observations at time t, execution at time t+1, and mark-to-market at time t+1, while incorporating realistic costs such as spread, commission, slippage, rollover financing, and margin-triggered liquidation; a decomposable 11-component reward architecture with fixed weights and per-step diagnostic logging to enable systematic ablation and component-level attribution; and a 10-action discrete interface with legal-action masking that encodes explicit trading primitives while enforcing margin-aware feasibility constraints. Empirical evaluation on EURUSD focuses on learning dynamics rather than generalization and reveals strongly non-monotonic reward interactions, where additional penalties do not reliably improve outcomes; the full reward configuration achieves the highest training Sharpe (0.765) and cumulative return (57.09 percent). The expanded action space increases return but also turnover and reduces Sharpe relative to a conservative 3-action baseline, indicating a return-activity trade-off under a fixed training budget, while scaling-enabled variants consistently reduce drawdown, with the combined configuration achieving the strongest endpoint performance. --- ### Smartphone-Based Identification of Unknown Liquids via Active Vibration Sensing Authors: Yongzhi Huang Published: 2026-03-20 Categories: eess.SP, cs.AI, cs.CE, cs.DC, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2603.28787v1 Traditional liquid identification instruments are often unavailable to the general public. This paper shows the feasibility of identifying unknown liquids with commercial lightweight devices, such as a smartphone. The key insight is that different liquid molecules have different viscosity coefficients and therefore must overcome different energy barriers during relative motion. With this intuition in mind, we introduce a novel model that measures liquids' viscosity based on active vibration. However, building a robust system using built-in smartphone accelerometers is challenging. Practical issues include under-sampling, self-interference, and the impact of liquid-volume changes. Instead of machine learning, we tackle these issues through multiple signal processing stages to reconstruct the original signals and cancel out the interference. Our approach estimates liquid viscosity with a mean relative error of 2.9% and distinguishes 30 types of liquids with an average accuracy of 95.47%. --- ### AI in Work-Based Learning: Understanding the Purposes and Effects of Intelligent Tools Among Student Interns Authors: John Paul P. Miranda, Rhiziel P. Manalese, Sheila M. Geronimo, Vernon Grace M. Maniago, Charlie K. Padilla, Aileen P. De Leon, Santa L. Merle, Mark Anthony A. Castro Published: 2026-03-20 Categories: cs.CY, cs.AI, stat.AP arXiv: https://arxiv.org/abs/2603.28786v1 This study examined how student interns in Philippine higher education use intelligent tools during their OJT. Data were collected from 384 respondents using a structured questionnaire that asked about AI tool usage, task-specific applications, and perceptions of confidence, ethics, and support. Analysis of task-based usage identified four main purposes: productivity and report writing, communication and content drafting, technical assistance and code support, and independent task completion. ChatGPT was the most commonly used AI tool, followed by Quillbot, Canva AI, and Grammarly. Students reported moderate confidence in using AI and applied these tools selectively and ethically during OJT tasks. This indicate that AI tools assist student interns in various OJT activities related to work-readiness. The study suggests that higher education programs include AI literacy and onboarding. Clear policies and fair access to AI tools are important to support responsible use and prepare students for future careers. --- ### Mixture of Experts with Soft Nearest Neighbor Loss: Resolving Expert Collapse via Representation Disentanglement Authors: Abien Fred Agarap, Arnulfo P. Azcarraga Published: 2026-03-20 Categories: cs.NE, cs.LG arXiv: https://arxiv.org/abs/2603.26734v1 The Mixture-of-Experts (MoE) model uses a set of expert networks that specialize on subsets of a dataset under the supervision of a gating network. A common issue in MoE architectures is ``expert collapse'' where overlapping class boundaries in the raw input feature space cause multiple experts to learn redundant representations, thus forcing the gating network into rigid routing to compensate. We propose an enhanced MoE architecture that utilizes a feature extractor network optimized using Soft Nearest Neighbor Loss (SNNL) prior to feeding input features to the gating and expert networks. By pre-conditioning the latent space to minimize distances among class-similar data points, we resolve structural expert collapse which results to experts learning highly orthogonal weights. We employ Expert Specialization Entropy and Pairwise Embedding Similarity to quantify this dynamic. We evaluate our experimental approach across four benchmark image classification datasets (MNIST, FashionMNIST, CIFAR10, and CIFAR100), and we show our SNNL-augmented MoE models demonstrate structurally diverse experts which allow the gating network to adopt a more flexible routing strategy. This paradigm significantly improves classification accuracy on the FashionMNIST, CIFAR10, and CIFAR100 datasets. --- ### Contextual inference from single objects in Vision-Language models Authors: Martina G. Vilas, Timothy Schaumlöffel, Gemma Roig Published: 2026-03-20 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.26731v1 How much scene context a single object carries is a well-studied question in human scene perception, yet how this capacity is organized in vision-language models (VLMs) remains poorly understood, with direct implications for the robustness of these models. We investigate this question through a systematic behavioral and mechanistic analysis of contextual inference from single objects. Presenting VLMs with single objects on masked backgrounds, we probe their ability to infer both fine-grained scene category and coarse superordinate context (indoor vs. outdoor). We found that single objects support above-chance inference at both levels, with performance modulated by the same object properties that predict human scene categorization. Object identity, scene, and superordinate predictions are partially dissociable: accurate inference at one level neither requires nor guarantees accurate inference at the others, and the degree of coupling differs markedly across models. Mechanistically, object representations that remain stable when background context is removed are more predictive of successful contextual inference. Scene and superordinate schemas are grounded in fundamentally different ways: scene identity is encoded in image tokens throughout the network, while superordinate information emerges only late or not at all. Together, these results reveal that the organization of contextual inference in VLMs is more complex than accuracy alone suggests, with behavioral and mechanistic signatures --- ### Multi-view Graph Convolutional Network with Fully Leveraging Consistency via Granular-ball-based Topology Construction, Feature Enhancement and Interactive Fusion Authors: Chengjie Cui, Taihua Xu, Shuyin Xia, Qinghua Zhang, Yun Cui, Shiping Wang Published: 2026-03-20 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.26729v2 The effective utilization of consistency is crucial for multi-view learning. GCNs leverage node connections to propagate information across the graph, facilitating the exploitation of consistency in multi-view data. However, most existing GCN-based multi-view methods suffer from several limitations. First, current approaches predominantly rely on KNN for topology construction, where the artificial selection of the k value significantly constrains the effective exploitation of inter-node consistency. Second, the inter-feature consistency within individual views is often overlooked, which adversely affects the quality of the final embedding representations. Moreover, these methods fail to fully utilize inter-view consistency as the fusion of embedded representations from multiple views is often implemented after the intra-view graph convolutional operation. Collectively, these issues limit the model's capacity to fully capture inter-node, inter-feature and inter-view consistency. To address these issues, this paper proposes the multi-view graph convolutional network with fully leveraging consistency via GB-based topology construction, feature enhancement and interactive fusion (MGCN-FLC). MGCN-FLC can fully utilize three types of consistency via the following three modules to enhance learning ability:The topology construction module based on the granular ball algorithm, which clusters nodes into granular balls with high internal similarity to capture inter-node consistency;The feature enhancement module that improves feature representations by capturing inter-feature consistency;The interactive fusion module that enables each view to deeply interact with all other views, thereby obtaining more comprehensive inter-view consistency. Experimental results on nine datasets show that the proposed MGCN-FLC outperforms state-of-the-art semi-supervised node classification methods. --- ### SEAR: Schema-Based Evaluation and Routing for LLM Gateways Authors: Zecheng Zhang, Han Zheng, Yue Xu Published: 2026-03-20 Categories: cs.DB, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2603.26728v1 Evaluating production LLM responses and routing requests across providers in LLM gateways requires fine-grained quality signals and operationally grounded decisions. To address this gap, we present SEAR, a schema-based evaluation and routing system for multi-model, multi-provider LLM gateways. SEAR defines an extensible relational schema covering both LLM evaluation signals (context, intent, response characteristics, issue attribution, and quality scores) and gateway operational metrics (latency, cost, throughput), with cross-table consistency links across around one hundred typed, SQL-queryable columns. To populate the evaluation signals reliably, SEAR proposes self-contained signal instructions, in-schema reasoning, and multi-stage generation that produces database-ready structured outputs. Because signals are derived through LLM reasoning rather than shallow classifiers, SEAR captures complex request semantics, enables human-interpretable routing explanations, and unifies evaluation and routing in a single query layer. Across thousands of production sessions, SEAR achieves strong signal accuracy on human-labeled data and supports practical routing decisions, including large cost reductions with comparable quality. --- ### The Nonverbal Gap: Toward Affective Computer Vision for Safer and More Equitable Online Dating Authors: Ratna Kandala, Niva Manchanda, Akshata Kishore Moharir Published: 2026-03-20 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.26727v1 Online dating has become the dominant way romantic relationships begin, yet current platforms strip the nonverbal cues: gaze, facial expression, body posture, response timing, that humans rely on to signal comfort, disinterest, and consent, creating a communication gap with disproportionate safety consequences for women. We argue that this gap represents both a technical opportunity and a moral responsibility for the computer vision community, which has developed the affective tools, facial action unit detection, gaze estimation, engagement modeling, and multimodal affect recognition, needed to begin addressing it, yet has largely ignored the dating domain as a research context. We propose a fairness-first research agenda organized around four capability areas: real-time discomfort detection, engagement asymmetry modeling between partners, consent-aware interaction design, and longitudinal interaction summarization, each grounded in established CV methodology and motivated by the social psychology of romantic communication. We argue that responsible pursuit of this agenda requires purpose-built datasets collected under dyadic consent protocols, fairness evaluation disaggregated across race, gender identity, neurotype, and cultural background, and architectural commitments to on-device processing that prevent affective data from becoming platform surveillance infrastructure. This vision paper calls on the WICV community, whose members are uniquely positioned to understand both the technical opportunity and the human stakes, to establish online dating safety as a first-class research domain before commercial deployment outpaces ethical deliberation. --- ### A Multimodal Deep Learning Framework for Edema Classification Using HCT and Clinical Data Authors: Aram Ansary Ogholbake, Hannah Choi, Spencer Brandenburg, Alyssa Antuna, Zahraa Al-Sharshahi, Makayla Cox, Haseeb Ahmed, Jacqueline Frank, Nathan Millson, Luke Bauerle, Jessica Lee, David Dornbos et al. Published: 2026-03-20 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.26726v1 We propose AttentionMixer, a unified deep learning framework for multimodal detection of brain edema that combines structural head CT (HCT) with routine clinical metadata. While HCT provides rich spatial information, clinical variables such as age, laboratory values, and scan timing capture complementary context that might be ignored or naively concatenated. AttentionMixer is designed to fuse these heterogeneous sources in a principled and efficient manner. HCT volumes are first encoded using a self-supervised Vision Transformer Autoencoder (ViT-AE++), without requiring large labeled datasets. Clinical metadata are mapped into the same feature space and used as keys and values in a cross-attention module, where HCT-derived feature vector serves as queries. This cross-attention fusion allows the network to dynamically modulate imaging features based on patient-specific context and provides an interpretable mechanism for multimodal integration. A lightweight MLP-Mixer then refines the fused representation before final classification, enabling global dependency modeling with substantially reduced parameter overhead. Missing or incomplete metadata are handled via a learnable embedding, promoting robustness to real-world clinical data quality. We evaluate AttentionMixer on a curated brain HCT cohort with expert edema annotations using five-fold cross-validation. Compared with strong HCT-only, metadata-only, and prior multimodal baselines, AttentionMixer achieves superior performance (accuracy 87.32%, precision 92.10%, F1-score 85.37%, AUC 94.14%). Ablation studies confirm the benefit of both cross-attention and MLP-Mixer refinement, and permutation-based metadata importance analysis highlights clinically meaningful variables driving predictions. These results demonstrate that structured, interpretable multimodal fusion can substantially improve edema detection in clinical practice. --- ### Capability Safety as Datalog: A Foundational Equivalence Authors: Cosimo Spera Published: 2026-03-20 Categories: cs.LO, cs.AI arXiv: https://arxiv.org/abs/2603.26725v1 We prove that capability safety admits an exact representation as propositional Datalog evaluation (Datalogprop: the monadic, ground, function-free fragment of first-order logic), enabling the transfer of algorithmic and structural results unavailable in the native formulation. This addresses two structural limitations of the capability hypergraph framework of Spera (2026): the absence of efficient incremental maintenance, and the absence of a decision procedure for audit surface containment. The equivalence is tight: capability hypergraphs correspond to exactly this fragment, no more. --- ### An Annotation-to-Detection Framework for Autonomous and Robust Vine Trunk Localization in the Field by Mobile Agricultural Robots Authors: Dimitrios Chatziparaschis, Elia Scudiero, Brent Sams, Konstantinos Karydis Published: 2026-03-20 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2603.26724v1 The dynamic and heterogeneous nature of agricultural fields presents significant challenges for object detection and localization, particularly for autonomous mobile robots that are tasked with surveying previously unseen unstructured environments. Concurrently, there is a growing need for real-time detection systems that do not depend on large-scale manually labeled real-world datasets. In this work, we introduce a comprehensive annotation-to-detection framework designed to train a robust multi-modal detector using limited and partially labeled training data. The proposed methodology incorporates cross-modal annotation transfer and an early-stage sensor fusion pipeline, which, in conjunction with a multi-stage detection architecture, effectively trains and enhances the system's multi-modal detection capabilities. The effectiveness of the framework was demonstrated through vine trunk detection in novel vineyard settings that featured diverse lighting conditions and varying crop densities to validate performance. When integrated with a customized multi-modal LiDAR and Odometry Mapping (LOAM) algorithm and a tree association module, the system demonstrated high-performance trunk localization, successfully identifying over 70% of trees in a single traversal with a mean distance error of less than 0.37m. The results reveal that by leveraging multi-modal, incremental-stage annotation and training, the proposed framework achieves robust detection performance regardless of limited starting annotations, showcasing its potential for real-world and near-ground agricultural applications. --- ### Sommelier: Scalable Open Multi-turn Audio Pre-processing for Full-duplex Speech Language Models Authors: Kyudan Jung, Jihwan Kim, Soyoon Kim, Jeonghoon Kim, Jaegul Choo, Cheonbok Park Published: 2026-03-20 Categories: cs.SD, cs.AI, eess.AS arXiv: https://arxiv.org/abs/2603.25750v2 As the paradigm of AI shifts from text-based LLMs to Speech Language Models (SLMs), there is a growing demand for full-duplex systems capable of real-time, natural human-computer interaction. However, the development of such models is constrained by the scarcity of high-quality, multi-speaker conversational data, as existing large-scale resources are predominantly single-speaker or limited in volume. Addressing the complex dynamics of natural dialogue, such as overlapping and back-channeling remains a challenge, with standard processing pipelines suffering from diarization errors and ASR hallucinations. To bridge this gap, we present a robust and scalable open-source data processing pipeline designed for full-duplex model. --- ### PDGMM-VAE: A Variational Autoencoder with Adaptive Per-Dimension Gaussian Mixture Model Priors for Nonlinear ICA Authors: Yuan-Hao Wei, Yan-Jie Sun Published: 2026-03-20 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2603.23547v2 Independent component analysis is a core framework within blind source separation for recovering latent source signals from observed mixtures under statistical independence assumptions. In this work, we propose PDGMM-VAE, a source-oriented variational autoencoder in which each latent dimension, interpreted explicitly as an individual source component, is assigned its own adaptive Gaussian mixture model prior. The proposed framework imposes heterogeneous per-dimension prior constraints, enabling different latent dimensions to model different non-Gaussian source marginals within a unified probabilistic encoder-decoder architecture. The parameters of these source-specific GMM priors are not fixed in advance, but are jointly learned together with the encoder and decoder under the overall training objective. Beyond the model construction itself, we provide a theoretical analysis clarifying why adaptive per-dimension prior design is meaningful in this setting. In particular, we show that heterogeneous per-dimension priors reduce latent permutation symmetry relative to homogeneous shared priors, and we further show that the KL regularization induced by the adaptive GMM prior creates source-specific attraction behavior that helps explain source-wise specialization during training. We also clarify the relation of the proposed model to the standard VAE and provide a weak recovery statement in an idealized linear low-noise regime. Experimental results on both linear and nonlinear mixing problems show that PDGMM-VAE can recover latent source signals and fit source-specific non-Gaussian marginals effectively. These results suggest that adaptive per-dimension mixture-prior design provides a principled and promising direction for VAE-based ICA and source-oriented generative modeling. --- ### Graph Signal Processing Meets Mamba2: Adaptive Filter Bank via Delta Modulation Authors: Yehjin Shin, Seojin Kim, Noseong Park Published: 2026-03-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.22333v1 State-space models (SSMs) offer efficient alternatives to attention with linear-time recurrence. Mamba2, a recent SSM-based language model, uses selective input gating and a multi-head structure, enabling parallel computation and strong benchmark performance. However, its multi-head recurrence operates independently without structured utilization or analysis. In this work, we propose a novel method called Hierarchical ADaptive filter bank for Efficient SSMs (HADES), a Graph Signal Processing (GSP)-inspired framework that reinterprets Mamba2 as an adaptive filter bank on a line graph. Our hierarchical architecture introduces two filter types: shared filters for global low-pass behavior and expert filters for local high-pass behavior, achieved through structured bias on the parameter Δ. HADES achieves comparable performance to baseline models including Mamba2 across various benchmarks in language modeling, commonsense reasoning, and long-context retrieval, while using only 58.9% of the original parameters. In this regard, HADES bridges GSP and neural sequence modeling, enabling efficient, hierarchical, and interpretable filtering within state-space models. --- ### Large Language Models for Missing Data Imputation: Understanding Behavior, Hallucination Effects, and Control Mechanisms Authors: Arthur Dantas Mangussi, Ricardo Cardoso Pereira, Ana Carolina Lorena, Pedro Henriques Abreu Published: 2026-03-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.22332v1 Data imputation is a cornerstone technique for handling missing values in real-world datasets, which are often plagued by missingness. Despite recent progress, prior studies on Large Language Models-based imputation remain limited by scalability challenges, restricted cross-model comparisons, and evaluations conducted on small or domain-specific datasets. Furthermore, heterogeneous experimental protocols and inconsistent treatment of missingness mechanisms (MCAR, MAR, and MNAR) hinder systematic benchmarking across methods. This work investigates the robustness of Large Language Models for missing data imputation in tabular datasets using a zero-shot prompt engineering approach. To this end, we present a comprehensive benchmarking study comparing five widely used LLMs against six state-of-the-art imputation baselines. The experimental design evaluates these methods across 29 datasets (including nine synthetic datasets) under MCAR, MAR, and MNAR mechanisms, with missing rates of up to 20\%. The results demonstrate that leading LLMs, particularly Gemini 3.0 Flash and Claude 4.5 Sonnet, consistently achieve superior performance on real-world open-source datasets compared to traditional methods. However, this advantage appears to be closely tied to the models' prior exposure to domain-specific patterns learned during pre-training on internet-scale corpora. In contrast, on synthetic datasets, traditional methods such as MICE outperform LLMs, suggesting that LLM effectiveness is driven by semantic context rather than purely statistical reconstruction. Furthermore, we identify a clear trade-off: while LLMs excel in imputation quality, they incur significantly higher computational time and monetary costs. Overall, this study provides a large-scale comparative analysis, positioning LLMs as promising semantics-driven imputers for complex tabular data. --- ### Conformal Risk Control for Safety-Critical Wildfire Evacuation Mapping: A Comparative Study of Tabular, Spatial, and Graph-Based Models Authors: Baljinnyam Dayan Published: 2026-03-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.22331v1 Every wildfire prediction model deployed today shares a dangerous property: none of these methods provides formal guarantees on how much fire spread is missed. Despite extensive work on wildfire spread prediction using deep learning, no prior study has applied distribution-free safety guarantees to this domain, leaving evacuation planners reliant on probability thresholds with no formal assurance. We address this gap by presenting, to our knowledge, the first application of conformal risk control (CRC) to wildfire spread prediction, providing finite-sample guarantees on false negative rate (FNR <= 0.05). We expose a stark failure: across three model families of increasing complexity (tabular: LightGBM, AUROC 0.854; convolutional: Tiny U-Net, AUROC 0.969; and graph-based: Hybrid ResGNN-UNet, AUROC 0.964), standard thresholds capture only 7-72% of true fire spread. CRC eliminates this failure uniformly. Our central finding is that model architecture determines evacuation efficiency, while CRC determines safety: both spatial models with CRC achieve approximately 95% fire coverage while flagging only approximately 15% of total pixels, making them 4.2x more efficient than LightGBM, while the graph model's additional complexity over a simple U-Net yields no meaningful efficiency gain. We propose a shift-aware three-way CRC framework that assigns SAFE/MONITOR/EVACUATE zones for operational triage, and characterize a fundamental limitation of prevalence-weighted bounds under extreme class imbalance (approximately 5% fire prevalence). All models, calibration code, and evaluation pipelines are released for reproducibility. --- ### Fair splits flip the leaderboard: CHANRG reveals limited generalization in RNA secondary-structure prediction Authors: Zhiyuan Chen, Zhenfeng Deng, Pan Deng, Yue Liao, Xiu Su, Peng Ye, Xihui Liu Published: 2026-03-20 Categories: q-bio.BM, cs.LG arXiv: https://arxiv.org/abs/2603.22330v1 Accurate prediction of RNA secondary structure underpins transcriptome annotation, mechanistic analysis of non-coding RNAs, and RNA therapeutic design. Recent gains from deep learning and RNA foundation models are difficult to interpret because current benchmarks may overestimate generalization across RNA families. We present the Comprehensive Hierarchical Annotation of Non-coding RNA Groups (CHANRG), a benchmark of 170{,}083 structurally non-redundant RNAs curated from more than 10 million sequences in Rfam~15.0 using structure-aware deduplication, genome-aware split design and multiscale structural evaluation. Across 29 predictors, foundation-model methods achieved the highest held-out accuracy but lost most of that advantage out of distribution, whereas structured decoders and direct neural predictors remained markedly more robust. This gap persisted after controlling for sequence length and reflected both loss of structural coverage and incorrect higher-order wiring. Together, CHANRG and a padding-free, symmetry-aware evaluation stack provide a stricter and batch-invariant framework for developing RNA structure predictors with demonstrable out-of-distribution robustness. --- ### Trained Persistent Memory for Frozen Decoder-Only LLMs Authors: Hong Jeong Published: 2026-03-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.22329v1 Decoder-only language models are stateless: hidden representations are discarded after every forward pass and nothing persists across sessions. Jeong (2026a) showed that trained memory adapters give a frozen encoder-decoder backbone persistent latent-space memory, building on the lateral-memory framework of Jeong (2026b,c). Here we ask whether the same principle transfers to the decoder-only setting, where no cross-attention pathway exists and memory must enter through self-attention alone. We adapt six methods -- prefix, parallel cross-attention, KV extension, Hebbian memory, context-gated branch, and slot-based sparse write -- to a frozen GPT-2, training only a small adapter $θ_{mem}$. The write rule is shared; only the read injection changes from decoder cross-attention to self-attention KV prefix or parallel branch. On LoCoMo we find a striking inductive-bias dichotomy: at $1\times$ capacity, three methods with strong architectural priors -- cross-attention (M.2), Hebbian (M.4), and slot write (M.6) -- achieve retained-memory scores of $7-18\%$ and knowledge gains $ΔK$ of $7-10$, while the other three fail ($< 0.4\%$). At $10\times$ capacity all six converge, showing the gap is architectural, not fundamental. Together with the encoder-decoder results of Jeong (2026a) and the brain-inspired modules of Jeong (2026b,c), these findings establish persistent latent-space memory as a general paradigm spanning major transformer families. --- ### Beyond the Mean: Distribution-Aware Loss Functions for Bimodal Regression Authors: Abolfazl Mohammadi-Seif, Carlos Soares, Rita P. Ribeiro, Ricardo Baeza-Yates Published: 2026-03-20 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2603.22328v1 Despite the strong predictive performance achieved by machine learning models across many application domains, assessing their trustworthiness through reliable estimates of predictive confidence remains a critical challenge. This issue arises in scenarios where the likelihood of error inferred from learned representations follows a bimodal distribution, resulting from the coexistence of confident and ambiguous predictions. Standard regression approaches often struggle to adequately express this predictive uncertainty, as they implicitly assume unimodal Gaussian noise, leading to mean-collapse behavior in such settings. Although Mixture Density Networks (MDNs) can represent different distributions, they suffer from severe optimization instability. We propose a family of distribution-aware loss functions integrating normalized RMSE with Wasserstein and Cramér distances. When applied to standard deep regression models, our approach recovers bimodal distributions without the volatility of mixture models. Validated across four experimental stages, our results show that the proposed Wasserstein loss establishes a new Pareto efficiency frontier: matching the stability of standard regression losses like MSE in unimodal tasks while reducing Jensen-Shannon Divergence by 45% on complex bimodal datasets. Our framework strictly dominates MDNs in both fidelity and robustness, offering a reliable tool for aleatoric uncertainty estimation in trustworthy AI systems. --- ### Evaluating AI-based Scientific Knowledge Synthesis with Epidemiological Systematic Reviews Authors: Shreyansh Padarha, Ryan Othniel Kearns, Tristan Naidoo, Lingyi Yang, Łukasz Borchmann, Piotr BŁaszczyk, Christian Morgenstern, Ruth McCabe, Sangeeta Bhatia, Philip H. Torr, Jakob Foerster, Scott A. Hale et al. Published: 2026-03-20 Categories: cs.IR, cs.AI, cs.DL arXiv: https://arxiv.org/abs/2603.22327v2 Systematic literature reviews (SLRs) are a demanding and high-stakes form of scientific knowledge synthesis that remains underspecified as an evaluation setting for large language models (LLMs). We introduce AgentSLR, a large-scale evaluation harness comprising an SLR automation workflow and an expert annotated dataset covering 16,248 articles, designed to test LLM capabilities across the stages of SLRs in epidemiology. Reference annotations were derived from peer-reviewed studies on WHO priority pathogens and produced by domain experts. The harness evaluates each review stage as a separate unit with dedicated metrics enabling targeted failure analysis. We evaluated five frontier reasoning models and found that no single model dominated across all tasks, showing sub-task specialisation often hidden by aggregate benchmarks. Structured data extraction is a major bottleneck, with no model exceeding an average field-level F1 of 0.67. Estimated costs vary substantially, by up to 96 times across evaluated models. Documented failure modes suggest that the evaluated models are not yet reliable enough for unsupervised deployment in epidemiology, where findings can inform public policy. --- ### A Direct Classification Approach for Reliable Wind Ramp Event Forecasting under Severe Class Imbalance Authors: Alejandro Morales-Hernández, Fabrizio De Caroa, Gian Marco Paldino, Pascal Tribel, Alfredo Vaccaro, Gianluca Bontempi Published: 2026-03-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.22326v1 Decision support systems are essential for maintaining grid stability in low-carbon power systems, such as wind power plants, by providing real-time alerts to control room operators regarding potential events, including Wind Power Ramp Events (WPREs). These early warnings enable the timely initiation of more detailed system stability assessments and preventive actions. However, forecasting these events is challenging due to the inherent class imbalance in WPRE datasets, where ramp events are less frequent (typically less than 15\% of observed events) compared to normal conditions. Ignoring this characteristic undermines the performance of conventional machine learning models, which often favor the majority class. This paper introduces a novel methodology for WPRE forecasting as a multivariate time series classification task and proposes a data preprocessing strategy that extracts features from recent power observations and masks unavailable ramp information, making it integrable with traditional real-time ramp identification tools. Particularly, the proposed methodology combines majority-class undersampling and ensemble learning to enhance wind ramp event forecasting under class imbalance. Numerical simulations conducted on a real-world dataset demonstrate the superiority of our approach, achieving over 85% accuracy and 88% weighted F1 score, outperforming benchmark classifiers. --- ### Hybrid Associative Memories Authors: Leon Lufkin, Tomás Figliolia, Beren Millidge, Kamesh Krishnamurthy Published: 2026-03-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.22325v2 Recurrent neural networks (RNNs) and self-attention are both widely used sequence-mixing layers that maintain an internal memory. However, this memory is constructed using two orthogonal mechanisms: RNNs compress the entire past into a fixed-size state, whereas self-attention's state stores every past time step growing its state (the KV cache) linearly with the sequence length. This results in orthogonal strengths and weaknesses. Self-attention layers excel at retrieving information in the context but have large memory and computational costs, while RNNs are more efficient but degrade over longer contexts and underperform for precise recall tasks. Prior work combining these mechanisms has focused primarily on naively interleaving them to reduce computational cost without regard to their complementary mechanisms. We propose the Hybrid Associative Memory (HAM) layer, which combines self-attention and RNNs while leveraging their individual strengths: the RNN compresses the entire sequence, while attention supplements it *only* with information that is difficult for the RNN to predict, which is hence the most valuable information to explicitly store. HAM layers enable data-dependent growth of the KV cache, which can be precisely controlled by the user with a single, continuous threshold. We find that this fine-grained control of the KV cache growth rate has a smooth trade-off with loss and performance. Empirically, we show that our hybrid architecture offers strong, competitive performance relative to RNNs and Transformers even at substantially lower KV-cache usage. --- ### DAQ: Delta-Aware Quantization for Post-Training LLM Weight Compression Authors: Xiaoming Yu, Shize Tang, Guanghua Yu, Linchuan Xie, Song Liu, Jianchen Zhu, Feng Li Published: 2026-03-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.22324v1 We introduce Delta-Aware Quantization (DAQ), a data-free post-training quantization framework that preserves the knowledge acquired during post-training. Standard quantization objectives minimize reconstruction error but are agnostic to the base model, allowing quantization noise to disproportionately corrupt the small-magnitude parameter deltas ($ΔW$) that encode post-training behavior -- an effect we analyze through the lens of quantization as implicit regularization. DAQ replaces reconstruction-based objectives with two delta-aware metrics -- Sign Preservation Rate and Cosine Similarity -- that directly optimize for directional fidelity of $ΔW$, requiring only the base and post-trained weight matrices. In a pilot FP8 study, DAQ recovers style-specific capabilities lost under standard quantization while maintaining general performance. --- ### A Multi-Task Targeted Learning Framework for Lithium-Ion Battery State-of-Health and Remaining Useful Life Authors: Chenhan Wang, Zhengyi Bao, Huipin Lin, Jiahao Nie, Chunxiang Zhu Published: 2026-03-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.22323v1 Accurately predicting the state-of-health (SOH) and remaining useful life (RUL) of lithium-ion batteries is crucial for ensuring the safe and efficient operation of electric vehicles while minimizing associated risks. However, current deep learning methods are limited in their ability to selectively extract features and model time dependencies for these two parameters. Moreover, most existing methods rely on traditional recurrent neural networks, which have inherent shortcomings in long-term time-series modeling. To address these issues, this paper proposes a multi-task targeted learning framework for SOH and RUL prediction, which integrates multiple neural networks, including a multi-scale feature extraction module, an improved extended LSTM, and a dual-stream attention module. First, a feature extraction module with multi-scale CNNs is designed to capture detailed local battery decline patterns. Secondly, an improved extended LSTM network is employed to enhance the model's ability to retain long-term temporal information, thus improving temporal relationship modeling. Building on this, the dual-stream attention module-comprising polarized attention and sparse attention to selectively focus on key information relevant to SOH and RUL, respectively, by assigning higher weights to important features. Finally, a many-to-two mapping is achieved through the dual-task layer. To optimize the model's performance and reduce the need for manual hyperparameter tuning, the Hyperopt optimization algorithm is used. Extensive comparative experiments on battery aging datasets demonstrate that the proposed method reduces the average RMSE for SOH and RUL predictions by 111.3\% and 33.0\%, respectively, compared to traditional and state-of-the-art methods. ---