# Research Papers — 2026-03 · bundle page 156 of 243 > Full-text markdown of papers #7751–7800 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/156 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/155.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/157.md) ### Entropy Guided Diversification and Preference Elicitation in Agentic Recommendation Systems Authors: Dat Tran, Yongce Li, Hannah Clay, Negin Golrezaei, Sajjad Beygi, Amin Saberi Published: 2026-03-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.11399v1 Users on e-commerce platforms can be uncertain about their preferences early in their search. Queries to recommendation systems are frequently ambiguous, incomplete, or weakly specified. Agentic systems are expected to proactively reason, ask clarifying questions, and act on the user's behalf, which makes handling such ambiguity increasingly important. In existing platforms, ambiguity led to excessive interactions and question fatigue or overconfident recommendations prematurely collapsing the search space. We present an Interactive Decision Support System (IDSS) that addresses ambiguous user queries using entropy as a unifying signal. IDSS maintains a dynamically filtered candidate product set and quantifies uncertainty over item attributes using entropy. This uncertainty guides adaptive preference elicitation by selecting follow-up questions that maximize expected information gain. When preferences remain incomplete, IDSS explicitly incorporates residual uncertainty into downstream recommendations through uncertainty-aware ranking and entropy-based diversification, rather than forcing premature resolution. We evaluate IDSS using review-driven simulated users grounded in real user reviews, enabling a controlled study of diverse shopping behaviors. Our evaluation measures both interaction efficiency and recommendation quality. Results show that entropy-guided elicitation reduces unnecessary follow-up questions, while uncertainty-aware ranking and presentation yield more informative, diverse, and transparent recommendation sets under ambiguous intent. These findings demonstrate that entropy-guided reasoning provides an effective foundation for agentic recommendation systems operating under uncertainty. --- ### Efficient Cross-View Localization in 6G Space-Air-Ground Integrated Network Authors: Min Hao, Yanbing Xu, Maoqiang Wu, Jinglin Huang, Chen Shang, Jiacheng Wang, Jiawen Kang, Zhu Han, Wei Ni Published: 2026-03-12 Categories: cs.NI, cs.AI arXiv: https://arxiv.org/abs/2603.11398v3 Recently, visual localization has become an important supplement to improve localization reliability, and cross-view approaches can greatly enhance coverage and adaptability. Meanwhile, future 6G will enable a globally covered mobile communication system, with a space-air-ground integrated network (SAGIN) serving as key supporting architecture. Inspired by this, we explore an integration of cross-view localization (CVL) with 6G SAGIN, thereby enhancing its performance in latency, energy consumption, and privacy protection. First, we provide a comprehensive review of CVL and SAGIN, highlighting their capabilities, integration opportunities, and potential applications. Benefiting from the fast and extensive image collection and transmission capabilities of the 6G SAGIN architecture, CVL achieves higher localization accuracy and faster processing speed. Then, we propose a split-inference framework for implementing CVL, which fully leverages the distributed communication and computing resources of the 6G SAGIN architecture. Subsequently, we conduct joint optimization of communication, computation, and confidentiality within the proposed split-inference framework, aiming to provide a paradigm and a direction for making CVL efficient. Experimental results validate the effectiveness of the proposed framework and provide solutions to the optimization problem. Finally, we discuss potential research directions for 6G SAGIN-enabled CVL. --- ### Efficient Cross-View Localization in 6G Space-Air-Ground Integrated Network Authors: Min Hao, Yanbing Xu, Maoqiang Wu, Jinglin Huang, Chen Shang, Jiacheng Wang, Ruichen Zhang, Jiawen Kang, Dusit Niyato, Zhu Han, Wei Ni Published: 2026-03-12 Categories: cs.NI, cs.AI arXiv: https://arxiv.org/abs/2603.11398v1 Recently, visual localization has become an important supplement to improve localization reliability, and cross-view approaches can greatly enhance coverage and adaptability. Meanwhile, future 6G will enable a globally covered mobile communication system, with a space-air-ground integrated network (SAGIN) serving as key supporting architecture. Inspired by this, we explore an integration of cross-view localization (CVL) with 6G SAGIN, thereby enhancing its performance in latency, energy consumption, and privacy protection. First, we provide a comprehensive review of CVL and SAGIN, highlighting their capabilities, integration opportunities, and potential applications. Benefiting from the fast and extensive image collection and transmission capabilities of the 6G SAGIN architecture, CVL achieves higher localization accuracy and faster processing speed. Then, we propose a split-inference framework for implementing CVL, which fully leverages the distributed communication and computing resources of the 6G SAGIN architecture. Subsequently, we conduct joint optimization of communication, computation, and confidentiality within the proposed split-inference framework, aiming to provide a paradigm and a direction for making CVL efficient. Experimental results validate the effectiveness of the proposed framework and provide solutions to the optimization problem. Finally, we discuss potential research directions for 6G SAGIN-enabled CVL. --- ### Harnessing Data Asymmetry: Manifold Learning in the Finsler World Authors: Thomas Dagès, Simon Weber, Daniel Cremers, Ron Kimmel Published: 2026-03-12 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2603.11396v1 Manifold learning is a fundamental task at the core of data analysis and visualisation. It aims to capture the simple underlying structure of complex high-dimensional data by preserving pairwise dissimilarities in low-dimensional embeddings. Traditional methods rely on symmetric Riemannian geometry, thus forcing symmetric dissimilarities and embedding spaces, e.g. Euclidean. However, this discards in practice valuable asymmetric information inherent to the non-uniformity of data samples. We suggest to harness this asymmetry by switching to Finsler geometry, an asymmetric generalisation of Riemannian geometry, and propose a Finsler manifold learning pipeline that constructs asymmetric dissimilarities and embeds in a Finsler space. This greatly broadens the applicability of existing asymmetric embedders beyond traditionally directed data to any data. We also modernise asymmetric embedders by generalising current reference methods to asymmetry, like Finsler t-SNE and Finsler Umap. On controlled synthetic and large real datasets, we show that our asymmetric pipeline reveals valuable information lost in the traditional pipeline, e.g. density hierarchies, and consistently provides superior quality embeddings than their Euclidean counterparts. --- ### ARROW: Augmented Replay for RObust World models Authors: Abdulaziz Alyahya, Abdallah Al Siyabi, Markus R. Ernst, Luke Yang, Levin Kuhlmann, Gideon Kowadlo Published: 2026-03-12 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.11395v3 Continual reinforcement learning challenges agents to acquire new skills while retaining previously learned ones with the goal of improving performance in both past and future tasks. Most existing approaches rely on model-free methods with replay buffers to mitigate catastrophic forgetting; however, these solutions often face significant scalability challenges due to large memory demands. Drawing inspiration from neuroscience, where the brain replays experiences to a predictive World Model rather than directly to the policy, we present ARROW (Augmented Replay for RObust World models), a model-based continual RL algorithm that extends DreamerV3 with a memory-efficient, distribution-matching replay buffer. Unlike standard fixed-size FIFO buffers, ARROW maintains two complementary buffers: a short-term buffer for recent experiences and a long-term buffer that preserves task diversity through intelligent sampling. We evaluate ARROW on two challenging continual RL settings: Tasks without shared structure (Atari), and tasks with shared structure, where knowledge transfer is possible (Procgen CoinRun variants). Compared to model-free and model-based baselines with replay buffers of the same-size, ARROW demonstrates substantially less forgetting on tasks without shared structure, while maintaining comparable forward transfer. Our findings highlight the potential of model-based RL and bio-inspired approaches for continual reinforcement learning, warranting further research. --- ### Stop Listening to Me! How Multi-turn Conversations Can Degrade LLM Reliability Authors: Kevin H. Guo, Chao Yan, Avinash Baidya, Katherine Brown, Xiang Gao, Juming Xiong, Zhijun Yin, Bradley A. Malin Published: 2026-03-12 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.11394v3 Large language models (LLMs) excel on static benchmarks, but their performance across multi-turn conversations, which better reflect real-world usage, remains understudied. Addressing this gap is critical in high-stakes settings like healthcare, where patients and clinicians are turning to LLM chatbots to address their medical inquiries. Here, we introduce the "stick-or-switch" (SoS) framework, which partitions a question-answer space into multiple sequential presentations to model two safety-centric behaviors: conviction (i.e., sticking to a correct answer selection or abstention against incorrect suggestions) and flexibility (i.e., switching to a correct suggestion when it is introduced). Evaluating 17 LLMs across three clinical benchmarks, we observe a pervasive conversation tax, where partitioning an answer-space into sequential presentations reduces end-to-end accuracy and abstention against incorrect suggestions by an average of up to 30%, reaching 65% in certain models. We also observe blind switching, where models transition an initial abstention to incorrect and correct suggestions at near-identical rates reaching 50%. Finally, we show that increasing model scale mitigates some of these conversational inefficacies while exacerbating others, such as a higher propensity to adopt an incorrect suggestion from an initial abstention. Together our findings demonstrate that the general proficiency captured by static benchmarks do not translate over multi-turn dialogues. --- ### Agentic AI for Embodied-enhanced Beam Prediction in Low-Altitude Economy Networks Authors: Min Hao, Zhizhuo Li, Zirui Zhang, Maoqiang Wu, Han Zhang, Rong Yu Published: 2026-03-12 Categories: cs.NI, cs.AI arXiv: https://arxiv.org/abs/2603.11392v1 Millimeter-wave or terahertz communications can meet demands of low-altitude economy networks for high-throughput sensing and real-time decision making. However, high-frequency characteristics of wireless channels result in severe propagation loss and strong beam directivity, which make beam prediction challenging in highly mobile uncrewed aerial vehicles (UAV) scenarios. In this paper, we employ agentic AI to enable the transformation of mmWave base stations toward embodied intelligence. We innovatively design a multi-agent collaborative reasoning architecture for UAV-to-ground mmWave communications and propose a hybrid beam prediction model system based on bimodal data. The multi-agent architecture is designed to overcome the limited context window and weak controllability of large language model (LLM)-based reasoning by decomposing beam prediction into task analysis, solution planning, and completeness assessment. To align with the agentic reasoning process, a hybrid beam prediction model system is developed to process multimodal UAV data, including numeric mobility information and visual observations. The proposed hybrid model system integrates Mamba-based temporal modelling, convolutional visual encoding, and cross-attention-based multimodal fusion, and dynamically switches data-flow strategies under multi-agent guidance. Extensive simulations on a real UAV mmWave communication dataset demonstrate that proposed architecture and system achieve high prediction accuracy and robustness under diverse data conditions, with maximum top-1 accuracy reaching 96.57%. --- ### High-Precision 6DOF Pose Estimation via Global Phase Retrieval in Fringe Projection Profilometry for 3D Mapping Authors: Sehoon Tak, Keunhee Cho, Sangpil Kim, Jae-Sang Hyun Published: 2026-03-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.11389v1 Digital fringe projection (DFP) enables micrometer-level 3D reconstruction, yet extending it to large-scale mapping remains challenging because six-degree-of-freedom pose estimation often cannot match the reconstruction's precision. Conventional iterative closest point (ICP) registration becomes inefficient on multi-million-point clouds and typically relies on downsampling or feature-based selection, which can reduce local detail and degrade pose precision. Drift-correction methods improve long-term consistency but do not resolve sampling sensitivity in dense DFP point clouds.We propose a high-precision pose estimation method that augments a moving DFP system with a fixed, intrinsically calibrated global projector. Using the global projector's phase-derived pixel constraints and a PnP-style reprojection objective, the method estimates the DFP system pose in a fixed reference frame without relying on deterministic feature extraction, and we experimentally demonstrate sampling invariance under coordinate-preserving subsampling. Experiments demonstrate sub-millimeter pose accuracy against a reference with quantified uncertainty bounds, high repeatability under aggressive subsampling, robust operation on homogeneous surfaces and low-overlap views, and reduced error accumulation when used to correct ICP-based trajectories. The method extends DFP toward accurate 3D mapping in quasi-static scenarios such as inspection and metrology, with the trade-off of time-multiplexed acquisition for the additional projector measurements. --- ### Deactivating Refusal Triggers: Understanding and Mitigating Overrefusal in Safety Alignment Authors: Zhiyu Xue, Zimo Qi, Guangliang Liu, Bocheng Chen, Ramtin Pedarsani Published: 2026-03-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.11388v1 Safety alignment aims to ensure that large language models (LLMs) refuse harmful requests by post-training on harmful queries paired with refusal answers. Although safety alignment is widely adopted in industry, the overrefusal problem where aligned LLMs also reject benign queries after safety alignment post-training, remains insufficiently studied. Such an issue degrades the usability of safety alignment in real-world applications. In this paper, we examine how overrefusal arises under safety alignment, and propose a mitigation strategy inspired by our findings. We define refusal triggers as linguistic cues in the training data that elicit refusal responses, safety alignment encourages LLMs to associate refusal triggers within a training sample with refusal responses, leading aligned LLMs to refuse harmful queries. However, the refusal triggers include not only harmful linguistic cues but also non-harmful cues, therefore causing overrefusal to benign queries. Building on this mechanistic analysis, we propose a method that explicitly considers refusal triggers in the safety alignment fine-tuning. Empirical results demonstrate that our approach achieves a more favorable trade-off between defense against jailbreak attacks and responsiveness to benign queries, outperforming prior methods. Warning: this paper contains harmful and biased sentences. --- ### FedACT: Concurrent Federated Intelligence across Heterogeneous Data Sources Authors: Md Sirajul Islam, Isabelle G Chapman, N I Md Ashafuddula, Xu Yuan, Li Chen, Nian-Feng Tzeng, Klara Nahrstedt Published: 2026-03-11 Categories: cs.LG, cs.AI, cs.DC arXiv: https://arxiv.org/abs/2605.00011v1 Federated Learning (FL) enables collaborative intelligence across decentralized data source devices in a privacy-preserving way. While substantial research attention has been drawn to optimizing the learning process for an individual task, real-world applications increasingly require multiple machine learning tasks simultaneously training their models across a shared pool of devices. Naively applying single-FL optimization techniques in multi-FL systems results in suboptimal system performance, particularly due to device heterogeneity and resource inefficiency. To address such a critical open challenge, we introduce {\em FedACT}, a novel resource heterogeneity-aware device scheduling approach designed to efficiently schedule heterogeneous devices across multiple concurrent FL jobs, with the goal of minimizing their average job completion time (JCT). {\em FedACT} dynamically assigns devices to FL jobs based on an alignment scoring mechanism that evaluates the compatibility between available resources of devices and resource demands of jobs. Additionally, it incorporates participation fairness to ensure balanced contributions from devices across jobs, further enhancing the accuracy levels of learned global models. An optimal scheduling plan is formulated in {\em FedACT} by prioritizing devices with higher alignment scores, while ensuring fair participation across jobs. To evaluate the effectiveness of the proposed scheduling algorithm, we carried out comprehensive experiments using diverse FL jobs and benchmark datasets. Experimental results demonstrate that {\em FedACT} reduces the average JCT by up to 8.3\(\times\) and improves model accuracy by up to 44.5\%, compared to the state-of-the-art baselines. --- ### Beyond Single-Score Ranking: Facet-Aware Reranking for Controllable Diversity in Paper Recommendation Authors: Duan Ming Tao Published: 2026-03-11 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2604.16329v1 Current paper recommendation systems output a single similarity score that mixes different notions of relatedness, so users cannot specify why papers should be similar. We present SciFACE (Scientific Faceted Cross-Encoder), a reranking framework that models two independent facets: Background (what problem is studied) and Method (how it is solved). SciFACE trains two separate cross-encoders on 5,891 real seed-candidate paper pairs labeled by GPT-4o-mini with facet-specific criteria and validated against human judgments. On CSFCube, SciFACE reaches 70.63 NDCG@20 on Background (5.9 points above SPECTER) and 49.06 NDCG@20 on Method (31.1 points above SPECTER), competitive with state-of-the-art results. Compared with FaBLE without citation pre-training, SciFACE improves Method NDCG@20 by 4.1 points while using 5,891 labeled pairs versus 40K synthetic augmentations. These results show that high-quality grounded facet labels can be more data-efficient than large-scale synthetic augmentation for learning fine-grained scientific similarity. --- ### General-purpose LLMs as Models of Human Driver Behavior: The Case of Simplified Merging Authors: Samir H. A. Mohammad, Wouter Mooi, Arkady Zgonnikov Published: 2026-03-11 Categories: cs.AI, cs.RO arXiv: https://arxiv.org/abs/2604.09609v2 Human behavior models are essential as behavior references and for simulating human agents in virtual safety assessment of automated vehicles (AVs), yet current models face a trade-off between interpretability and flexibility. General-purpose large language models (LLMs) offer a promising alternative: a single model potentially deployable without parameter fitting across diverse scenarios. However, what LLMs can and cannot capture about human driving behavior remains poorly understood. We address this gap by embedding two general-purpose LLMs (OpenAI o3 and Google Gemini 2.5 Pro) as standalone, closed-loop driver agents in a simplified one-dimensional merging scenario and comparing their behavior against human data using quantitative and qualitative analyses. Both models reproduce human-like intermittent operational control and tactical dependencies on spatial cues. However, neither consistently captures the human response to dynamic velocity cues, and safety performance diverges sharply between models. A systematic prompt ablation study reveals that prompt components act as model-specific inductive biases that do not transfer across LLMs. These findings suggest that general-purpose LLMs could potentially serve as standalone, ready-to-use human behavior models in AV evaluation pipelines, but future research is needed to better understand their failure modes and ensure their validity as models of human driving behavior. --- ### Unifying Ontology Construction and Semantic Alignment for Deterministic Enterprise Reasoning at Scale Authors: Hongyin Zhu Published: 2026-03-11 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.09608v1 While enterprises amass vast quantities of data, much of it remains chaotic and effectively dormant, preventing decision-making based on comprehensive information. Existing neuro-symbolic approaches rely on disjoint pipelines and struggle with error propagation. We introduce the large ontology model (LOM), a unified framework that seamlessly integrates ontology construction, semantic alignment, and logical reasoning into a single end-to-end architecture. LOM employs a construct-align-reason (CAR) pipeline, leveraging its unified architecture across all three stages: it first autonomously constructs a domain-specific ontological universe from raw data, then aligns neural generation with this structural reality using a graph-aware encoder and reinforcement learning, and finally executes deterministic reasoning over the constructed topology, node attributes and relation types. We evaluate LOM on a comprehensive benchmark constructed from diverse real-world enterprise datasets. Experimental results demonstrate that LOM-4B achieves 88.8% accuracy in ontology completion and 94% in complex graph reasoning tasks, significantly outperforming state-of-the-art LLMs. These findings validate that autonomous logical construction is essential for achieving deterministic, enterprise-grade intelligence. --- ### GNN-as-Judge: Unleashing the Power of LLMs for Graph Learning with GNN Feedback Authors: Ruiyao Xu, Kaize Ding Published: 2026-03-11 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.08553v1 Large Language Models (LLMs) have shown strong performance on text-attributed graphs (TAGs) due to their superior semantic understanding ability on textual node features. However, their effectiveness as predictors in the low-resource setting, where labeled nodes are severely limited and scarce, remains constrained since fine-tuning LLMs usually requires sufficient labeled data, especially when the TAG shows complex structural patterns. In essence, this paper targets two key challenges: (i) the difficulty of generating and selecting reliable pseudo labels on TAGs for LLMs, and (ii) the need to mitigate potential label noise when fine-tuning LLMs with pseudo labels. To counter the challenges, we propose a new framework, GNN-as-Judge, which can unleash the power of LLMs for few-shot semi-supervised learning on TAGs by incorporating the structural inductive bias of Graph Neural Networks (GNNs). Specifically, GNN-as-Judge introduces a collaborative pseudo-labeling strategy that first identifies the most influenced unlabeled nodes from labeled nodes, then exploits both the agreement and disagreement patterns between LLMs and GNNs to generate reliable labels. Furthermore, we develop a weakly-supervised LLM fine-tuning algorithm that can distill the knowledge from informative pseudo labels while mitigating the potential label noise. Experiments on multiple TAG datasets demonstrate that GNN-as-Judge significantly outperforms existing methods, particularly in low-resource regimes where labeled data are scarce. --- ### Depression Detection at the Point of Care: Automated Analysis of Linguistic Signals from Routine Primary Care Encounters Authors: Feng Chen, Manas Bedmutha, Janice Sabin, Andrea Hartzler, Nadir Weibel, Trevor Cohen Published: 2026-03-11 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.06193v1 Depression is underdiagnosed in primary care, yet timely identification remains critical. Recorded clinical encounters, increasingly common with digital scribing technologies, present an opportunity to detect depression from naturalistic dialogue. We investigated automated depression detection from 1,108 audio-recorded primary care encounters in the Establishing Focus study, with depression defined by PHQ-9 (n=253 depressed, n=855 non-depressed). We compared three supervised approaches, Sentence-BERT + Logistic Regression (LR), LIWC+LR and ModernBERT, against a zero-shot GPT-OSS. GPT-OSS achieved the strongest performance (AUPRC=0.510, AUROC=0.774), with LIWC+LR competitive among supervised models (AUPRC=0.500, AUROC=0.742). Combined dyadic transcripts outperformed single-speaker configurations, with providers linguistically mirroring patients in depression encounters, an additive signal not captured by either speaker alone. Meaningful detection is achievable from the first 128 patient tokens (AUPRC=0.356, AUROC=0.675), supporting in-the-moment clinical decision support. These findings argue for passively collected clinical audio as a low-burden complement to existing screening workflows. --- ### The Stepwise Informativeness Assumption: Why are Entropy Dynamics and Reasoning Correlated in LLMs? Authors: Mar Gonzàlez I Català, Haitz Sáez de Ocáriz Borde, George D. Montañez, Pietro Liò Published: 2026-03-11 Categories: cs.CL, cs.AI, cs.IT, cs.LG arXiv: https://arxiv.org/abs/2604.06192v1 Recent work uses entropy-based signals at multiple representation levels to study reasoning in large language models, but the field remains largely empirical. A central unresolved puzzle is why internal entropy dynamics, defined under the predictive distribution of a model, correlate so robustly with external correctness given by the ground-truth answer. In this paper, we argue that this correlation arises because autoregressive models reason correctly when they accumulate information about the true answer via answer-informative prefixes. We formalize this intuition via the Stepwise Informativeness Assumption (SIA), which states that reasoning prefixes accumulate answer-relevant information in expectation as generation progresses. We show that SIA naturally emerges from maximum-likelihood optimization on human reasoning traces and is reinforced by standard fine-tuning and reinforcement-learning pipelines. We then derive observable signatures of SIA linking conditional answer entropy dynamics to correctness. We empirically test SIA across multiple reasoning benchmarks (GSM8K, ARC, SVAMP) and a diverse set of open-weight LLMs (Gemma-2, LLaMA-3.2, Qwen-2.5, DeepSeek and Olmo variants), showing that training induces it and that correct traces exhibit characteristic conditional answer entropy patterns. --- ### Harf-Speech: A Clinically Aligned Framework for Arabic Phoneme-Level Speech Assessment Authors: Asif Azad, MD Sadik Hossain Shanto, Mohammad Sadat Hossain, Bdour Alwuqaysi, Sabri Boughorbel, Yahya Bokhari, Abdulrhman Aljouie, Ayah Othman Sindi, Ehsan Hoque Published: 2026-03-11 Categories: eess.AS, cs.AI, cs.CL, cs.SD arXiv: https://arxiv.org/abs/2604.06191v2 Automated phoneme-level pronunciation assessment is vital for scalable speech therapy and language learning, yet validated tools for Arabic remain scarce. We present Harf-Speech, a modular system scoring Arabic pronunciation at the phoneme level on a clinical scale. It combines an MSA phonetizer, a fine-tuned speech-to-phoneme model, Levenshtein alignment, and a blended scorer using longest common subsequence and edit-distance metrics. We fine-tune three ASR architectures on Arabic phoneme data and benchmark them with zero-shot multimodal models; the best, OmniASR-CTC-1B-v2, achieves 8.92% phoneme error rate. Three certified speech-language pathologists independently scored 40 utterances for clinical validation. Harf-Speech attains a Pearson correlation of 0.791 and ICC(2,1) of 0.659 with mean expert scores, outperforming existing end-to-end assessment frameworks. These results show Harf-Speech yields clinically aligned, interpretable scores comparable to inter-rater expert agreement. --- ### Robust LLM Performance Certification via Constrained Maximum Likelihood Estimation Authors: Minghe Shen, Ananth Balashankar, Adam Fisch, David Madras, Miguel Rodrigues Published: 2026-03-11 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.03257v1 The ability to rigorously estimate the failure rates of large language models (LLMs) is a prerequisite for their safe deployment. Currently, however, practitioners often face a tradeoff between expensive human gold standards and potentially severely-biased automatic annotation schemes such as "LLM-as-a-Judge" labeling. In this paper, we propose a new, practical, and efficient approach to LLM failure rate estimation based on constrained maximum-likelihood estimation (MLE). Our method integrates three distinct signal sources: (i) a small, high-quality human-labeled calibration set, (ii) a large corpus of LLM-judge annotations, and, most importantly, (iii) additional side information via domain-specific constraints derived from known bounds on judge performance statistics. We validate our approach through a comprehensive empirical study, benchmarking it against state-of-the-art baselines like Prediction-Powered Inference (PPI). Across diverse experimental regimes -- spanning varying judge accuracies, calibration set sizes, and LLM failure rates -- our constrained MLE consistently delivers more accurate and lower-variance estimates than existing methods. By moving beyond the "black-box" use of automated judges to a flexible framework, we provide a principled, interpretable, and scalable pathway towards LLM failure-rate certification. --- ### Is your AI Model Accurate Enough? The Difficult Choices Behind Rigorous AI Development and the EU AI Act Authors: Lucas G. Uberti-Bona Marin, Bram Rijsbosch, Kristof Meding, Gerasimos Spanakis, Gijs van Dijck, Konrad Kollnig Published: 2026-03-11 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2604.03254v2 Technical and legal debates frequently suggest that "accuracy" is an objective, measurable, and purely technical property. We challenge this view, showing that evaluating AI performance fundamentally depends on context-dependent normative decisions. These techno-normative choices are crucial for rigorous AI deployment, as they determine which errors are prioritised, how risks are distributed, and how trade-offs between competing objectives are resolved. This paper provides a legal-technical analysis of the choices that shape how accuracy is defined, measured, and assessed, using the 2024 European Union AI Act -- which mandates an "appropriate level of accuracy" for high-risk systems -- as a primary case study. We identify and analyse four choices central to any robust performance evaluation: (1) selecting metrics, (2) balancing multiple metrics, (3) measuring metrics against representative data, and (4) determining acceptance thresholds. For each choice, we study its relationship to the AI Act's accuracy requirement and associated documentation obligations, show how its technical implementation embeds implicit or explicit assumptions about acceptable risks, errors, and trade-offs, and discuss the implications for the practical implementation of the AI Act by examples and related technical standards. By making the techno-normative dimensions of accuracy explicit, this paper contributes to broader interdisciplinary debates on AI governance and regulation, and offers specific guidance for regulators, auditors, and developers tasked with translating (legal) safety requirements into technical practice. --- ### Self-Execution Simulation Improves Coding Models Authors: Gallil Maimon, Ori Yoran, Felix Kreuk, Michael Hassid, Gal Cohen, Pierre Chambon, Yossi Adi Published: 2026-03-11 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.03253v1 A promising research direction in enabling LLMs to generate consistently correct code involves addressing their inability to properly estimate program execution, particularly for code they generate. In this work, we demonstrate that Code LLMs can be trained to simulate program execution in a step-by-step manner and that this capability can be leveraged to improve competitive programming performance. Our approach combines supervised fine-tuning on natural language execution traces, textual explanations grounded in true execution, with reinforcement learning using verifiable rewards. We introduce two complementary objectives: output prediction given code and inputs, and solving competitive programming tasks with either ground-truth or self-predicted execution feedback. These objectives enable models to perform self-verification over multiple candidate solutions, and iterative self-fixing by simulating test execution. Across multiple competitive programming benchmarks, our method yields consistent improvements over standard reasoning approaches. We further present ablations and analysis to elucidate the role of execution simulation and its limitations. --- ### Evaluating Digital Inclusiveness of Digital Agri-Food Tools Using Large Language Models: A Comparative Analysis Between Human and AI-Based Evaluations Authors: Githma Pewinya, Carolina Martins, Garcia Mariangel Published: 2026-03-11 Categories: cs.CY, cs.CL arXiv: https://arxiv.org/abs/2604.03252v1 Ensuring digital inclusiveness is a critical priority in agri-food systems, particularly in the Global South, where digital divides persist. The Multidimensional Digital Inclusiveness Index (MDII) offers a comprehensive, human-led framework to assess how inclusive digital agricultural tools (agritools) are. However, the current evaluation process is resource intensive, often requiring months to complete. This study explores whether large language models (LLMs) can support a rapid, AI-enabled assessment of digital inclusiveness, complementing the MDII's existing workflow. Using a comparative analysis, the research benchmarks the performance of four LLMs (Grok, Gemini, GPT-4o, and GPT-5) against prior expert-led evaluations. The study investigates model alignment with human scores, sensitivity to temperature settings, and potential sources of bias. Findings suggest that LLMs can generate evaluative outputs that approximate expert judgment in some dimensions, though reliability varies across models and contexts. This exploratory work provides early evidence for the integration of GenAI into inclusive digital development monitoring, with implications for scaling evaluations in time-sensitive or resource-constrained environments. --- ### Multi-lingual Multi-institutional Electronic Health Record based Predictive Model Authors: Kyunghoon Hur, Heeyoung Kwak, Jinsu Jang, Nakhwan Kim, Edward Choi Published: 2026-03-11 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.00027v1 Large-scale EHR prediction across institutions is hindered by substantial heterogeneity in schemas and code systems. Although Common Data Models (CDMs) can standardize records for multi-institutional learning, the manual harmonization and vocabulary mapping are costly and difficult to scale. Text-based harmonization provides an alternative by converting raw EHR into a unified textual form, enabling pooled learning without explicit standardization. However, applying this paradigm to multi-national datasets introduces an additional layer of heterogeneity, which is "language" that must be addressed for truly scalable EHRs learning. In this work, we investigate multilingual multi-institutional learning for EHR prediction, aiming to enable pooled training across multinational ICU datasets without manual standardization. We compare two practical strategies for handling language barriers: (i) directly modeling multilingual records with multilingual encoders, and (ii) translating non-English records into English via LLM-based word-level translation. Across seven public ICU datasets, ten clinical tasks with multiple prediction windows, translation-based lingual alignment yields more reliable cross-dataset performance than multilingual encoders. The multi-institutional learning model consistently outperforms strong baselines that require manual feature selection and harmonization, and also surpasses single-dataset training. We further demonstrate that text-based framework with lingual alignment effectively performs transfer learning via few-shot fine-tuning, with additional gains. To our knowledge, this is the first study to aggregate multilingual multinational ICU EHR datasets into one predictive model, providing a scalable path toward language-agnostic clinical prediction and future global multi-institutional EHR research. --- ### "Who Am I, and Who Else Is Here?" Behavioral Differentiation Without Role Assignment in Multi-Agent LLM Systems Authors: Houssam EL Kandoussi Published: 2026-03-11 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.00026v1 When multiple large language models interact in a shared conversation, do they develop differentiated social roles or converge toward uniform behavior? We present a controlled experimental platform that orchestrates simultaneous multi-agent discussions among 7 heterogeneous LLMs on a unified inference backend, systematically varying group composition, naming conventions, and prompt structure across 12 experimental series (208 runs, 13,786 coded messages). Each message is independently coded on six behavioral flags by two LLM judges from distinct model families (Gemini 3.1 Pro and Claude Sonnet 4.6), achieving mean Cohen's kappa = 0.78 with conservative intersection-based adjudication. Human validation on 609 randomly stratified messages confirmed coding reliability (mean kappa = 0.73 vs. Gemini). We find that (1) heterogeneous groups exhibit significantly richer behavioral differentiation than homogeneous groups (cosine similarity 0.56 vs. 0.85; p < 10^-5, r = 0.70); (2) groups spontaneously exhibit compensatory response patterns when an agent crashes; (3) revealing real model names significantly increases behavioral convergence (cosine 0.56 to 0.77, p = 0.001); and (4) removing all prompt scaffolding converges profiles to homogeneous-level similarity (p < 0.001). Critically, these behaviors are absent when agents operate in isolation, confirming that behavioral diversity is a structured, reproducible phenomenon driven by the interaction of architectural heterogeneity, group context, and prompt-level scaffolding. --- ### Brevity Constraints Reverse Performance Hierarchies in Language Models Authors: MD Azizul Hakim Published: 2026-03-11 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.00025v1 Standard evaluation protocols reveal a counterintuitive phenomenon: on 7.7% of benchmark problems spanning five datasets, larger language models underperform smaller ones by 28.4 percentage points despite 10-100x more parameters. Through systematic evaluation of 31 models (0.5B-405B parameters) across 1,485 problems, we identify the mechanism as spontaneous scale-dependent verbosity that introduces errors through overelaboration. Causal intervention experiments demonstrate this reflects correctable prompt design rather than fundamental capability limitations. Constraining large models to produce brief responses improves accuracy by 26 percentage points and reduces performance gaps by up to two-thirds. Most critically, brevity constraints completely reverse performance hierarchies on mathematical reasoning and scientific knowledge benchmarks, with large models achieving 7.7-15.9 percentage point advantages over small models -- direct inversions of the original gaps. These reversals prove large models possess superior latent capabilities that universal prompting masks. We validate findings through three independent contamination tests and demonstrate inverse scaling operates continuously across the full parameter spectrum, with dataset-specific optimal scales ranging from 0.5B to 3.0B parameters. Our results establish that maximizing large model performance requires scale-aware prompt engineering rather than universal evaluation protocols, with immediate implications for deployment: prompt adaptation simultaneously improves accuracy and reduces computational costs. --- ### WHBench: Evaluating Frontier LLMs with Expert-in-the-Loop Validation on Women's Health Topics Authors: Sneha Maurya, Spandana Govindgari, Girish Kumar, Akhara AI Published: 2026-03-11 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2604.00024v2 Large language models are increasingly used for medical guidance, but women's health remains under-evaluated in benchmark design. We present the Women's Health Benchmark (WHBench), a targeted evaluation suite of 47 expert-crafted scenarios across 10 women's health topics, designed to expose clinically meaningful failure modes including outdated guidelines, unsafe omissions, dosing errors, and equity-related blind spots. We evaluate 22 models using a 23-criterion rubric spanning clinical accuracy, completeness, safety, communication quality, instruction following, equity, uncertainty handling, and guideline adherence, with safety-weighted penalties and server-side score recalculation. Across 3,102 attempted responses (3,100 scored), no model mean performance exceeds 75 percent; the best model reaches 72.1 percent. Even top models show low fully correct rates and substantial variation in harm rates. Inter-rater reliability is moderate at the response label level but high for model ranking, supporting WHBench utility for comparative system evaluation while highlighting the need for expert oversight in clinical deployment. WHBench provides a public, failure-mode-aware benchmark to track safer and more equitable progress in womens health AI. --- ### Phonological Fossils: Machine Learning Detection of Non-Mainstream Vocabulary in Sulawesi Basic Lexicon Authors: Mukhlis Amien, Go Frendi Gunawan Published: 2026-03-11 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.00023v1 Basic vocabulary in many Sulawesi Austronesian languages includes forms resisting reconstruction to any proto-form with phonological patterns inconsistent with inherited roots, but whether this non-conforming vocabulary represents pre-Austronesian substrate or independent innovation has not been tested computationally. We combine rule-based cognate subtraction with a machine learning classifier trained on phonological features. Using 1,357 forms from six Sulawesi languages in the Austronesian Basic Vocabulary Database, we identify 438 candidate substrate forms (26.5%) through cognate subtraction and Proto-Austronesian cross-checking. An XGBoost classifier trained on 26 phonological features distinguishes inherited from non-mainstream forms with AUC=0.763, revealing a phonological fingerprint: longer forms, more consonant clusters, higher glottal stop rates, and fewer Austronesian prefixes. Cross-method consensus (Cohen's kappa=0.61) identifies 266 high-confidence non-mainstream candidates. However, clustering yields no coherent word families (silhouette=0.114; cross-linguistic cognate test p=0.569), providing no evidence for a single pre-Austronesian language layer. Application to 16 additional languages confirms geographic patterning: Sulawesi languages show higher predicted non-mainstream rates (mean P_sub=0.606) than Western Indonesian languages (0.393). This study demonstrates that phonological machine learning can complement traditional comparative methods in detecting non-mainstream lexical layers, while cautioning against interpreting phonological non-conformity as evidence for a shared substrate language. --- ### Criterion Validity of LLM-as-Judge for Business Outcomes in Conversational Commerce Authors: Liang Chen, Qi Liu, Wenhuan Lin, Feng Liang Published: 2026-03-11 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.00022v1 Multi-dimensional rubric-based dialogue evaluation is widely used to assess conversational AI, yet its criterion validity -- whether quality scores are associated with the downstream outcomes they are meant to serve -- remains largely untested. We address this gap through a two-phase study on a major Chinese matchmaking platform, testing a 7-dimension evaluation rubric (implemented via LLM-as-Judge) against verified business conversion. Our findings concern rubric design and weighting, not LLM scoring accuracy: any judge using the same rubric would face the same structural issue. The core finding is dimension-level heterogeneity: in Phase 2 (n=60 human conversations, stratified sample, verified labels), Need Elicitation (D1: rho=0.368, p=0.004) and Pacing Strategy (D3: rho=0.354, p=0.006) are significantly associated with conversion after Bonferroni correction, while Contextual Memory (D5: rho=0.018, n.s.) shows no detectable association. This heterogeneity causes the equal-weighted composite (rho=0.272) to underperform its best dimensions -- a composite dilution effect that conversion-informed reweighting partially corrects (rho=0.351). Logistic regression controlling for conversation length confirms D3's association strengthens (OR=3.18, p=0.006), ruling out a length confound. An initial pilot (n=14) mixing human and AI conversations had produced a misleading "evaluation-outcome paradox," which Phase 2 revealed as an agent-type confound artifact. Behavioral analysis of 130 conversations through a Trust-Funnel framework identifies a candidate mechanism: AI agents execute sales behaviors without building user trust. We operationalize these findings in a three-layer evaluation architecture and advocate criterion validity testing as standard practice in applied dialogue evaluation. --- ### How Do Language Models Process Ethical Instructions? Deliberation, Consistency, and Other-Recognition Across Four Models Authors: Hiroki Fukui Published: 2026-03-11 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2604.00021v1 Alignment safety research assumes that ethical instructions improve model behavior, but how language models internally process such instructions remains unknown. We conducted over 600 multi-agent simulations across four models (Llama 3.3 70B, GPT-4o mini, Qwen3-Next-80B-A3B, Sonnet 4.5), four ethical instruction formats (none, minimal norm, reasoned norm, virtue framing), and two languages (Japanese, English). Confirmatory analysis fully replicated the Llama Japanese dissociation pattern from a prior study ($\mathrm{BF}_{10} > 10$ for all three hypotheses), but none of the other three models reproduced this pattern, establishing it as model-specific. Three new metrics -- Deliberation Depth (DD), Value Consistency Across Dilemmas (VCAD), and Other-Recognition Index (ORI) -- revealed four distinct ethical processing types: Output Filter (GPT; safe outputs, no processing), Defensive Repetition (Llama; high consistency through formulaic repetition), Critical Internalization (Qwen; deep deliberation, incomplete integration), and Principled Consistency (Sonnet; deliberation, consistency, and other-recognition co-occurring). The central finding is an interaction between processing capacity and instruction format: in low-DD models, instruction format has no effect on internal processing; in high-DD models, reasoned norms and virtue framing produce opposite effects. Lexical compliance with ethical instructions did not correlate with any processing metric at the cell level ($r = -0.161$ to $+0.256$, all $p > .22$; $N = 24$; power limited), suggesting that safety, compliance, and ethical processing are largely dissociable. These processing types show structural correspondence to patterns observed in clinical offender treatment, where formal compliance without internal processing is a recognized risk signal. --- ### Detecting Abnormal User Feedback Patterns through Temporal Sentiment Aggregation Authors: Yalun Qi, Sichen Zhao, Zhiming Xue, Xianling Zeng, Zihan Yu Published: 2026-03-11 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.00020v1 In many real-world applications, such as customer feedback monitoring, brand reputation management, and product health tracking, understanding the temporal dynamics of user sentiment is crucial for early detection of anomalous events such as malicious review campaigns or sudden declines in user satisfaction. Traditional sentiment analysis methods focus on individual text classification, which is insufficient to capture collective behavioral shifts over time due to inherent noise and class imbalance in short user comments. In this work, we propose a temporal sentiment aggregation framework that leverages pretrained transformer-based language models to extract per-comment sentiment signals and aggregates them into time-window-level scores. Significant downward shifts in these aggregated scores are interpreted as potential anomalies in user feedback patterns. We adopt RoBERTa as our core semantic feature extractor and demonstrate, through empirical evaluation on real social media data, that the aggregated sentiment scores reveal meaningful trends and support effective anomaly detection. Experiments on real-world social media data demonstrate that our method successfully identifies statistically significant sentiment drops that correspond to coherent complaint patterns, providing an effective and interpretable solution for feedback anomaly monitoring. --- ### The Chronicles of RiDiC: Generating Datasets with Controlled Popularity Distribution for Long-form Factuality Evaluation Authors: Pavel Braslavski, Dmitrii Iarosh, Nikita Sushko, Andrey Sakhovskiy, Vasily Konovalov, Elena Tutubalina, Alexander Panchenko Published: 2026-03-11 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.00019v1 We present a configurable pipeline for generating multilingual sets of entities with specified characteristics, such as domain, geographical location and popularity, using data from Wikipedia and Wikidata. These datasets are intended for evaluating the factuality of LLMs' long-form generation, thereby complementing evaluation based on short-form QA datasets. We present the RiDiC dataset as an example of this approach. RiDiC contains 3,000 entities from three domains -- rivers, natural disasters, and car models -- spanning different popularity tiers. Each entity is accompanied by its geographical location, English and Chinese names (if available) and relevant English and Chinese Wikipedia content, which is used to evaluate LLMs' responses. Generations about RiDiC entities were obtained from three LLMs in English and Chinese. These were then evaluated using a third-party factuality checker, which showed that entities from our dataset caused even frontier models to hallucinate. To facilitate the evaluation of LLMs' long-form factuality in multiple languages, the code, data, and generation/evaluation scripts have been released. --- ### TIPS: Turn-Level Information-Potential Reward Shaping for Search-Augmented LLMs Authors: Yutao Xie, Nathaniel Thomas, Nicklas Hansen, Yang Fu, Li Erran Li, Xiaolong Wang Published: 2026-03-11 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.22293v1 Search-augmented large language models (LLMs) trained with reinforcement learning (RL) have achieved strong results on open-domain question answering (QA), but training still remains a significant challenge. The optimization is often unstable due to sparse rewards and difficult credit assignments across reasoning and tool calls. To address this, we introduce Turn-Level Information Potential Reward Shaping (TIPS), a simple framework that assigns dense, turn-level rewards to each reasoning + tool-call segment based on the increased likelihood of the correct answer under a teacher model. By leveraging the potential-based reward shaping, TIPS offers fine-grained and policy-invariant guidance that overcomes the limitations of outcome-only optimization. Evaluated on seven QA benchmarks, TIPS consistently outperforms GRPO/PPO baselines and substantially improves training stability. For instance, with a Qwen-2.5 7B Instruct model, TIPS improves the average Exact Match score by 11.8% and F1 by 13.6% relative to PPO. Our results demonstrate that turn-level information-potential reward shaping provides an effective and general solution to sparse-reward credit assignment for multi-turn LLM reasoning. --- ### The Deep-Match Framework for Event-Related Potential Detection in EEG Authors: Marek Zylinski, Bartosz Tomasz Smigielski, Gerard Cybulski Published: 2026-03-11 Categories: eess.SP, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.20258v1 Reliable detection of event-related potentials (ERPs) at the single-trial level remains a major challenge due to the low signal-to-noise ratio EEG recordings. In this work, we investigate whether incorporating prior knowledge about ERP templates into deep learning models can improve detection performance. We employ the Deep-Match framework for ERP detection using multi-channel EEG signals. The model is trained in two stages. First, an encoder-decoder architecture is trained to reconstruct input EEG signals, enabling the network to learn compact signal representations. In the second stage, the decoder is replaced with a detection module, and the network is fine-tuned for ERP identification. Two model variants are evaluated: a standard model with randomly initialized filters and a Deep-MF model in which input kernels are initialized using ERP templates. Model performance is assessed on a single-trial ERP detection task using leave-one-subject-out validation. The proposed Deep-MF model slightly outperforms the detector with standard kernel initialization for most held-out subjects. Despite substantial inter-subject variability, Deep-MF achieves a higher average F1-score (0.37) compared to the standard network (0.34), indicating improved robustness to cross-subject differences. The best performance obtained by Deep-MF reaches an F1-score of 0.71, exceeding the maximum score achieved by the standard model (0.59). These results demonstrate that ERP-informed kernel initialization can provide consistent improvements in subject-independent single-trial ERP detection. Overall, the findings highlight the potential of integrating domain knowledge with deep learning architectures for EEG analysis. The proposed approach represents a step toward practical wearable EEG and passive brain-computer interface systems capable of real-time monitoring of cognitive processes. --- ### SciNav: A General Agent Framework for Scientific Coding Tasks Authors: Tianshu Zhang, Huan Sun Published: 2026-03-11 Categories: cs.CL, cs.AI, cs.CE, cs.LG, cs.MA, eess.SY arXiv: https://arxiv.org/abs/2603.20256v1 Autonomous science agents built on large language models (LLMs) are increasingly used to generate hypotheses, design experiments, and produce reports. However, prior work mainly targets open-ended scientific problems with subjective outputs that are difficult to evaluate. Scientific coding benchmarks, by contrast, provide executable outputs for objective assessment. Existing approaches remain engineering-driven pipelines, revealing the need for structured, end-to-end science agent frameworks for scientific coding tasks. We address this gap by focusing on scientific coding tasks, where evaluation can be made rigorously, and introducing an agent framework SciNav (Scientific Navigator) that enables more effective solution exploration. Our framework is designed to operate under constrained search budgets, moving beyond reliance on pre-defined success metrics and prolonged search cycles. Inspired by findings that comparative judgments often reveal finer-grained quality differences and therefore provide greater discriminative power than absolute scoring, our framework leverages pairwise relative judgments within a tree search process to select top-K promising solution branches, prune low-potential ones, and progressively narrow down the solution candidates on the selected branches guided by relative comparisons. We demonstrate our agent's effectiveness across different types of tasks on two benchmarks. Experiments show that SciNav significantly outperforms direct prompting and prior agents like OpenHands and Self-Debug across different base models, task types, and difficulty levels, and exceeds different frontier comparators such as random selection and LLM absolute scoring. These results confirm the strength of our agent design and highlight the effectiveness of relative judgment-guided top-K search for high-quality scientific coding, marking a step toward more practical science agents. --- ### Abjad-Kids: An Arabic Speech Classification Dataset for Primary Education Authors: Abdul Aziz Snoubara, Baraa Al_Maradni, Haya Al_Naal, Malek Al_Madrmani, Roaa Jdini, Seedra Zarzour, Khloud Al Jallad Published: 2026-03-11 Categories: cs.CL, cs.HC, cs.LG, cs.SD, eess.AS arXiv: https://arxiv.org/abs/2603.20255v1 Speech-based AI educational applications have gained significant interest in recent years, particularly for children. However, children speech research remains limited due to the lack of publicly available datasets, especially for low-resource languages such as Arabic.This paper presents Abjad-Kids, an Arabic speech dataset designed for kindergarten and primary education, focusing on fundamental learning of alphabets, numbers, and colors. The dataset consists of 46397 audio samples collected from children aged 3 - 12 years, covering 141 classes. All samples were recorded under controlled specifications to ensure consistency in duration, sampling rate, and format. To address high intra-class similarity among Arabic phonemes and the limited samples per class, we propose a hierarchical audio classification based on CNN-LSTM architectures. Our proposed methodology decomposes alphabet recognition into a two-stage process: an initial grouping classification model followed by specialized classifiers for each group. Both strategies: static linguistic-based grouping and dynamic clustering-based grouping, were evaluated. Experimental results demonstrate that static linguistic-based grouping achieves superior performance. Comparisons between traditional machine learning with deep learning approaches, highlight the effectiveness of CNN-LSTM models combined with data augmentation. Despite achieving promising results, most of our experiments indicate a challenge with overfitting, which is likely due to the limited number of samples, even after data augmentation and model regularization. Thus, future work may focus on collecting additional data to address this issue. Abjad-Kids will be publicly available. We hope that Abjad-Kids enrich children representation in speech dataset, and be a good resource for future research in Arabic speech classification for kids. --- ### AI Detectors Fail Diverse Student Populations: A Mathematical Framing of Structural Detection Limits Authors: Nathan Garland Published: 2026-03-11 Categories: cs.CY, cs.AI, stat.OT arXiv: https://arxiv.org/abs/2603.20254v1 Student experiences and empirical studies report that "black box" AI text detectors produce high false positive rates with disproportionate errors against certain student populations, yet typically theoretical analyses model detection as a test between two known distributions for human and AI prose. This framing omits the structural feature of university assessment whereby an assessor generally does not know the individual student's writing distribution, making the null hypothesis composite. Standard application of the variational characterisation of total variation distance to this composite null shows trade-off bounds that any text-only, one-shot detector with useful power must produce false accusations at a rate governed by the distributional overlap between student writing and AI output. This is a constraint arising from population diversity that is logically independent of AI model quality and cannot be overcome by better detector engineering or technology. A subgroup mixture bound connects these quantities to observable demographic groups, providing a theoretical basis for the disparate impact patterns documented empirically. We propose suggestions to improve policy and practice, and argue that detection scores should not serve as sole evidence in misconduct proceedings. --- ### SimulCost: A Cost-Aware Benchmark and Toolkit for Automating Physics Simulations with LLMs Authors: Yadi Cao, Sicheng Lai, Jiahe Huang, Yang Zhang, Zach Lawrence, Rohan Bhakta, Izzy F. Thomas, Mingyun Cao, Chung-Hao Tsai, Zihao Zhou, Yidong Zhao, Hao Liu et al. Published: 2026-03-11 Categories: physics.comp-ph, cs.AI, cs.DC, cs.LG arXiv: https://arxiv.org/abs/2603.20253v4 Evaluating LLM agents for scientific tasks has focused on token costs while ignoring tool-use costs like simulation time and experimental resources. As a result, metrics like pass@k become impractical under realistic budget constraints. To address this gap, we introduce SimulCost, the first benchmark targeting cost-sensitive parameter tuning in physics simulations. SimulCost compares LLM tuning cost-sensitive parameters against traditional scanning approach in both accuracy and computational cost, spanning 2,643 single-round (initial guess) and 2,304 multi-round (adjustment by trial-and-error) tasks across 11 simulators from fluid dynamics, solid mechanics, and plasma physics, whose costs are analytically defined and platform-independent. A twelfth simulator, a production plasma code measurable only by wall clock, is reported separately. Frontier LLMs achieve 45-62% success rates in single-round mode, dropping to 34-50% under high accuracy requirements, rendering their initial guesses unreliable especially for high accuracy tasks. Multi-round mode improves rates to 66-81%, but LLMs are 1.5-2.7x slower than traditional scanning, making them uneconomical choices. We also investigate parameter group correlations for knowledge transfer potential, and the impact of in-context examples and reasoning effort, providing practical implications for deployment and fine-tuning. We open-source SimulCost as a static benchmark and extensible toolkit to facilitate research on improving cost-aware agentic designs for physics simulations, and for expanding new simulation environments. Code and data are available at https://github.com/Rose-STL-Lab/SimulCost-Bench --- ### SimulCost: A Cost-Aware Benchmark and Toolkit for Automating Physics Simulations with LLMs Authors: Yadi Cao, Sicheng Lai, Jiahe Huang, Yang Zhang, Zach Lawrence, Rohan Bhakta, Izzy F. Thomas, Mingyun Cao, Chung-Hao Tsai, Zihao Zhou, Yidong Zhao, Hao Liu et al. Published: 2026-03-11 Categories: physics.comp-ph, cs.AI, cs.DC, cs.LG arXiv: https://arxiv.org/abs/2603.20253v2 Evaluating LLM agents for scientific tasks has focused on token costs while ignoring tool-use costs like simulation time and experimental resources. As a result, metrics like pass@k become impractical under realistic budget constraints. To address this gap, we introduce SimulCost, the first benchmark targeting cost-sensitive parameter tuning in physics simulations. SimulCost compares LLM tuning cost-sensitive parameters against traditional scanning approach in both accuracy and computational cost, spanning 2,947 single-round (initial guess) and 1,931 multi-round (adjustment by trial-and-error) tasks across 13 simulators from fluid dynamics, solid mechanics, and plasma physics. Each simulator's cost is analytically defined and platform-independent. Frontier LLMs achieve 46-65% success rates in single-round mode, dropping to 35-55% under high accuracy requirements, rendering their initial guesses unreliable especially for high accuracy tasks. Multi-round mode improves rates to 72-81%, but LLMs are 1.5-2.5x slower than traditional scanning, making them uneconomical choices. We also investigate parameter group correlations for knowledge transfer potential, and the impact of in-context examples and reasoning effort, providing practical implications for deployment and fine-tuning. We open-source SimulCost as a static benchmark and extensible toolkit to facilitate research on improving cost-aware agentic designs for physics simulations, and for expanding new simulation environments. Code and data are available at https://github.com/Rose-STL-Lab/SimulCost-Bench. --- ### FinReflectKG -- HalluBench: GraphRAG Hallucination Benchmark for Financial Question Answering Systems Authors: Mahesh Kumar, Bhaskarjit Sarmah, Stefano Pasquali Published: 2026-03-11 Categories: cs.CL, q-fin.CP arXiv: https://arxiv.org/abs/2603.20252v1 As organizations increasingly integrate AI-powered question-answering systems into financial information systems for compliance, risk assessment, and decision support, ensuring the factual accuracy of AI-generated outputs becomes a critical engineering challenge. Current Knowledge Graph (KG)-augmented QA systems lack systematic mechanisms to detect hallucinations - factually incorrect outputs that undermine reliability and user trust. We introduce FinBench-QA-Hallucination, a benchmark for evaluating hallucination detection methods in KG-augmented financial QA over SEC 10-K filings. The dataset contains 755 annotated examples from 300 pages, each labeled for groundedness using a conservative evidence-linkage protocol requiring support from both textual chunks and extracted relational triplets. We evaluate six detection approaches - LLM judges, fine-tuned classifiers, Natural Language Inference (NLI) models, span detectors, and embedding-based methods under two conditions: with and without KG triplets. Results show that LLM-based judges and embedding approaches achieve the highest performance (F1: 0.82-0.86) under clean conditions. However, most methods degrade significantly when noisy triplets are introduced, with Matthews Correlation Coefficient (MCC) dropping 44-84 percent, while embedding methods remain relatively robust with only 9 percent degradation. Statistical tests (Cochran's Q and McNemar) confirm significant performance differences (p < 0.001). Our findings highlight vulnerabilities in current KG-augmented systems and provide insights for building reliable financial information systems, where hallucinations can lead to regulatory violations and flawed decisions. The benchmark also offers a framework for integrating AI reliability evaluation into information system design across other high-stakes domains such as healthcare, legal, and government. --- ### Parameter-Efficient Token Embedding Editing for Clinical Class-Level Unlearning Authors: Iyad Ait Hou, Shrenik Borad, Harsh Sharma, Pooja Srinivasan, Rebecca Hwa, Aya Zirikly Published: 2026-03-11 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.19302v1 Machine unlearning is increasingly important for clinical language models, where privacy regulations and institutional policies may require removing sensitive information from deployed systems without retraining from scratch. In practice, deletion requests must balance effective forgetting of targeted information with preservation of model utility and minimal parameter modification. We introduce Sparse Token Embedding Unlearning (STEU), a parameter-efficient method for behavioral class-level unlearning that updates only PMI-selected token embeddings together with a small classifier head while keeping all encoder layers frozen. Across experiments on MIMIC-IV, MIMIC-III, and eICU using BioClinicalBERT, BERT-base, and DistilBERT, STEU consistently suppresses the target class while largely preserving retained task performance. In the primary MIMIC-IV setting, STEU achieves near-complete forgetting (forget F1 = 0.0004) while maintaining competitive retained utility (retain avg F1 = 0.4766) after modifying only 0.19\% of model parameters. These results suggest that targeted behavioral unlearning can be achieved through sparse embedding edits without modifying deeper encoder representations. --- ### PRIME-CVD: A Parametrically Rendered Informatics Medical Environment for Education in Cardiovascular Risk Modelling Authors: Nicholas I-Hsien Kuo, Marzia Hoque Tania, Blanca Gallego, Louisa Jorm Published: 2026-03-11 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.19299v1 In recent years, progress in medical informatics and machine learning has been accelerated by the availability of openly accessible benchmark datasets. However, patient-level electronic medical record (EMR) data are rarely available for teaching or methodological development due to privacy, governance, and re-identification risks. This has limited reproducibility, transparency, and hands-on training in cardiovascular risk modelling. Here we introduce PRIME-CVD, a parametrically rendered informatics medical environment designed explicitly for medical education. PRIME-CVD comprises two openly accessible synthetic data assets representing a cohort of 50,000 adults undergoing primary prevention for cardiovascular disease. The datasets are generated entirely from a user-specified causal directed acyclic graph parameterised using publicly available Australian population statistics and published epidemiologic effect estimates, rather than from patient-level EMR data or trained generative models. Data Asset 1 provides a clean, analysis-ready cohort suitable for exploratory analysis, stratification, and survival modelling, while Data Asset 2 restructures the same cohort into a relational, EMR-style database with realistic structural and lexical heterogeneity. Together, these assets enable instruction in data cleaning, harmonisation, causal reasoning, and policy-relevant risk modelling without exposing sensitive information. Because all individuals and events are generated de novo, PRIME-CVD preserves realistic subgroup imbalance and risk gradients while ensuring negligible disclosure risk. PRIME-CVD is released under a Creative Commons Attribution 4.0 licence to support reproducible research and scalable medical education. --- ### A Dynamic Bayesian and Machine Learning Framework for Quantitative Evaluation and Prediction of Operator Situation Awareness in Nuclear Power Plants Authors: Shuai Chen, Huiqiao Jia, Tao Qing, Li Zhang, Xingyu Xiao Published: 2026-03-11 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.19298v2 Operator situation awareness is a pivotal yet elusive determinant of human reliability in complex nuclear control environments. Existing assessment methods, such as SAGAT and SART, remain static, retrospective, and detached from the evolving cognitive dynamics that drive operational risk. To overcome these limitations, this study introduces the dynamic Bayesian machine learning framework for situation awareness (DBML SA), a unified approach that fuses probabilistic reasoning and data driven intelligence to achieve quantitative, interpretable, and predictive situation awareness modeling. Leveraging 212 operational event reports (2007 to 2021), the framework reconstructs the causal temporal structure of 11 performance shaping factors across multiple cognitive layers. The Bayesian component enables time evolving inference of situation awareness reliability under uncertainty, while the neural component establishes a nonlinear predictive mapping from PSFs to SART scores, achieving a mean absolute percentage error of 13.8 % with statistical consistency to subjective evaluations (p > 0.05). Results highlight training quality and stress dynamics as primary drivers of situation awareness degradation. Overall, DBML SA transcends traditional questionnaire-based assessments by enabling real-time cognitive monitoring, sensitivity analysis, and early-warning prediction, paving the way toward intelligent human machine reliability management in next-generation digital main control rooms. --- ### CLaRE-ty Amid Chaos: Quantifying Representational Entanglement to Predict Ripple Effects in LLM Editing Authors: Manit Baser, Alperen Yildiz, Dinil Mon Divakaran, Mohan Gurusamy Published: 2026-03-11 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.19297v2 The static knowledge representations of large language models (LLMs) inevitably become outdated or incorrect over time. While model-editing techniques offer a promising solution by modifying a model's factual associations, they often produce unpredictable ripple effects, which are unintended behavioral changes that propagate even to the hidden space. In this work, we introduce CLaRE, a lightweight representation-level technique to identify where these ripple effects may occur. Unlike prior gradient-based methods, CLaRE quantifies entanglement between facts using forward activations from a single intermediate layer, avoiding costly backward passes. To enable systematic study, we prepare and analyse a corpus of 11,427 facts drawn from three existing datasets. Using CLaRE, we compute large-scale entanglement graphs of this corpus for multiple models, capturing how local edits propagate through representational space. These graphs enable stronger preservation sets for model editing, audit trails, efficient red-teaming, and scalable post-edit evaluation. In comparison to baselines, CLaRE achieves an average of 62.2% improvement in Spearman correlation with ripple effects while being $2.74\times$ faster, and using $2.85\times$ less peak GPU memory. Besides, CLaRE requires only a fraction of the storage needed by the baselines to compute and preserve fact representations. Our entanglement graphs and corpus are available at https://github.com/manitbaser/CLaRE. --- ### TTQ: Activation-Aware Test-Time Quantization to Accelerate LLM Inference On The Fly Authors: Toshiaki Koike-Akino, Jing Liu, Ye Wang Published: 2026-03-11 Categories: cs.LG, cs.CL, eess.SP arXiv: https://arxiv.org/abs/2603.19296v1 To tackle the huge computational demand of large foundation models, activation-aware compression techniques without retraining have been introduced. However, since these methods highly rely on calibration data, domain shift issues may arise for unseen downstream tasks. We propose a test-time quantization (TTQ) framework which compresses large models on the fly at inference time to resolve this issue. With an efficient online calibration, instant activation-aware quantization can adapt every prompt regardless of the downstream tasks, yet achieving inference speedup. Several experiments demonstrate that TTQ can improve the quantization performance over state-of-the-art baselines. --- ### BrainSCL: Subtype-Guided Contrastive Learning for Brain Disorder Diagnosis Authors: Xiaolong Li, Guiliang Guo, Guangqi Wen, Peng Cao, Jinzhu Yang, Honglin Wu, Xiaoli Liu, Fei Wang, Osmar R. Zaiane Published: 2026-03-11 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.19295v1 Mental disorder populations exhibit pronounced heterogeneity -- that is, the significant differences between samples -- poses a significant challenge to the definition of positive pairs in contrastive learning. To address this, we propose a subtype-guided contrastive learning framework that models patient heterogeneity as latent subtypes and incorporates them as structural priors to guide discriminative representation learning. Specifically, we construct multi-view representations by combining patients' clinical text with graph structure adaptively learned from BOLD signals, to uncover latent subtypes via unsupervised spectral clustering. A dual-level attention mechanism is proposed to construct prototypes for capturing stable subtype-specific connectivity patterns. We further propose a subtype-guided contrastive learning strategy that pulls samples toward their subtype prototype graph, reinforcing intra-subtype consistency for providing effective supervisory signals to improve model performance. We evaluate our method on Major Depressive Disorder (MDD), Bipolar Disorder (BD), and Autism Spectrum Disorders (ASD). Experimental results confirm the effectiveness of subtype prototype graphs in guiding contrastive learning and demonstrate that the proposed approach outperforms state-of-the-art approaches. Our code is available at https://anonymous.4open.science/r/BrainSCL-06D7. --- ### Taming Epilepsy: Mean Field Control of Whole-Brain Dynamics Authors: Ming Li, Ting Gao, Jingqiao Dua Published: 2026-03-11 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.18035v1 Controlling the high-dimensional neural dynamics during epileptic seizures remains a significant challenge due to the nonlinear characteristics and complex connectivity of the brain. In this paper, we propose a novel framework, namely Graph-Regularized Koopman Mean-Field Game (GK-MFG), which integrates Reservoir Computing (RC) for Koopman operator approximation with Alternating Population and Agent Control Network (APAC-Net) for solving distributional control problems. By embedding Electroencephalogram (EEG) dynamics into a linear latent space and imposing graph Laplacian constraints derived from the Phase Locking Value (PLV), our method achieves robust seizure suppression while respecting the functional topological structure of the brain. --- ### SimulU: Training-free Policy for Long-form Simultaneous Speech-to-Speech Translation Authors: Amirbek Djanibekov, Luisa Bentivogli, Matteo Negri, Sara Papi Published: 2026-03-11 Categories: eess.AS, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2603.16924v1 Simultaneous speech-to-speech translation (SimulS2S) is essential for real-time multilingual communication, with increasing integration into meeting and streaming platforms. Despite this, SimulS2S remains underexplored in research, where current solutions often rely on resource-intensive training procedures and operate on short-form, pre-segmented utterances, failing to generalize to continuous speech. To bridge this gap, we propose SimulU, the first training-free policy for long-form SimulS2S. SimulU adopts history management and speech output selection strategies that exploit cross-attention in pre-trained end-to-end models to regulate both input history and output generation. Evaluations on MuST-C across 8 languages show that SimulU achieves a better or comparable quality-latency trade-off against strong cascaded models. By eliminating the need for ad-hoc training, SimulU offers a promising path to end-to-end SimulS2S in realistic, long-form scenarios. --- ### QV May Be Enough: Toward the Essence of Attention in LLMs Authors: Zhang Edward Published: 2026-03-11 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.15665v1 Starting from first principles and a linguistic perspective centered on part-of-speech (POS) and syntactic analysis, this paper explores and derives the underlying essence of the Query-Key-Value (QKV) mechanism within the Transformer architecture. Based on this theoretical foundation, we provide a unified explanatory framework for the efficacy of contemporary architectures, including MQA, GQA, and MLA, while identifying their inherent trade-offs and potential optimization trajectories. We introduce the QV paradigm and provide empirical evidence for its validity. Building upon this, we propose the QV-Ka optimization scheme, which is further substantiated through experimental validation. The interpretable theoretical analysis of the QKV mechanism presented in this work establishes a robust foundation for the future evolution of large language model architectures. --- ### Hybrid Intent-Aware Personalization with Machine Learning and RAG-Enabled Large Language Models for Financial Services Marketing Authors: Akhil Chandra Shanivendra Published: 2026-03-11 Categories: cs.LG, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2603.14173v1 Personalized marketing in financial services requires models that can both predict customer behavior and generate compliant, context-appropriate content. This paper presents a hybrid architecture that integrates classical machine learning for segmentation, latent intent modeling, and personalization prediction with retrieval-augmented large language models for grounded content generation. A synthetic, reproducible dataset is constructed to reflect temporal customer behavior, product interactions, and marketing responses. The proposed framework incorporates temporal encoders, latent representations, and multi-task classification to estimate segment membership, customer intent, and product-channel recommendations. A retrieval-augmented generation layer then produces customer-facing messages constrained by retrieved domain documents. Experiments show that temporal modeling and intent features improve personalization accuracy, while citation-based retrieval reduces unsupported generation and supports auditability in regulated settings. The contribution is primarily architectural, demonstrating how predictive modeling and RAG-based generation can be combined into a transparent, explainable pipeline for financial services personalization. --- ### Citation-Enforced RAG for Fiscal Document Intelligence: Cited, Explainable Knowledge Retrieval in Tax Compliance Authors: Akhil Chandra Shanivendra Published: 2026-03-11 Categories: cs.IR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2603.14170v1 Tax authorities and public-sector financial agencies rely on large volumes of unstructured and semi-structured fiscal documents - including tax forms, instructions, publications, and jurisdiction-specific guidance - to support compliance analysis and audit workflows. While recent advances in generative AI and retrieval-augmented generation (RAG) have shown promise for document-centric question answering, existing approaches often lack the transparency, citation fidelity, and conservative behaviour required in high-stakes regulatory domains. This paper presents a multimodal, citation-enforced RAG framework for fiscal document intelligence that prioritises explainability and auditability. The framework adopts a source-first ingestion strategy, preserves page-level provenance, enforces citations during generation, and supports abstention when evidence is insufficient. Evaluation on real IRS and state tax documents demonstrates improved citation fidelity, reduced hallucination, and analyst-usable explanations, illustrating a pathway toward trustworthy AI for tax compliance. --- ### MAD: Microenvironment-Aware Distillation -- A Pretraining Strategy for Virtual Spatial Omics from Microscopy Authors: Jiashu Han, Kunzan Liu, Yeojin Kim, Saurabh Sinha, Sixian You Published: 2026-03-11 Categories: cs.CV, cs.AI, physics.optics arXiv: https://arxiv.org/abs/2603.13401v1 Bridging microscopy and omics would allow us to read molecular states from images-at single-cell resolution and tissue scale-without the cost and throughput limits of omics technologies. Self-supervised pretraining offers a scalable approach with minimal labels, yet how to encode single-cell identity within tissue environments-and the extent of biological information such models can capture-remains an open question. Here, we introduce MAD (microenvironment-aware distillation), a pretraining strategy that learns cell-centric embeddings by jointly self-distilling the morphology view and the microenvironment view of the same indexed cell into a unified embedding space. Across diverse tissues and imaging modalities, MAD achieves state-of-the-art prediction performance on downstream tasks including cell subtyping, transcriptomic prediction, and bioinformatic inference. MAD even outperforms foundation models with a similar number of model parameters that have been trained on substantially larger datasets. These results demonstrate that MAD's dual-view joint self-distillation effectively captures the complexity and diversity of cells within tissues. Together, this establishes MAD as a general tool for representation learning in microscopy, enabling virtual spatial omics and biological insights from vast microscopy datasets. ---