# Research Papers — 2026-01 · bundle page 72 of 189 > Full-text markdown of papers #3551–3600 of 9403 published in 2026-01, newest first. AI-readable mirrors of the full 9403-paper 2026-01 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-01/page/72 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/71.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/73.md) ### From Generation to Collaboration: Using LLMs to Edit for Empathy in Healthcare Authors: Man Luo, Bahareh Harandizadeh, Amara Tariq, Halim Abbas, Umar Ghaffar, Christopher J Warren, Segun O. Kolade, Haidar M. Abdul-Muhsin Published: 2026-01-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.15558v1 Clinical empathy is essential for patient care, but physicians need continually balance emotional warmth with factual precision under the cognitive and emotional constraints of clinical practice. This study investigates how large language models (LLMs) can function as empathy editors, refining physicians' written responses to enhance empathetic tone while preserving underlying medical information. More importantly, we introduce novel quantitative metrics, an Empathy Ranking Score and a MedFactChecking Score to systematically assess both emotional and factual quality of the responses. Experimental results show that LLM edited responses significantly increase perceived empathy while preserving factual accuracy compared with fully LLM generated outputs. These findings suggest that using LLMs as editorial assistants, rather than autonomous generators, offers a safer, more effective pathway to empathetic and trustworthy AI-assisted healthcare communication. --- ### LLM or Human? Perceptions of Trust and Information Quality in Research Summaries Authors: Nil-Jana Akpinar, Sandeep Avula, CJ Lee, Brandon Dang, Kaza Razat, Vanessa Murdock Published: 2026-01-22 Categories: cs.CY, cs.CL arXiv: https://arxiv.org/abs/2601.15556v1 Large Language Models (LLMs) are increasingly used to generate and edit scientific abstracts, yet their integration into academic writing raises questions about trust, quality, and disclosure. Despite growing adoption, little is known about how readers perceive LLM-generated summaries and how these perceptions influence evaluations of scientific work. This paper presents a mixed-methods survey experiment investigating whether readers with ML expertise can distinguish between human- and LLM-generated abstracts, how actual and perceived LLM involvement affects judgments of quality and trustworthiness, and what orientations readers adopt toward AI-assisted writing. Our findings show that participants struggle to reliably identify LLM-generated content, yet their beliefs about LLM involvement significantly shape their evaluations. Notably, abstracts edited by LLMs are rated more favorably than those written solely by humans or LLMs. We also identify three distinct reader orientations toward LLM-assisted writing, offering insights into evolving norms and informing policy around disclosure and acceptable use in scientific communication. --- ### BanditLP: Large-Scale Stochastic Optimization for Personalized Recommendations Authors: Phuc Nguyen, Benjamin Zelditch, Joyce Chen, Rohit Patra, Changshuai Wei Published: 2026-01-22 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2601.15552v1 We present BanditLP, a scalable multi-stakeholder contextual bandit framework that unifies neural Thompson Sampling for learning objective-specific outcomes with a large-scale linear program for constrained action selection at serving time. The methodology is application-agnostic, compatible with arbitrary neural architectures, and deployable at web scale, with an LP solver capable of handling billions of variables. Experiments on public benchmarks and synthetic data show consistent gains over strong baselines. We apply this approach in LinkedIn's email marketing system and demonstrate business win, illustrating the value of integrated exploration and constrained optimization in production. --- ### ALIGNAgent: Adaptive Learner Intelligence for Gap Identification and Next-step guidance Authors: Bismack Tokoli, Luis Jaimes, Ayesha S. Dina Published: 2026-01-22 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2601.15551v1 Personalized learning systems have emerged as a promising approach to enhance student outcomes by tailoring educational content, pacing, and feedback to individual needs. However, most existing systems remain fragmented, specializing in either knowledge tracing, diagnostic modeling, or resource recommendation, but rarely integrating these components into a cohesive adaptive cycle. In this paper, we propose ALIGNAgent (Adaptive Learner Intelligence for Gap Identification and Next-step guidance), a multi-agent educational framework designed to deliver personalized learning through integrated knowledge estimation, skill-gap identification, and targeted resource recommendation.ALIGNAgent begins by processing student quiz performance, gradebook data, and learner preferences to generate topic-level proficiency estimates using a Skill Gap Agent that employs concept-level diagnostic reasoning to identify specific misconceptions and knowledge deficiencies. After identifying skill gaps, the Recommender Agent retrieves preference-aware learning materials aligned with diagnosed deficiencies, implementing a continuous feedback loop where interventions occur before advancing to subsequent topics. Extensive empirical evaluation on authentic datasets from two undergraduate computer science courses demonstrates ALIGNAgent's effectiveness, with GPT-4o-based agents achieving precision of 0.87-0.90 and F1 scores of 0.84-0.87 in knowledge proficiency estimation validated against actual exam performance. --- ### Common to Whom? Regional Cultural Commonsense and LLM Bias in India Authors: Sangmitra Madhusudan, Trush Shashank More, Steph Buongiorno, Renata Dividino, Jad Kabbara, Ali Emami Published: 2026-01-22 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.15550v3 Existing cultural commonsense benchmarks treat nations as monolithic, assuming uniform practices within national boundaries. But does cultural commonsense hold uniformly within a nation, or does it vary at the sub-national level? We introduce Indica, the first benchmark designed to test LLMs' ability to address this question, focusing on India - a nation of 28 states, 8 union territories, and 22 official languages. We collect human-annotated answers from five Indian regions (North, South, East, West, and Central) across 515 questions spanning 8 domains of everyday life, yielding 1,630 region-specific question-answer pairs. Strikingly, only 39.4% of questions elicit agreement across all five regions, demonstrating that cultural commonsense in India is predominantly regional, not national. We evaluate eight state-of-the-art LLMs and find two critical gaps: models achieve only 13.4%-20.9% accuracy on region-specific questions, and they exhibit geographic bias, over-selecting Central and North India as the "default" (selected 30-40% more often than expected) while under-representing East and West. Beyond India, our methodology provides a generalizable framework for evaluating cultural commonsense in any culturally heterogeneous nation, from question design grounded in anthropological taxonomy, to regional data collection, to bias measurement. --- ### VIOLA: Towards Video In-Context Learning with Minimal Annotations Authors: Ryo Fujii, Hideo Saito, Ryo Hachiuma Published: 2026-01-22 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.15549v1 Generalizing Multimodal Large Language Models (MLLMs) to novel video domains is essential for real-world deployment but remains challenging due to the scarcity of labeled data. While In-Context Learning (ICL) offers a training-free adaptation path, standard methods rely on large annotated pools, which are often impractical in specialized environments like industrial or surgical settings since they require the experts' annotations. To bridge this gap, we introduce VIOLA (Video In-cOntext Learning with minimal Annotation), a label-efficient framework that synergizes minimal expert supervision with abundant unlabeled data. First, to maximize the efficiency of a strict annotation budget, we propose density-uncertainty-weighted sampling. Unlike standard diversity or uncertainty strategies that risk selecting visual outliers, our method leverages density estimation to identify samples that are simultaneously diverse, representative, and informative. Second, to utilize the remaining unlabeled data without noise propagation, we construct a hybrid pool and introduce confidence-aware retrieval and confidence-aware prompting. These mechanisms explicitly model label reliability, retrieving demonstrations based on a composite score of similarity and confidence while enabling the MLLM to adaptively distinguish between verified ground truths and noisy pseudo-labels. Extensive experiments across nine diverse benchmarks using four MLLMs demonstrate that our framework significantly outperforms various baselines in low-resource settings, achieving robust adaptation with minimal annotation costs. --- ### Learning Neural Operators from Partial Observations via Latent Autoregressive Modeling Authors: Jingren Hou, Hong Wang, Pengyu Xu, Chang Gao, Huafeng Liu, Liping Jing Published: 2026-01-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.15547v2 Real-world scientific applications frequently encounter incomplete observational data due to sensor limitations, geographic constraints, or measurement costs. Although neural operators significantly advanced PDE solving in terms of computational efficiency and accuracy, their underlying assumption of fully-observed spatial inputs severely restricts applicability in real-world applications. We introduce the first systematic framework for learning neural operators from partial observation. We identify and formalize two fundamental obstacles: (i) the supervision gap in unobserved regions that prevents effective learning of physical correlations, and (ii) the dynamic spatial mismatch between incomplete inputs and complete solution fields. Specifically, our proposed Latent Autoregressive Neural Operator(LANO) introduces two novel components designed explicitly to address the core difficulties of partial observations: (i) a mask-to-predict training strategy that creates artificial supervision by strategically masking observed regions, and (ii) a Physics-Aware Latent Propagator that reconstructs solutions through boundary-first autoregressive generation in latent space. Additionally, we develop POBench-PDE, a dedicated and comprehensive benchmark designed specifically for evaluating neural operators under partial observation conditions across three PDE-governed tasks. LANO achieves state-of-the-art performance with 18--69$\%$ relative L2 error reduction across all benchmarks under patch-wise missingness with less than 50$\%$ missing rate, including real-world climate prediction. Our approach effectively addresses practical scenarios involving up to 75$\%$ missing rate, to some extent bridging the existing gap between idealized research settings and the complexities of real-world scientific computing. --- ### Beyond validation loss: Clinically-tailored optimization metrics improve a model's clinical performance Authors: Charles B. Delahunt, Courosh Mehanian, Daniel E. Shea, Matthew P. Horning Published: 2026-01-22 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.15546v1 A key task in ML is to optimize models at various stages, e.g. by choosing hyperparameters or picking a stopping point. A traditional ML approach is to use validation loss, i.e. to apply the training loss function on a validation set to guide these optimizations. However, ML for healthcare has a distinct goal from traditional ML: Models must perform well relative to specific clinical requirements, vs. relative to the loss function used for training. These clinical requirements can be captured more precisely by tailored metrics. Since many optimization tasks do not require the driving metric to be differentiable, they allow a wider range of options, including the use of metrics tailored to be clinically-relevant. In this paper we describe two controlled experiments which show how the use of clinically-tailored metrics provide superior model optimization compared to validation loss, in the sense of better performance on the clinical task. The use of clinically-relevant metrics for optimization entails some extra effort, to define the metrics and to code them into the pipeline. But it can yield models that better meet the central goal of ML for healthcare: strong performance in the clinic. --- ### RDumb++: Drift-Aware Continual Test-Time Adaptation Authors: Himanshu Mishra Published: 2026-01-22 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.15544v1 Continual Test-Time Adaptation (CTTA) seeks to update a pretrained model during deployment using only the incoming, unlabeled data stream. Although prior approaches such as Tent, EATA etc. provide meaningful improvements under short evolving shifts, they struggle when the test distribution changes rapidly or over extremely long horizons. This challenge is exemplified by the CCC benchmark, where models operate over streams of 7.5M samples with continually changing corruption types and severities. We propose RDumb++, a principled extension of RDumb that introduces two drift-detection mechanisms i.e entropy-based drift scoring and KL-divergence drift scoring, together with adaptive reset strategies. These mechanisms allow the model to detect when accumulated adaptation becomes harmful and to recover before prediction collapse occurs. Across CCC-medium with three speeds and three seeds (nine runs, each containing one million samples), RDumb++ consistently surpasses RDumb, yielding approx 3% absolute accuracy gains while maintaining stable adaptation throughout the entire stream. Ablation experiments on drift thresholds and reset strengths further show that drift-aware resetting is essential for preventing collapse and achieving reliable long-horizon CTTA. --- ### XInsight: Integrative Stage-Consistent Psychological Counseling Support Agents for Digital Well-Being Authors: Fei Wang, Jiangnan Yang, Junjie Chen, Yuxin Liu, Kun Li, Yanyan Wei, Dan Guo, Meng Wang Published: 2026-01-21 Categories: cs.HC, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2603.06583v1 Web-based platforms are becoming a primary channel for psychological support, yet most LLM-driven chatbots remain opaque, single-stage, and weakly grounded in established therapeutic practice, limiting their usefulness for web applications that promote digital well-being. To address this gap, we present \textbf{XInsight}, a counseling-inspired multi-agent framework that models psychological support as a stage-consistent workflow aligned with the classical \textit{Exploration-Insight-Action} paradigm. Building on structured client representations, XInsight orchestrates specialized agents under a unified \textit{Reason-Intervene-Reflect} cycle: an Exploration agent organizes background and concerns into a structured Case Conceptualization Form, a Routing agent performs Adaptive Therapeutic Routing (ATR) across SFBT, CBT, and MBCT, a unified Therapeutic agent executes school-consistent submodules, and a Consolidation agent guides review, skill integration, and relapse-prevention planning. A Recording agent continuously transforms open-ended web dialogues into standardized psychological artifacts, including case formulations, therapeutic records, and relapse-prevention plans, enhancing interpretability, continuity, and accountability. To support rigorous and transparent assessment, we introduce \textbf{XInsight-Bench} with a Scale-Guided LLM Evaluation (SGLE) protocol that combines therapy-specific clinical scales with general counseling criteria. Experiments show improved paradigm alignment, multi-therapy integration, interaction depth, and interpretability over existing multi-agent counseling systems, indicating that XInsight provides a practical blueprint for integrating counseling-inspired support agents into web applications for digital well-being. --- ### From Toil to Thought: Designing for Strategic Exploration and Responsible AI in Systematic Literature Reviews Authors: Runlong Ye, Naaz Sibia, Angela Zavaleta Bernuy, Tingting Zhu, Carolina Nobre, Viktoria Pammer-Schindler, Michael Liut Published: 2026-01-21 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2603.05514v2 Systematic Literature Reviews (SLRs) are fundamental to scientific progress, yet the process is hindered by a fragmented tool ecosystem that imposes a high cognitive load. This friction suppresses the iterative, exploratory nature of scholarly work. To investigate these challenges, we conducted an exploratory design study with 20 experienced researchers. This study identified key friction points: 1) the high cognitive load of managing iterative query refinement across multiple databases, 2) the overwhelming scale and pace of publication of modern literature, and 3) the tension between automation and scholarly agency. Informed by these findings, we developed ARC, a design probe that operationalizes solutions for multi-database integration, transparent iterative search, and verifiable AI-assisted screening. A comparative user study with 8 researchers suggests that an integrated environment facilitates a transition in scholarly work, moving researchers from managing administrative overhead to engaging in strategic exploration. By utilizing external representations to scaffold strategic exploration and transparent AI reasoning, our system supports verifiable judgment, aiming to augment expert contributions from initial creation through long-term maintenance of knowledge synthesis. --- ### Lost Before Translation: Social Information Transmission and Survival in AI-AI Communication Authors: Bijean Ghafouri, Emilio Ferrara Published: 2026-01-21 Categories: cs.HC, cs.CL arXiv: https://arxiv.org/abs/2602.17674v1 When AI systems summarize and relay information, they inevitably transform it. But how? We introduce an experimental paradigm based on the telephone game to study what happens when AI talks to AI. Across five studies tracking content through AI transmission chains, we find three consistent patterns. The first is convergence, where texts differing in certainty, emotional intensity, and perspectival balance collapse toward a shared default of moderate confidence, muted affect, and analytical structure. The second is selective survival, where narrative anchors persist while the texture of evidence, hedges, quotes, and attributions is stripped away. The third is competitive filtering, where strong arguments survive while weaker but valid considerations disappear when multiple viewpoints coexist. In downstream experiments, human participants rated AI-transmitted content as more credible and polished. Importantly, however, humans also showed degraded factual recall, reduced perception of balance, and diminished emotional resonance. We show that the properties that make AI-mediated content appear authoritative may systematically erode the cognitive and affective diversity on which informed judgment depends. --- ### Language Model Representations for Efficient Few-Shot Tabular Classification Authors: Inwon Kang, Parikshit Ram, Yi Zhou, Horst Samulowitz, Oshani Seneviratne Published: 2026-01-21 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.15844v1 The Web is a rich source of structured data in the form of tables, from product catalogs and knowledge bases to scientific datasets. However, the heterogeneity of the structure and semantics of these tables makes it challenging to build a unified method that can effectively leverage the information they contain. Meanwhile, Large language models (LLMs) are becoming an increasingly integral component of web infrastructure for tasks like semantic search. This raises a crucial question: can we leverage these already-deployed LLMs to classify structured data in web-native tables (e.g., product catalogs, knowledge base exports, scientific data portals), avoiding the need for specialized models or extensive retraining? This work investigates a lightweight paradigm, $\textbf{Ta}$ble $\textbf{R}$epresentation with $\textbf{L}$anguage Model~($\textbf{TaRL}$), for few-shot tabular classification that directly utilizes semantic embeddings of individual table rows. We first show that naive application of these embeddings underperforms compared to specialized tabular models. We then demonstrate that their potentials can be unlocked with two key techniques: removing the common component from all embeddings and calibrating the softmax temperature. We show that a simple meta-learner, trained on handcrafted features, can learn to predict an appropriate temperature. This approach achieves performance comparable to state-of-the-art models in low-data regimes ($k \leq 32$) of semantically-rich tables. Our findings demonstrate the viability of reusing existing LLM infrastructure for efficient semantics-driven pathway to reuse existing LLM infrastructure for Web table understanding. --- ### The Perplexity Paradox: Why Code Compresses Better Than Math in LLM Prompts Authors: Warren Johnson Published: 2026-01-21 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.15843v1 In "Compress or Route?" (Johnson, 2026), we found that code generation tolerates aggressive prompt compression (r >= 0.6) while chain-of-thought reasoning degrades gradually. That study was limited to HumanEval (164 problems), left the "perplexity paradox" mechanism unvalidated, and provided no adaptive algorithm. This paper addresses all three gaps. First, we validate across six code benchmarks (HumanEval, MBPP, HumanEval+, MultiPL-E) and four reasoning benchmarks (GSM8K, MATH, ARC-Challenge, MMLU-STEM), confirming the compression threshold generalizes across languages and difficulties. Second, we conduct the first per-token perplexity analysis (n=723 tokens), revealing a "perplexity paradox": code syntax tokens are preserved (high perplexity) while numerical values in math problems are pruned despite being task-critical (low perplexity). Signature injection recovers +34 percentage points in pass rate (5.3% to 39.3%; Cohen's h=0.890). Third, we propose TAAC (Task-Aware Adaptive Compression), achieving 22% cost reduction with 96% quality preservation, outperforming fixed-ratio compression by 7%. MBPP validation (n=1,800 trials) confirms systematic variation: 3.6% at r=0.3 to 54.6% at r=1.0. --- ### Memes-as-Replies: Can Models Select Humorous Manga Panel Responses? Authors: Ryosuke Kohita, Seiichiro Yoshioka Published: 2026-01-21 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2602.15842v1 Memes are a popular element of modern web communication, used not only as static artifacts but also as interactive replies within conversations. While computational research has focused on analyzing the intrinsic properties of memes, the dynamic and contextual use of memes to create humor remains an understudied area of web science. To address this gap, we introduce the Meme Reply Selection task and present MaMe-Re (Manga Meme Reply Benchmark), a benchmark of 100,000 human-annotated pairs (500,000 total annotations from 2,325 unique annotators) consisting of openly licensed Japanese manga panels and social media posts. Our analysis reveals three key insights: (1) large language models (LLMs) show preliminary evidence of capturing complex social cues such as exaggeration, moving beyond surface-level semantic matching; (2) the inclusion of visual information does not improve performance, revealing a gap between understanding visual content and effectively using it for contextual humor; (3) while LLMs can match human judgments in controlled settings, they struggle to distinguish subtle differences in wit among semantically similar candidates. These findings suggest that selecting contextually humorous replies remains an open challenge for current models. --- ### Agentic AI for Commercial Insurance Underwriting with Adversarial Self-Critique Authors: Joyjit Roy, Samaresh Kumar Singh Published: 2026-01-21 Categories: cs.AI, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2602.13213v2 Commercial insurance underwriting is a labor-intensive process that requires manual review of extensive documentation to assess risk and determine policy pricing. While AI offers substantial efficiency improvements, existing solutions lack comprehensive reasoning and internal mechanisms to ensure reliability in regulated, high-stakes environments. Full automation remains impractical and inadvisable when human judgment and accountability are critical. This study presents a decision-negative, human-in-the-loop agentic system that incorporates an adversarial self-critique mechanism as a bounded safety architecture for regulated underwriting workflows. In this system, a critic agent challenges the primary agent's conclusions prior to submitting recommendations to human reviewers. This internal system of checks and balances addresses a critical gap in AI safety for regulated workflows. Additionally, the research develops a formal taxonomy of failure modes to characterize potential errors by decision-negative agents. This taxonomy provides a structured framework for risk identification and management in high-stakes applications. Experimental evaluation using 500 expert-validated underwriting cases demonstrates that the adversarial critique mechanism reduces AI hallucination rates from 11.3% to 3.8% and increases decision accuracy from 92% to 96%. At the same time, the framework enforces strict human authority over all binding decisions by design. These findings indicate that adversarial self-critique supports safer AI deployment in regulated domains and offers a model for responsible integration where human oversight is indispensable. --- ### Retrieval-Augmented Self-Taught Reasoning Model with Adaptive Chain-of-Thought for ASR Named Entity Correction Authors: Junjie An, Jingguang Tian, Tianyi Wang, Yu Gao, Xiaofeng Mou, Yi Xu Published: 2026-01-21 Categories: cs.CL, cs.AI, eess.AS arXiv: https://arxiv.org/abs/2602.12287v1 End-to-end automatic speech recognition (ASR) systems frequently misrecognize domain-specific phrases like named entities, which can cause catastrophic failures in downstream tasks. A new family of named entity correction methods based on large language models (LLMs) has recently emerged. However, these approaches have yet to fully exploit the sophisticated reasoning capabilities inherent to LLMs. To bridge this gap, we propose a novel retrieval-augmented generation framework for correcting named entity errors in ASR. Our approach consists of two key components: (1) a rephrasing language model (RLM) for named entity recognition, followed by candidate retrieval using a phonetic-level edit distance; and (2) a novel self-taught reasoning model with adaptive chain-of-thought (A-STAR) that dynamically adjusts the depth of its reasoning based on task difficulty. Experiments on the AISHELL-1 and Homophone datasets demonstrate the effectiveness of our method, which achieves relative reductions in the named entity character error rate of 17.96\% and 34.42\%, respectively, compared to a strong baseline. --- ### Mind the Gap No More: Achieving Zero-Gap Multimodal Integration via One Tokenizer Authors: Yanan Li, Christina Yi Jin, Yuan Jin, Manli Luo, Tie Xu, Shuai Jiao, Wei He, Qing Zhang Published: 2026-01-21 Categories: q-bio.GN, cs.CL arXiv: https://arxiv.org/abs/2602.12286v2 A central challenge in developing Multimodal Large Language Models (MLLMs) is effectively integrating heterogeneous inputs into a cohesive reasoning engine. Current paradigms predominantly rely on modular architectures that introduce modality-specific encoders and cross-modal fusion mechanisms. However, these designs are fundamentally bottlenecked by a geometric modality gap, forcing the LLM to expend significant computational capacity on geometric reconciliation rather than deep cross-modal reasoning. In this work, we formally characterize this modality gap and theoretically demonstrate that native architectures, specifically those employing a unified vocabulary, intrinsically maintain a zero-gap state across all hidden layers. Guided by these theoretical findings, we propose \textit{One Tokenizer}, a native architecture that maps all modalities directly into a shared token space. We empirically validate this framework on a DNA--text multimodal testbed. Our extensive evaluations reveal that by achieving seamless integration within the LLM's native latent space, One Tokenizer consistently outperforms encoder-based modular counterparts, providing a fundamentally superior framework for deep biological reasoning. --- ### From Biased Chatbots to Biased Agents: Examining Role Assignment Effects on LLM Agent Robustness Authors: Linbo Cao, Lihao Sun, Yang Yue Published: 2026-01-21 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.12285v1 Large Language Models (LLMs) are increasingly deployed as autonomous agents capable of actions with real-world impacts beyond text generation. While persona-induced biases in text generation are well documented, their effects on agent task performance remain largely unexplored, even though such effects pose more direct operational risks. In this work, we present the first systematic case study showing that demographic-based persona assignments can alter LLM agents' behavior and degrade performance across diverse domains. Evaluating widely deployed models on agentic benchmarks spanning strategic reasoning, planning, and technical operations, we uncover substantial performance variations - up to 26.2% degradation, driven by task-irrelevant persona cues. These shifts appear across task types and model architectures, indicating that persona conditioning and simple prompt injections can distort an agent's decision-making reliability. Our findings reveal an overlooked vulnerability in current LLM agentic systems: persona assignments can introduce implicit biases and increase behavioral volatility, raising concerns for the safe and robust deployment of LLM agents. --- ### A Lightweight LLM Framework for Disaster Humanitarian Information Classification Authors: Han Jinzhen, Kim Jisung, Yang Jong Soo, Yun Hong Sik Published: 2026-01-21 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.12284v1 Timely classification of humanitarian information from social media is critical for effective disaster response. However, deploying large language models (LLMs) for this task faces challenges in resource-constrained emergency settings. This paper develops a lightweight, cost-effective framework for disaster tweet classification using parameter-efficient fine-tuning. We construct a unified experimental corpus by integrating and normalizing the HumAID dataset (76,484 tweets across 19 disaster events) into a dual-task benchmark: humanitarian information categorization and event type identification. Through systematic evaluation of prompting strategies, LoRA fine-tuning, and retrieval-augmented generation (RAG) on Llama 3.1 8B, we demonstrate that: (1) LoRA achieves 79.62% humanitarian classification accuracy (+37.79% over zero-shot) while training only ~2% of parameters; (2) QLoRA enables efficient deployment with 99.4% of LoRA performance at 50% memory cost; (3) contrary to common assumptions, RAG strategies degrade fine-tuned model performance due to label noise from retrieved examples. These findings establish a practical, reproducible pipeline for building reliable crisis intelligence systems with limited computational resources. --- ### Code Mixologist : A Practitioner's Guide to Building Code-Mixed LLMs Authors: Himanshu Gupta, Pratik Jayarao, Chaitanya Dwivedi, Neeraj Varshney Published: 2026-01-21 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.11181v2 Code-mixing and code-switching (CSW) remain challenging phenomena for large language models (LLMs). Despite recent advances in multilingual modeling, LLMs often struggle in mixed-language settings, exhibiting systematic degradation in grammaticality, factuality, and safety behavior. This work provides a comprehensive overview of CSW research in modern large language model settings. We introduce a unifying taxonomy that organizes prior work along dimensions of data, modeling, and evaluation, and we distill these findings into a practical playbook of actionable recommendations for building, adapting, and evaluating CSW-capable LLMs. We review modeling approaches ranging from CSW-tailored pre-training and task-specific post-training to prompting strategies and in-context learning. We analyze current evaluation practices, highlighting sources of instability and limited reproducibility, and we catalog existing benchmarks while critically examining their linguistic coverage and English-centric biases. Finally, we discuss emerging safety concerns, including use of code-mixing as a mechanism for bypassing model safeguards, and identify open research challenges. --- ### Mechanistic Interpretability for Large Language Model Alignment: Progress, Challenges, and Future Directions Authors: Usman Naseem Published: 2026-01-21 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.11180v1 Large language models (LLMs) have achieved remarkable capabilities across diverse tasks, yet their internal decision-making processes remain largely opaque. Mechanistic interpretability (i.e., the systematic study of how neural networks implement algorithms through their learned representations and computational structures) has emerged as a critical research direction for understanding and aligning these models. This paper surveys recent progress in mechanistic interpretability techniques applied to LLM alignment, examining methods ranging from circuit discovery to feature visualization, activation steering, and causal intervention. We analyze how interpretability insights have informed alignment strategies including reinforcement learning from human feedback (RLHF), constitutional AI, and scalable oversight. Key challenges are identified, including the superposition hypothesis, polysemanticity of neurons, and the difficulty of interpreting emergent behaviors in large-scale models. We propose future research directions focusing on automated interpretability, cross-model generalization of circuits, and the development of interpretability-driven alignment techniques that can scale to frontier models. --- ### From Instruction to Output: The Role of Prompting in Modern NLG Authors: Munazza Zaib, Elaf Alhazmi Published: 2026-01-21 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.11179v1 Prompt engineering has emerged as an integral technique for extending the strengths and abilities of Large Language Models (LLMs) to gain significant performance gains in various Natural Language Processing (NLP) tasks. This approach, which requires instructions to be composed in natural language to bring out the knowledge from LLMs in a structured way, has driven breakthroughs in various NLP tasks. Yet there is still no structured framework or coherent understanding of the varied prompt engineering methods and techniques, particularly in the field of Natural Language Generation (NLG). This survey aims to help fill that gap by outlining recent developments in prompt engineering, and their effect on different NLG tasks. It reviews recent advances in prompting methods and their impact on NLG tasks, presenting prompt design as an input-level control mechanism that complements fine-tuning and decoding approaches. The paper introduces a taxonomy of prompting paradigms, a decision framework for prompt selection based on varying factors for the practitioners, outlines emerging trends and challenges, and proposes a framework that links design, optimization, and evaluation to support more controllable and generalizable NLG. --- ### Joint single-shot ToA and DoA estimation for VAA-based BLE ranging with phase ambiguity: A deep learning-based approach Authors: Jincheng Xie, Yili Deng, Jiguang He, Pengyu Wang, Miaomiao Dong, Rui Tang, Zhongyi Huang Published: 2026-01-21 Categories: eess.SP, cs.AI, cs.IT arXiv: https://arxiv.org/abs/2602.02503v1 Conventional direction-of-arrival (DoA) estimation methods rely on multi-antenna arrays, which are costly to implement on size-constrained Bluetooth Low Energy (BLE) devices. Virtual antenna array (VAA) techniques enable DoA estimation with a single antenna, making angle estimation feasible on such devices. However, BLE only provides a single-shot two-way channel frequency response (CFR) with a binary phase ambiguity issue, which hinders the direct application of VAA. To address this challenge, we propose a unified model that combines VAA with BLE two-way CFR, and introduce a neural network based phase recovery framework that employs row / column predictors with a voting mechanism to resolve the ambiguity. The recovered one-way CFR then enables super resolution algorithms such as MUSIC for joint time of arrival (ToA) and DoA estimation. Simulation results demonstrate that the proposed method achieves superior performance under non-uniform VAAs, with mean square errors approaching the Cramer Rao bound at SNR $\geq$ 5 dB. --- ### Standards for trustworthy AI in the European Union: technical rationale, structural challenges, and an implementation path Authors: Piercosma Bisconti, Marcello Galisai Published: 2026-01-21 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2602.00078v1 This white paper examines the technical foundations of European AI standardization under the AI Act. It explains how harmonized standards enable the presumption of conformity mechanism, describes the CEN/CENELEC standardization process, and analyzes why AI poses unique standardization challenges including stochastic behavior, data dependencies, immature evaluation practices, and lifecycle dynamics. The paper argues that AI systems are typically components within larger sociotechnical systems, requiring a layered approach where horizontal standards define process obligations and evidence structures while sectoral profiles specify domain-specific thresholds and acceptance criteria. It proposes a workable scheme based on risk management, reproducible technical checks redefined as stability of measured properties, structured documentation, comprehensive logging, and assurance cases that evolve over the system lifecycle. The paper demonstrates that despite methodological difficulties, technical standards remain essential for translating legal obligations into auditable engineering practice and enabling scalable conformity assessment across providers, assessors, and enforcement authorities --- ### Automated univariate time series forecasting with regression trees Authors: Francisco Martínez, María P. Frías Published: 2026-01-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00077v1 This paper describes a methodology for automated univariate time series forecasting using regression trees and their ensembles: bagging and random forests. The key aspects that are addressed are: the use of an autoregressive approach and recursive forecasts, how to select the autoregressive features, how to deal with trending series and how to cope with seasonal behavior. Experimental results show a forecast accuracy comparable with well-established statistical models such as exponential smoothing or ARIMA. Furthermore, a publicly available software implementing all the proposed strategies has been developed and is described in the paper. --- ### Repair Brain Damage: Real-Numbered Error Correction Code for Neural Network Authors: Ziqing Li, Myung Cho, Qiutong Jin, Weiyu Xu Published: 2026-01-21 Categories: cs.NE, cs.LG arXiv: https://arxiv.org/abs/2602.00076v1 We consider a neural network (NN) that may experience memory faults and computational errors. In this paper, we propose a novel real-number-based error correction code (ECC) capable of detecting and correcting both memory errors and computational errors. The proposed approach introduces structures in the form of real-number-based linear constraints on the NN weights to enable error detection and correction, without sacrificing classification performance or increasing the number of real-valued NN parameters. --- ### Dimensional Peeking for Low-Variance Gradients in Zeroth-Order Discrete Optimization via Simulation Authors: Philipp Andelfinger, Wentong Cai Published: 2026-01-21 Categories: cs.LG, cs.MS arXiv: https://arxiv.org/abs/2602.00075v1 Gradient-based optimization methods are commonly used to identify local optima in high-dimensional spaces. When derivatives cannot be evaluated directly, stochastic estimators can provide approximate gradients. However, these estimators' perturbation-based sampling of the objective function introduces variance that can lead to slow convergence. In this paper, we present dimensional peeking, a variance reduction method for gradient estimation in discrete optimization via simulation. By lifting the sampling granularity from scalar values to classes of values that follow the same control flow path, we increase the information gathered per simulation evaluation. Our derivation from an established smoothed gradient estimator shows that the method does not introduce any bias. We present an implementation via a custom numerical data type to transparently carry out dimensional peeking over C++ programs. Variance reductions by factors of up to 7.9 are observed for three simulation-based optimization problems with high-dimensional input. The optimization progress compared to three meta-heuristics shows that dimensional peeking increases the competitiveness of zeroth-order optimization for discrete and non-convex simulations. --- ### Adoption and Use of LLMs at an Academic Medical Center Authors: Nigam H. Shah, Nerissa Ambers, Abby Pandya, Timothy Keyes, Juan M. Banda, Srikar Nallan, Carlene Lugtu, Artem A. Trotsyuk, Suhana Bedi, Alyssa Unell, Miguel Fuentes, Francois Grolleau et al. Published: 2026-01-21 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2602.00074v2 While large language models (LLMs) can support clinical documentation needs, standalone tools struggle with "workflow friction" from manual data entry. We developed ChatEHR, a system that enables the use of LLMs with the entire patient timeline spanning several years. ChatEHR enables automations - which are static combinations of prompts and data that perform a fixed task - and interactive use in the electronic health record (EHR) via a user interface (UI). The resulting ability to sift through patient medical records for diverse use-cases such as pre-visit chart review, screening for transfer eligibility, monitoring for surgical site infections, and chart abstraction, redefines LLM use as an institutional capability. This system, accessible after user-training, enables continuous monitoring and evaluation of LLM use. In 1.5 years, we built 7 automations and 1075 users have trained to become routine users of the UI, engaging in 23,000 sessions in the first 3 months of launch. For automations, being model-agnostic and accessing multiple types of data was essential for matching specific clinical or administrative tasks with the most appropriate LLM. Benchmark-based evaluations proved insufficient for monitoring and evaluation of the UI, requiring new methods to monitor performance. Generation of summaries was the most frequent task in the UI, with an estimated 0.73 hallucinations and 1.60 inaccuracies per generation. The resulting mix of cost savings, time savings, and revenue growth required a value assessment framework to prioritize work as well as quantify the impact of using LLMs. Initial estimates are $6M savings in the first year of use, without quantifying the benefit of the better care offered. Such a "build-from-within" strategy provides an opportunity for health systems to maintain agency via a vendor-agnostic, internally governed LLM platform. --- ### Large Language Models: A Mathematical Formulation Authors: Ricardo Baptista, Andrew Stuart, Son Tran Published: 2026-01-21 Categories: math.NA, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2601.22170v2 Large language models (LLMs) process and predict sequences containing text to answer questions, and address tasks including document summarization, providing recommendations, writing software and solving quantitative problems. We provide a mathematical framework for LLMs by describing the encoding of text sequences into sequences of tokens, defining the architecture for next-token prediction models, explaining how these models are learned from data, and demonstrating how they are deployed to address a variety of tasks. The mathematical sophistication required to understand this material is not high, and relies on straightforward ideas from information theory, probability and optimization. Nonetheless, the combination of ideas resting on these different components from the mathematical sciences yields a complex algorithmic structure; and this algorithmic structure has demonstrated remarkable empirical successes. The mathematical framework established here provides a platform from which it is possible to formulate and address questions concerning the accuracy, efficiency and robustness of the algorithms that constitute LLMs. The framework also suggests directions for development of modified and new methodologies. --- ### Ego4OOD: Rethinking Egocentric Video Domain Generalization via Covariate Shift Scoring Authors: Zahra Vaseqi, James Clark Published: 2026-01-21 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.17056v1 Egocentric video action recognition under domain shifts remains challenging due to large intra-class spatio-temporal variability, long-tailed feature distributions, and strong correlations between actions and environments. Existing benchmarks for egocentric domain generalization often conflate covariate shifts with concept shifts, making it difficult to reliably evaluate a model's ability to generalize across input distributions. To address this limitation, we introduce Ego4OOD, a domain generalization benchmark derived from Ego4D that emphasizes measurable covariate diversity while reducing concept shift through semantically coherent, moment-level action categories. Ego4OOD spans eight geographically distinct domains and is accompanied by a clustering-based covariate shift metric that provides a quantitative proxy for domain difficulty. We further leverage a one-vs-all binary training objective that decomposes multi-class action recognition into independent binary classification tasks. This formulation is particularly well-suited for covariate shift by reducing interference between visually similar classes under feature distribution shift. Using this formulation, we show that a lightweight two-layer fully connected network achieves performance competitive with state-of-the-art egocentric domain generalization methods on both Argo1M and Ego4OOD, despite using fewer parameters and no additional modalities. Our empirical analysis demonstrates a clear relationship between measured covariate shift and recognition performance, highlighting the importance of controlled benchmarks and quantitative domain characterization for studying out-of-distribution generalization in egocentric video. --- ### Failing on Bias Mitigation: A Case Study on the Challenges of Fairness in Government Data Authors: Hongbo Bo, Jingyu Hu, Debbie Watson, Weiru Liu Published: 2026-01-21 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2601.17054v2 The potential for bias and unfairness in AI-supporting government services raises ethical and legal concerns. Using crime rate prediction with the Bristol City Council data as a case study, we examine how these issues persist. Rather than auditing real-world deployed systems, our goal is to understand why widely adopted bias mitigation techniques often fail when applied to government data. Our findings reveal that bias mitigation approaches applied to government data are not always effective -- not because of flaws in model architecture or metric selection, but due to the inherent properties of the data itself. Through comparing a set of comprehensive models and fairness methods, our experiments consistently show that the mitigation efforts cannot overcome the embedded unfairness in the data -- further reinforcing that the origin of bias lies in the structure and history of government datasets. We then explore the reasons for the mitigation failures in predictive models on government data and highlight the potential sources of unfairness posed by data distribution shifts, the accumulation of historical bias, and delays in data release. We also discover the limitations of the blind spots in fairness analysis and bias mitigation methods when only targeting a single sensitive feature through a set of intersectional fairness experiments. Although this study is limited to one city, the findings are highly suggestive, which can contribute to an early warning that biases in government data may persist even with standard mitigation methods. --- ### Synthetic Data Guided Feature Selection for Robust Activity Recognition in Older Adults Authors: Shuhao Que, Dieuwke van Dartel, Ilse Heeringa, Han Hegeman, Miriam Vollenbroek-Hutten, Ying Wang Published: 2026-01-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.17053v2 Physical activity during hip fracture rehabilitation is essential for mitigating long-term functional decline in geriatric patients. However, it is rarely quantified in clinical practice. Existing continuous monitoring systems with commercially available wearable activity trackers are typically developed in middle-aged adults and therefore perform unreliably in older adults with slower and more variable gait patterns. This study aimed to develop a robust human activity recognition (HAR) system to improve continuous physical activity recognition in the context of hip fracture rehabilitation. 24 healthy older adults aged over 80 years were included to perform activities of daily living (walking, standing, sitting, lying down, and postural transfers) under simulated free-living conditions for 75 minutes while wearing two accelerometers positioned on the lower back and anterior upper thigh. Model robustness was evaluated using leave-one-subject-out cross-validation. The synthetic data demonstrated potential to improve generalization across participants. The resulting feature intervention model (FIM), aided by synthetic data guidance, achieved reliable activity recognition with mean F1-scores of 0.896 for walking, 0.927 for standing, 0.997 for sitting, 0.937 for lying down, and 0.816 for postural transfers. Compared with a control condition model without synthetic data, the FIM significantly improved the postural transfer detection, i.e., an activity class of high clinical relevance that is often overlooked in existing HAR literature. In conclusion, these preliminary results demonstrate the feasibility of robust activity recognition in older adults. Further validation in hip fracture patient populations is required to assess the clinical utility of the proposed monitoring system. --- ### Single-Pixel Vision-Language Model for Intrinsic Privacy-Preserving Behavioral Intelligence Authors: Hongjun An, Yiliang Song, Jiawei Shao, Zhe Sun, Xuelong Li Published: 2026-01-21 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.17050v1 Adverse social interactions, such as bullying, harassment, and other illicit activities, pose significant threats to individual well-being and public safety, leaving profound impacts on physical and mental health. However, these critical events frequently occur in privacy-sensitive environments like restrooms, and changing rooms, where conventional surveillance is prohibited or severely restricted by stringent privacy regulations and ethical concerns. Here, we propose the Single-Pixel Vision-Language Model (SP-VLM), a novel framework that reimagines secure environmental monitoring. It achieves intrinsic privacy-by-design by capturing human dynamics through inherently low-dimensional single-pixel modalities and inferring complex behavioral patterns via seamless vision-language integration. Building on this framework, we demonstrate that single-pixel sensing intrinsically suppresses identity recoverability, rendering state-of-the-art face recognition systems ineffective below a critical sampling rate. We further show that SP-VLM can nonetheless extract meaningful behavioral semantics, enabling robust anomaly detection, people counting, and activity understanding from severely degraded single-pixel observations. Combining these findings, we identify a practical sampling-rate regime in which behavioral intelligence emerges while personal identity remains strongly protected. Together, these results point to a human-rights-aligned pathway for safety monitoring that can support timely intervention without normalizing intrusive surveillance in privacy-sensitive spaces. --- ### Summary of the Unusual Activity Recognition Challenge for Developmental Disability Support Authors: Christina Garcia, Nhat Tan Le, Taihei Fujioka, Umang Dobhal, Milyun Ni'ma Shoumi, Thanh Nha Nguyen, Sozo Inoue Published: 2026-01-21 Categories: cs.CV, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2601.17049v1 This paper presents an overview of the Recognize the Unseen: Unusual Behavior Recognition from Pose Data Challenge, hosted at ISAS 2025. The challenge aims to address the critical need for automated recognition of unusual behaviors in facilities for individuals with developmental disabilities using non-invasive pose estimation data. Participating teams were tasked with distinguishing between normal and unusual activities based on skeleton keypoints extracted from video recordings of simulated scenarios. The dataset reflects real-world imbalance and temporal irregularities in behavior, and the evaluation adopted a Leave-One-Subject-Out (LOSO) strategy to ensure subject-agnostic generalization. The challenge attracted broad participation from 40 teams applying diverse approaches ranging from classical machine learning to deep learning architectures. Submissions were assessed primarily using macro-averaged F1 scores to account for class imbalance. The results highlight the difficulty of modeling rare, abrupt actions in noisy, low-dimensional data, and emphasize the importance of capturing both temporal and contextual nuances in behavior modeling. Insights from this challenge may contribute to future developments in socially responsible AI applications for healthcare and behavior monitoring. --- ### SiMiC: Context-Aware Silicon Microstructure Characterization Using Attention-Based Convolutional Neural Networks for Field-Emission Tip Analysis Authors: Jing Jie Tan, Rupert Schreiner, Matthias Hausladen, Ali Asgharzade, Simon Edler, Julian Bartsch, Michael Bachmann, Andreas Schels, Ban-Hoe Kwan, Danny Wee-Kiat Ng, Yan-Chai Hum Published: 2026-01-21 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.17048v1 Accurate characterization of silicon microstructures is essential for advancing microscale fabrication, quality control, and device performance. Traditional analysis using Scanning Electron Microscopy (SEM) often requires labor-intensive, manual evaluation of feature geometry, limiting throughput and reproducibility. In this study, we propose SiMiC: Context-Aware Silicon Microstructure Characterization Using Attention-Based Convolutional Neural Networks for Field-Emission Tip Analysis. By leveraging deep learning, our approach efficiently extracts morphological features-such as size, shape, and apex curvature-from SEM images, significantly reducing human intervention while improving measurement consistency. A specialized dataset of silicon-based field-emitter tips was developed, and a customized CNN architecture incorporating attention mechanisms was trained for multi-class microstructure classification and dimensional prediction. Comparative analysis with classical image processing techniques demonstrates that SiMiC achieves high accuracy while maintaining interpretability. The proposed framework establishes a foundation for data-driven microstructure analysis directly linked to field-emission performance, opening avenues for correlating emitter geometry with emission behavior and guiding the design of optimized cold-cathode and SEM electron sources. The related dataset and algorithm repository that could serve as a baseline in this area can be found at https://research.jingjietan.com/?q=SIMIC --- ### A Contrastive Pre-trained Foundation Model for Deciphering Imaging Noisomics across Modalities Authors: Yuanjie Gu, Yiqun Wang, Chaohui Yu, Ang Xuan, Fan Wang, Zhi Lu, Biqin Dong Published: 2026-01-21 Categories: cs.CV, cs.LG, eess.IV arXiv: https://arxiv.org/abs/2601.17047v1 Characterizing imaging noise is notoriously data-intensive and device-dependent, as modern sensors entangle physical signals with complex algorithmic artifacts. Current paradigms struggle to disentangle these factors without massive supervised datasets, often reducing noise to mere interference rather than an information resource. Here, we introduce "Noisomics", a framework shifting the focus from suppression to systematic noise decoding via the Contrastive Pre-trained (CoP) Foundation Model. By leveraging the manifold hypothesis and synthetic noise genome, CoP employs contrastive learning to disentangle semantic signals from stochastic perturbations. Crucially, CoP breaks traditional deep learning scaling laws, achieving superior performance with only 100 training samples, outperforming supervised baselines trained on 100,000 samples, thereby reducing data and computational dependency by three orders of magnitude. Extensive benchmarking across 12 diverse out-of-domain datasets confirms its robust zero-shot generalization, demonstrating a 63.8% reduction in estimation error and an 85.1% improvement in the coefficient of determination compared to the conventional training strategy. We demonstrate CoP's utility across scales: from deciphering non-linear hardware-noise interplay in consumer photography to optimizing photon-efficient protocols for deep-tissue microscopy. By decoding noise as a multi-parametric footprint, our work redefines stochastic degradation as a vital information resource, empowering precise imaging diagnostics without prior device calibration. --- ### Automated Road Crack Localization for Spatially Guided Highway Maintenance Authors: Steffen Knoblauch, Ram Kumar Muthusamy, Pedram Ghamisi, Alexander Zipf Published: 2026-01-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.16737v3 Highway networks are crucial for economic prosperity. Climate change-induced temperature fluctuations are exacerbating stress on road pavements, resulting in elevated maintenance costs. This underscores the need for targeted and efficient maintenance strategies. This study investigates the potential of open-source data to guide highway infrastructure maintenance. The proposed framework integrates airborne imagery and OpenStreetMap (OSM) to fine-tune YOLOv11 for highway crack localization. To demonstrate the framework's real-world applicability, a Swiss Relative Highway Crack Density (RHCD) index was calculated to inform nationwide highway maintenance. The crack classification model achieved an F1-score of $0.84$ for the positive class (crack) and $0.97$ for the negative class (no crack). The Swiss RHCD index exhibited weak correlations with Long-term Land Surface Temperature Amplitudes (LT-LST-A) (Pearson's $r\ = -0.05$) and Traffic Volume (TV) (Pearson's $r\ = 0.17$), underlining the added value of this novel index for guiding maintenance over other data. Significantly high RHCD values were observed near urban centers and intersections, providing contextual validation for the predictions. These findings highlight the value of open-source data sharing to drive innovation, ultimately enabling more efficient solutions in the public sector. --- ### Test-Time Adaptation for Speech Emotion Recognition Authors: Jiaheng Dong, Hong Jia, Ting Dang Published: 2026-01-21 Categories: eess.AS, cs.LG, cs.SD arXiv: https://arxiv.org/abs/2601.16240v1 The practical utility of Speech Emotion Recognition (SER) systems is undermined by their fragility to domain shifts, such as speaker variability, the distinction between acted and naturalistic emotions, and cross-corpus variations. While domain adaptation and fine-tuning are widely studied, they require either source data or labelled target data, which are often unavailable or raise privacy concerns in SER. Test-time adaptation (TTA) bridges this gap by adapting models at inference using only unlabeled target data. Yet, having been predominantly designed for image classification and speech recognition, the efficacy of TTA for mitigating the unique domain shifts in SER has not been investigated. In this paper, we present the first systematic evaluation and comparison covering 11 TTA methods across three representative SER tasks. The results indicate that backpropagation-free TTA methods are the most promising. Conversely, entropy minimization and pseudo-labeling generally fail, as their core assumption of a single, confident ground-truth label is incompatible with the inherent ambiguity of emotional expression. Further, no single method universally excels, and its effectiveness is highly dependent on the distributional shifts and tasks. --- ### VibeTensor: System Software for Deep Learning, Fully Generated by AI Agents Authors: Bing Xu, Terry Chen, Fengzhe Zhou, Tianqi Chen, Yangqing Jia, Vinod Grover, Haicheng Wu, Wei Liu, Craig Wittenbrink, Wen-mei Hwu, Roger Bringmann, Ming-Yu Liu et al. Published: 2026-01-21 Categories: cs.SE, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.16238v1 VIBETENSOR is an open-source research system software stack for deep learning, generated by LLM-powered coding agents under high-level human guidance. In this paper, "fully generated" refers to code provenance: implementation changes were produced and applied as agent-proposed diffs; validation relied on agent-run builds, tests, and differential checks, without per-change manual diff review. It implements a PyTorch-style eager tensor library with a C++20 core (CPU+CUDA), a torch-like Python overlay via nanobind, and an experimental Node.js/TypeScript interface. Unlike thin bindings, VIBETENSOR includes its own tensor/storage system, schema-lite dispatcher, reverse-mode autograd, CUDA runtime (streams/events/graphs), a stream-ordered caching allocator with diagnostics, and a stable C ABI for dynamically loaded operator plugins. We view this release as a milestone for AI-assisted software engineering: it shows coding agents can generate a coherent deep learning runtime spanning language bindings down to CUDA memory management, validated primarily by builds and tests. We describe the architecture, summarize the workflow used to produce and validate the system, and evaluate the artifact. We report repository scale and test-suite composition, and summarize reproducible microbenchmarks from an accompanying AI-generated kernel suite, including fused attention versus PyTorch SDPA/FlashAttention. We also report end-to-end training sanity checks on 3 small workloads (sequence reversal, ViT, miniGPT) on NVIDIA H100 (Hopper, SM90) and Blackwell-class GPUs; multi-GPU results are Blackwell-only and use an optional CUTLASS-based ring-allreduce plugin gated on CUDA 13+ and sm103a toolchain support. Finally, we discuss failure modes in generated system software, including a "Frankenstein" composition effect where locally correct subsystems interact to yield globally suboptimal performance. --- ### Computational Foundations for Strategic Coopetition: Formalizing Collective Action and Loyalty Authors: Vik Pant, Eric Yu Published: 2026-01-21 Categories: cs.MA, cs.AI, cs.CY, cs.SE arXiv: https://arxiv.org/abs/2601.16237v1 Mixed-motive multi-agent settings are rife with persistent free-riding because individual effort benefits all members equally, yet each member bears the full cost of their own contribution. Classical work by Holmström established that under pure self-interest, Nash equilibrium is universal shirking. While i* represents teams as composite actors, it lacks scalable computational mechanisms for analyzing how collective action problems emerge and resolve in coopetitive settings. This technical report extends computational foundations for strategic coopetition to team-level dynamics, building on companion work formalizing interdependence/complementarity (arXiv:2510.18802) and trust dynamics (arXiv:2510.24909). We develop loyalty-moderated utility functions with two mechanisms: loyalty benefit (welfare internalization plus intrinsic contribution satisfaction) and cost tolerance (reduced effort burden for loyal members). We integrate i* structural dependencies through dependency-weighted team cohesion, connecting member incentives to team-level positioning. The framework applies to both human teams (loyalty as psychological identification) and multi-agent systems (alignment coefficients and adjusted cost functions). Experimental validation across 3,125 configurations demonstrates robust loyalty effects (15.04x median effort differentiation). All six behavioral targets achieve thresholds: free-riding baseline (96.5%), loyalty monotonicity (100%), effort differentiation (100%), team size effect (100%), mechanism synergy (99.5%), and bounded outcomes (100%). Empirical validation using published Apache HTTP Server (1995-2023) case study achieves 60/60 points, reproducing contribution patterns across formation, growth, maturation, and governance phases. Statistical significance confirmed at p<0.001, Cohen's d=0.71. --- ### PRISM: Deriving a White-Box Transformer as a Signal-Noise Decomposition Operator via Maximum Coding Rate Reduction Authors: Dongchen Huang Published: 2026-01-21 Categories: cs.LG, cs.AI, cs.CL, physics.data-an arXiv: https://arxiv.org/abs/2601.15540v2 Deep learning models, particularly Transformers, are often criticized as "black boxes" and lack interpretability. We propose Prism, a white-box attention-based architecture derived from the principles of Maximizing Coding Rate Reduction ($\text{MCR}^2$). By modeling the attention mechanism as a gradient ascent process on a distinct signal-noise manifold, we introduce a specific irrational frequency separation ($π$-RoPE) to enforce incoherence between signal (semantic) and noise (syntactic) subspaces. We show empirical evidence that these geometric inductive biases can induce unsupervised functional disentanglement alone. Prism spontaneously specializes its attention heads into spectrally distinct regimes: low-frequency heads capturing long-range causal dependencies (signal) and high-frequency heads handling local syntactic constraints and structural artifacts. To provide a theoretical grounding for these spectral phenomena, we draw an analogy between attention mechanism and a Hamiltonian dynamical system and identify that the standard geometric progression of Rotary Positional Embeddings (RoPE) induces dense resonance networks (Arnold Tongues), leading to feature rank collapse. Empirical validation on 124M-parameter models trained on OpenWebText demonstrates that Prism spontaneously isolates the Attention Sink pathology and maintains isentropic information flow across layers. Further, we suggest a physics-informed plug-and-play intervention KAM-RoPE for large language models (LLMs). Our results suggest that interpretability and performance can be unified through principled geometric construction, offering a theoretically grounded alternative to heuristic architectural modifications --- ### A Machine Vision Approach to Preliminary Skin Lesion Assessments Authors: Ali Khreis, Ro'Yah Radaideh, Quinn McGill Published: 2026-01-21 Categories: eess.IV, cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2601.15539v1 Early detection of malignant skin lesions is critical for improving patient outcomes in aggressive, metastatic skin cancers. This study evaluates a comprehensive system for preliminary skin lesion assessment that combines the clinically established ABCD rule of dermoscopy (analyzing Asymmetry, Borders, Color, and Dermoscopic Structures) with machine learning classification. Using a 1,000-image subset of the HAM10000 dataset, the system implements an automated, rule-based pipeline to compute a Total Dermoscopy Score (TDS) for each lesion. This handcrafted approach is compared against various machine learning solutions, including traditional classifiers (Logistic Regression, Random Forest, and SVM) and deep learning models. While the rule-based system provides high clinical interpretability, results indicate a performance bottleneck when reducing complex morphology to five numerical features. Experimental findings show that transfer learning with EfficientNet-B0 failed significantly due to domain shift between natural and medical images. In contrast, a custom three-layer Convolutional Neural Network (CNN) trained from scratch achieved 78.5% accuracy and 86.5% recall on median-filtered images, representing a 19-point accuracy improvement over traditional methods. The results demonstrate that direct pixel-level learning captures diagnostic patterns beyond handcrafted features and that purpose-built lightweight architectures can outperform large pretrained models for small, domain-specific medical datasets. --- ### QUAIL: Quantization Aware Unlearning for Mitigating Misinformation in LLMs Authors: Himanshu Mishra, Kanwal Mehreen Published: 2026-01-21 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.15538v1 Machine unlearning aims to remove specific knowledge (e.g., copyrighted or private data) from a trained model without full retraining. In practice, models are often quantized (e.g., 4-bit) for deployment, but we find that quantization can catastrophically restore forgotten information (1). In this paper, we (1) analyze why low-bit quantization undermines unlearning, and (2) propose a quantization-aware unlearning method to mitigate this. We first compute weight-change statistics and bucket overlaps in quantization to show that typical unlearning updates are too small to cross quantization thresholds. Building on this insight, we introduce a logits space hinge loss: for each forget example, we force the output logits of the unlearned model to differ from the original model by at least a margin (half the quantization step). This ensures forgotten examples remain distinguishable even after quantization. We evaluate on language and classification tasks (including a Twitter misinformation dataset) and show our method preserves forgetting under 4-bit quantization, whereas existing methods almost entirely recover the forgotten knowledge. --- ### From Generative Engines to Actionable Simulators: The Imperative of Physical Grounding in World Models Authors: Zhikang Chen, Tingting Zhu Published: 2026-01-21 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.15533v1 A world model is an AI system that simulates how an environment evolves under actions, enabling planning through imagined futures rather than reactive perception. Current world models, however, suffer from visual conflation: the mistaken assumption that high-fidelity video generation implies an understanding of physical and causal dynamics. We show that while modern models excel at predicting pixels, they frequently violate invariant constraints, fail under intervention, and break down in safety-critical decision-making. This survey argues that visual realism is an unreliable proxy for world understanding. Instead, effective world models must encode causal structure, respect domain-specific constraints, and remain stable over long horizons. We propose a reframing of world models as actionable simulators rather than visual engines, emphasizing structured 4D interfaces, constraint-aware dynamics, and closed-loop evaluation. Using medical decision-making as an epistemic stress test, where trial-and-error is impossible and errors are irreversible, we demonstrate that a world model's value is determined not by how realistic its rollouts appear, but by its ability to support counterfactual reasoning, intervention planning, and robust long-horizon foresight. --- ### Machine learning-enhanced non-amnestic Alzheimer's disease diagnosis from MRI and clinical features Authors: Megan A. Witherow, Michael L. Evans, Ahmed Temtam, Hamid R. Okhravi, Khan M. Iftekharuddin Published: 2026-01-21 Categories: cs.LG, q-bio.NC, q-bio.QM arXiv: https://arxiv.org/abs/2601.15530v2 Alzheimer's disease (AD), defined as an abnormal buildup of amyloid plaques and tau tangles in the brain can be diagnosed with high accuracy based on protein biomarkers via PET or CSF analysis. However, due to the invasive nature of biomarker collection, most AD diagnoses are made in memory clinics using cognitive tests and evaluation of hippocampal atrophy based on MRI. While clinical assessment and hippocampal volume show high diagnostic accuracy for amnestic or typical AD (tAD), a substantial subgroup of AD patients with atypical presentation (atAD) are routinely misdiagnosed. To improve diagnosis of atAD patients, we propose a machine learning approach to distinguish between atAD and non-AD cognitive impairment using clinical testing battery and MRI data collected as standard-of-care. We develop and evaluate our approach using 1410 subjects across four groups (273 tAD, 184 atAD, 235 non-AD, and 685 cognitively normal) collected from one private data set and two public data sets from the National Alzheimer's Coordinating Center (NACC) and the Alzheimer's Disease Neuroimaging Initiative (ADNI). We perform multiple atAD vs. non-AD classification experiments using clinical features and hippocampal volume as well as a comprehensive set of MRI features from across the brain. The best performance is achieved by incorporating additional important MRI features, which outperforms using hippocampal volume alone. Furthermore, we use the Boruta statistical approach to identify and visualize significant brain regions distinguishing between diagnostic groups. Our ML approach improves the percentage of correctly diagnosed atAD cases (the recall) from 52% to 69% for NACC and from 34% to 77% for ADNI, while achieving high precision. The proposed approach has important implications for improving diagnostic accuracy for non-amnestic atAD in clinical settings using only clinical testing battery and MRI. --- ### TransportAgents: a multi-agents LLM framework for traffic accident severity prediction Authors: Zhichao Yang, Jiashu He, Jinxuan Fan, Cirillo Cinzia Published: 2026-01-21 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.15519v2 Accurate prediction of traffic crash severity is critical for improving emergency response and public safety planning. Although recent large language models (LLMs) exhibit strong reasoning capabilities, their single-agent architectures often struggle with heterogeneous, domain-specific crash data and tend to generate biased or unstable predictions. To address these limitations, this paper proposes TransportAgents, a hybrid multi-agent framework that integrates category-specific LLM reasoning with a multilayer perceptron (MLP) integration module. Each specialized agent focuses on a particular subset of traffic information, such as demographics, environmental context, or incident details, to produce intermediate severity assessments that are subsequently fused into a unified prediction. Extensive experiments on two complementary U.S. datasets, the Consumer Product Safety Risk Management System (CPSRMS) and the National Electronic Injury Surveillance System (NEISS), demonstrate that TransportAgents consistently outperforms both traditional machine learning and advanced LLM-based baselines. Across three representative backbones, including closed-source models such as GPT-3.5 and GPT-4o, as well as open-source models such as LLaMA-3.3, the framework exhibits strong robustness, scalability, and cross-dataset generalizability. A supplementary distributional analysis further shows that TransportAgents produces more balanced and well-calibrated severity predictions than standard single-agent LLM approaches, highlighting its interpretability and reliability for safety-critical decision support applications. --- ### DS@GT at TREC TOT 2025: Bridging Vague Recollection with Fusion Retrieval and Learned Reranking Authors: Wenxin Zhou, Ritesh Mehta, Anthony Miyaguchi Published: 2026-01-21 Categories: cs.IR, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2601.15518v2 We develop a two-stage retrieval system that combines multiple complementary retrieval methods with a learned reranker and LLM-based reranking, to address the TREC Tip-of-the-Tongue (ToT) task. In the first stage, we employ hybrid retrieval that merges LLM-based retrieval, sparse (BM25), and dense (BGE-M3) retrieval methods. We also introduce topic-aware multi-index dense retrieval that partitions the Wikipedia corpus into 24 topical domains. In the second stage, we evaluate both a trained LambdaMART reranker and LLM-based reranking. To support model training, we generate 5000 synthetic ToT queries using LLMs. Our best system achieves recall of 0.66 and NDCG@1000 of 0.41 on the test set by combining hybrid retrieval with Gemini-2.5-flash reranking, demonstrating the effectiveness of fusion retrieval. --- ### DeltaDorsal: Enhancing Hand Pose Estimation with Dorsal Features in Egocentric Views Authors: William Huang, Siyou Pei, Leyi Zou, Eric J. Gonzalez, Ishan Chatterjee, Yang Zhang Published: 2026-01-21 Categories: cs.CV, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2601.15516v2 The proliferation of XR devices has made egocentric hand pose estimation a vital task, yet this perspective is inherently challenged by frequent finger occlusions. To address this, we propose a novel approach that leverages the rich information in dorsal hand skin deformation, unlocked by recent advances in dense visual featurizers. We introduce a dual-stream delta encoder that learns pose by contrasting features from a dynamic hand with a baseline relaxed position. Our evaluation demonstrates that, using only cropped dorsal images, our method reduces the Mean Per Joint Angle Error (MPJAE) by 18% in self-occluded scenarios (fingers >= 50% occluded) compared to state-of-the-art techniques that depend on the whole hand's geometry and large model backbones. Consequently, our method not only enhances the reliability of downstream tasks like index finger pinch and tap estimation in occluded scenarios but also unlocks new interaction paradigms, such as detecting isometric force for a surface "click" without visible movement while minimizing model size. --- ### Assessing the informative value of macroeconomic indicators for public health forecasting Authors: Shome Chakraborty, Fardil Khan, Soutik Ghosal Published: 2026-01-21 Categories: stat.AP, stat.ML arXiv: https://arxiv.org/abs/2601.15514v1 Macroeconomic conditions influence the environments in which health systems operate, yet their value as leading signals of health system capacity has not been systematically evaluated. In this study, we examine whether selected macroeconomic indicators contain predictive information for several capacity-related public health targets, including employment in the health and social assistance workforce, new business applications in the sector, and health care construction spending. Using monthly U.S. time series data, we evaluate multiple forecasting approaches, including neural network models with different optimization strategies, generalized additive models, random forests, and time series models with exogenous macroeconomic indicators, under alternative model fitting designs. Across evaluation settings, we find that macroeconomic indicators provide a consistent and reproducible predictive signal for some public health targets, particularly workforce and infrastructure measures, while other targets exhibit weaker or less stable predictability. Models emphasizing stability and implicit regularization tend to perform more reliably during periods of economic volatility. These findings suggest that macroeconomic indicators may serve as useful upstream signals for digital public health monitoring, while underscoring the need for careful model selection and validation when translating economic trends into health system forecasting tools. ---