# Research Papers — 2026-01 · bundle page 54 of 189 > Full-text markdown of papers #2651–2700 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/54 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/53.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/55.md) ### Cross-Domain Transfer with Self-Supervised Spectral-Spatial Modeling for Hyperspectral Image Classification Authors: Jianshu Chao, Tianhua Lv, Qiqiong Ma, Yunfei Qiu, Li Fang, Huifang Shen, Wei Yao Published: 2026-01-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18088v1 Self-supervised learning has demonstrated considerable potential in hyperspectral representation, yet its application in cross-domain transfer scenarios remains under-explored. Existing methods, however, still rely on source domain annotations and are susceptible to distribution shifts, leading to degraded generalization performance in the target domain. To address this, this paper proposes a self-supervised cross-domain transfer framework that learns transferable spectral-spatial joint representations without source labels and achieves efficient adaptation under few samples in the target domain. During the self-supervised pre-training phase, a Spatial-Spectral Transformer (S2Former) module is designed. It adopts a dual-branch spatial-spectral transformer and introduces a bidirectional cross-attention mechanism to achieve spectral-spatial collaborative modeling: the spatial branch enhances structural awareness through random masking, while the spectral branch captures fine-grained differences. Both branches mutually guide each other to improve semantic consistency. We further propose a Frequency Domain Constraint (FDC) to maintain frequency-domain consistency through real Fast Fourier Transform (rFFT) and high-frequency magnitude loss, thereby enhancing the model's capability to discern fine details and boundaries. During the fine-tuning phase, we introduce a Diffusion-Aligned Fine-tuning (DAFT) distillation mechanism. This aligns semantic evolution trajectories through a teacher-student structure, enabling robust transfer learning under low-label conditions. Experimental results demonstrate stable classification performance and strong cross-domain adaptability across four hyperspectral datasets, validating the method's effectiveness under resource-constrained conditions. --- ### "Crash Test Dummies" for AI-Enabled Clinical Assessment: Validating Virtual Patient Scenarios with Virtual Learners Authors: Brian Gin, Ahreum Lim, Flávia Silva e Oliveira, Kuan Xing, Xiaomei Song, Gayana Amiyangoda, Thilanka Seneviratne, Alison F. Doubleday, Ananya Gangopadhyaya, Bob Kiser, Lukas Shum-Tim, Dhruva Patel et al. Published: 2026-01-26 Categories: cs.HC, cs.AI, stat.AP arXiv: https://arxiv.org/abs/2601.18085v1 Background: In medical and health professions education (HPE), AI is increasingly used to assess clinical competencies, including via virtual standardized patients. However, most evaluations rely on AI-human interrater reliability and lack a measurement framework for how cases, learners, and raters jointly shape scores. This leaves robustness uncertain and can expose learners to misguidance from unvalidated systems. We address this by using AI "simulated learners" to stress-test and psychometrically characterize assessment pipelines before human use. Objective: Develop an open-source AI virtual patient platform and measurement model for robust competency evaluation across cases and rating conditions. Methods: We built a platform with virtual patients, virtual learners with tunable ACGME-aligned competency profiles, and multiple independent AI raters scoring encounters with structured Key-Features items. Transcripts were analyzed with a Bayesian HRM-SDT model that treats ratings as decisions under uncertainty and separates learner ability, case performance, and rater behavior; parameters were estimated with MCMC. Results: The model recovered simulated learners' competencies, with significant correlations to the generating competencies across all ACGME domains despite a non-deterministic pipeline. It estimated case difficulty by competency and showed stable rater detection (sensitivity) and criteria (severity/leniency thresholds) across AI raters using identical models/prompts but different seeds. We also propose a staged "safety blueprint" for deploying AI tools with learners, tied to entrustment-based validation milestones. Conclusions: Combining a purpose-built virtual patient platform with a principled psychometric model enables robust, interpretable, generalizable competency estimates and supports validation of AI-assisted assessment prior to use with human learners. --- ### DRPG (Decompose, Retrieve, Plan, Generate): An Agentic Framework for Academic Rebuttal Authors: Peixuan Han, Yingjie Yu, Jingjun Xu, Jiaxuan You Published: 2026-01-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18081v2 Despite the growing adoption of large language models (LLMs) in scientific research workflows, automated support for academic rebuttal, a crucial step in academic communication and peer review, remains largely underexplored. Existing approaches typically rely on off-the-shelf LLMs or simple pipelines, which struggle with long-context understanding and often fail to produce targeted and persuasive responses. In this paper, we propose DRPG, an agentic framework for automatic academic rebuttal generation that operates through four steps: Decompose reviews into atomic concerns, Retrieve relevant evidence from the paper, Plan rebuttal strategies, and Generate responses accordingly. Notably, the Planner in DRPG reaches over 98% accuracy in identifying the most feasible rebuttal direction. Experiments on data from top-tier conferences demonstrate that DRPG significantly outperforms existing rebuttal pipelines and achieves performance beyond the average human level using only an 8B model. Our analysis further demonstrates the effectiveness of the planner design and its value in providing multi-perspective and explainable suggestions. We also showed that DRPG works well in a more complex multi-round setting. These results highlight the effectiveness of DRPG and its potential to provide high-quality rebuttal content and support the scaling of academic discussions. Codes for this work are available at https://github.com/ulab-uiuc/DRPG-RebuttalAgent. --- ### Use of operator defect identities in multi-channel signal plus residual-analysis via iterated products and telescoping energy-residuals: Applications to kernels in machine learning Authors: Palle E. T. Jorgensen, Myung-Sin Song, James F. Tian Published: 2026-01-26 Categories: math.FA, cs.LG, math.OA arXiv: https://arxiv.org/abs/2601.18080v1 We present a new operator theoretic framework for analysis of complex systems with intrinsic subdivisions into components, taking the form of "residuals" in general, and "telescoping energy residuals" in particular. We prove new results which yield admissibility/effectiveness, and new a priori bounds on energy residuals. Applications include infinite-dimensional Kaczmarz theory for $λ_{n}$-relaxed variants, and $λ_{n}$-effectiveness. And we give applications of our framework to generalized machine learning algorithms, greedy Kernel Principal Component Analysis (KPCA), proving explicit convergence results, residual energy decomposition, and criteria for stability under noise. --- ### Sparks of Cooperative Reasoning: LLMs as Strategic Hanabi Agents Authors: Mahesh Ramesh, Kaousheik Jayakumar, Aswinkumar Ramkumar, Pavan Thodima, Aniket Rege, Emmanouil-Vasileios Vlatakis-Gkaragkounis Published: 2026-01-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.18077v3 Cooperative reasoning under incomplete information remains challenging for both humans and multi-agent systems. The card game Hanabi embodies this challenge, requiring theory-of-mind reasoning and strategic communication. We benchmark 17 state-of-the-art LLM agents in 2-5 player games and study the impact of context engineering across model scales (4B to 600B+) to understand persistent coordination failures and robustness to scaffolding: from a minimal prompt with only explicit card details (Watson setting), to scaffolding with programmatic, Bayesian-motivated deductions (Sherlock), to multi-turn state tracking via working memory (Mycroft setting). We show that (1) agents can maintain an internal working memory for state tracking and (2) cross-play performance between different LLMs smoothly interpolates with model strength. In the Sherlock setting, the strongest reasoning models exceed 15 points on average across player counts, yet still trail experienced humans and specialist Hanabi agents, both consistently scoring above 20. We release the first public Hanabi datasets with annotated trajectories and move utilities: (1) HanabiLogs, containing 1,520 full game logs for instruction tuning, and (2) HanabiRewards, containing 560 games with dense move-level value annotations for all candidate moves. Supervised and RL finetuning of a 4B open-weight model (Qwen3-Instruct) on our datasets improves cooperative Hanabi play by 21% and 156% respectively, bringing performance to within ~3 points of a strong proprietary reasoning model (o4-mini) and surpassing the best non-reasoning model (GPT-4.1) by 52%. The HanabiRewards RL-finetuned model further generalizes beyond Hanabi, improving performance on a cooperative group-guessing benchmark by 11%, temporal reasoning on EventQA by 6.4 points, instruction-following on IFBench by 1.7 Pass@10, and matching AIME 2025 mathematical reasoning Pass@10. --- ### Comparison requires valid measurement: Rethinking attack success rate comparisons in AI red teaming Authors: Alexandra Chouldechova, A. Feder Cooper, Solon Barocas, Abhinav Palia, Dan Vann, Hanna Wallach Published: 2026-01-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18076v1 We argue that conclusions drawn about relative system safety or attack method efficacy via AI red teaming are often not supported by evidence provided by attack success rate (ASR) comparisons. We show, through conceptual, theoretical, and empirical contributions, that many conclusions are founded on apples-to-oranges comparisons or low-validity measurements. Our arguments are grounded in asking a simple question: When can attack success rates be meaningfully compared? To answer this question, we draw on ideas from social science measurement theory and inferential statistics, which, taken together, provide a conceptual grounding for understanding when numerical values obtained through the quantification of system attributes can be meaningfully compared. Through this lens, we articulate conditions under which ASRs can and cannot be meaningfully compared. Using jailbreaking as a running example, we provide examples and extensive discussion of apples-to-oranges ASR comparisons and measurement validity challenges. --- ### Diffusion Model-based Reinforcement Learning for Version Age of Information Scheduling: Average and Tail-Risk-Sensitive Control Authors: Haoyuan Pan, Sizhao Chen, Zhaorui Wang, Tse-Tin Chan Published: 2026-01-26 Categories: cs.NI, cs.AI arXiv: https://arxiv.org/abs/2601.18069v1 Ensuring timely and semantically accurate information delivery is critical in real-time wireless systems. While Age of Information (AoI) quantifies temporal freshness, Version Age of Information (VAoI) captures semantic staleness by accounting for version evolution between transmitters and receivers. Existing VAoI scheduling approaches primarily focus on minimizing average VAoI, overlooking rare but severe staleness events that can compromise reliability under stochastic packet arrivals and unreliable channels. This paper investigates both average-oriented and tail-risk-sensitive VAoI scheduling in a multi-user status update system with long-term transmission cost constraints. We first formulate the average VAoI minimization problem as a constrained Markov decision process and introduce a deep diffusion-based Soft Actor-Critic (D2SAC) algorithm. By generating actions through a diffusion-based denoising process, D2SAC enhances policy expressiveness and establishes a strong baseline for mean performance. Building on this foundation, we put forth RS-D3SAC, a risk-sensitive deep distributional diffusion-based Soft Actor-Critic algorithm. RS-D3SAC integrates a diffusion-based actor with a quantile-based distributional critic, explicitly modeling the full VAoI return distribution. This enables principled tail-risk optimization via Conditional Value-at-Risk (CVaR) while satisfying long-term transmission cost constraints. Extensive simulations show that, while D2SAC reduces average VAoI, RS-D3SAC consistently achieves substantial reductions in CVaR without sacrificing mean performance. The dominant gain in tail-risk reduction stems from the distributional critic, with the diffusion-based actor providing complementary refinement to stabilize and enrich policy decisions, highlighting their effectiveness for robust and risk-aware VAoI scheduling in multi-user wireless systems. --- ### XGuardian: Towards Explainable and Generalized AI Anti-Cheat on FPS Games Authors: Jiayi Zhang, Chenxin Sun, Chenxiong Qian Published: 2026-01-26 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2601.18068v1 Aim-assist cheats are the most prevalent and infamous form of cheating in First-Person Shooter (FPS) games, which help cheaters illegally reveal the opponent's location and auto-aim and shoot, and thereby pose significant threats to the game industry. Although a considerable research effort has been made to automatically detect aim-assist cheats, existing works suffer from unreliable frameworks, limited generalizability, high overhead, low detection performance, and a lack of explainability of detection results. In this paper, we propose XGuardian, a server-side generalized and explainable system for detecting aim-assist cheats to overcome these limitations. It requires only two raw data inputs, pitch and yaw, which are all FPS games' must-haves, to construct novel temporal features and describe aim trajectories, which are essential for distinguishing cheaters and normal players. XGuardian is evaluated with the latest mainstream FPS game CS2, and validates its generalizability with another two different games. It achieves high detection performance and low overhead compared to prior works across different games with real-world and large-scale datasets, demonstrating wide generalizability and high effectiveness. It is able to justify its predictions and thereby shorten the ban cycle. We make XGuardian as well as our datasets publicly available. --- ### EvolVE: Evolutionary Search for LLM-based Verilog Generation and Optimization Authors: Wei-Po Hsin, Ren-Hao Deng, Yao-Ting Hsieh, En-Ming Huang, Shih-Hao Hung Published: 2026-01-26 Categories: cs.AI, cs.NE, cs.PL arXiv: https://arxiv.org/abs/2601.18067v1 Verilog's design cycle is inherently labor-intensive and necessitates extensive domain expertise. Although Large Language Models (LLMs) offer a promising pathway toward automation, their limited training data and intrinsic sequential reasoning fail to capture the strict formal logic and concurrency inherent in hardware systems. To overcome these barriers, we present EvolVE, the first framework to analyze multiple evolution strategies on chip design tasks, revealing that Monte Carlo Tree Search (MCTS) excels at maximizing functional correctness, while Idea-Guided Refinement (IGR) proves superior for optimization. We further leverage Structured Testbench Generation (STG) to accelerate the evolutionary process. To address the lack of complex optimization benchmarks, we introduce IC-RTL, targeting industry-scale problems derived from the National Integrated Circuit Contest. Evaluations establish EvolVE as the new state-of-the-art, achieving 98.1% on VerilogEval v2 and 92% on RTLLM v2. Furthermore, on the industry-scale IC-RTL suite, our framework surpasses reference implementations authored by contest participants, reducing the Power, Performance, Area (PPA) product by up to 66% in Huffman Coding and 17% in the geometric mean across all problems. The source code of the IC-RTL benchmark is available at https://github.com/weiber2002/ICRTL. --- ### Grounded Concreteness: Human-Like Concreteness Sensitivity in Vision-Language Models Authors: Aryan Roy, Zekun Wang, Christopher J. MacLellan Published: 2026-01-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.18065v1 Do vision--language models (VLMs) develop more human-like sensitivity to linguistic concreteness than text-only large language models (LLMs) when both are evaluated with text-only prompts? We study this question with a controlled comparison between matched Llama text backbones and their Llama Vision counterparts across multiple model scales, treating multimodal pretraining as an ablation on perceptual grounding rather than access to images at inference. We measure concreteness effects at three complementary levels: (i) output behavior, by relating question-level concreteness to QA accuracy; (ii) embedding geometry, by testing whether representations organize along a concreteness axis; and (iii) attention dynamics, by quantifying context reliance via attention-entropy measures. In addition, we elicit token-level concreteness ratings from models and evaluate alignment to human norm distributions, testing whether multimodal training yields more human-consistent judgments. Across benchmarks and scales, VLMs show larger gains on more concrete inputs, exhibit clearer concreteness-structured representations, produce ratings that better match human norms, and display systematically different attention patterns consistent with increased grounding. --- ### Resonant Sparse Geometry Networks Authors: Hasi Hays Published: 2026-01-26 Categories: cs.LG, cs.AI, cs.NE arXiv: https://arxiv.org/abs/2601.18064v1 We introduce Resonant Sparse Geometry Networks (RSGN), a brain-inspired architecture with self-organizing sparse hierarchical input-dependent connectivity. Unlike Transformer architectures that employ dense attention mechanisms with O(n^2) computational complexity, RSGN embeds computational nodes in learned hyperbolic space where connection strength decays with geodesic distance, achieving dynamic sparsity that adapts to each input. The architecture operates on two distinct timescales: fast differentiable activation propagation optimized through gradient descent, and slow Hebbian-inspired structural learning for connectivity adaptation through local correlation rules. We provide rigorous mathematical analysis demonstrating that RSGN achieves O(n*k) computational complexity, where k << n represents the average active neighborhood size. Experimental evaluation on hierarchical classification and long-range dependency tasks demonstrates that RSGN achieves 96.5% accuracy on long-range dependency tasks while using approximately 15x fewer parameters than standard Transformers. On challenging hierarchical classification with 20 classes, RSGN achieves 23.8% accuracy (compared to 5% random baseline) with only 41,672 parameters, nearly 10x fewer than the Transformer baselines which require 403,348 parameters to achieve 30.1% accuracy. Our ablation studies confirm the contribution of each architectural component, with Hebbian learning providing consistent improvements. These results suggest that brain-inspired principles of sparse, geometrically-organized computation offer a promising direction toward more efficient and biologically plausible neural architectures. --- ### Expert Evaluation and the Limits of Human Feedback in Mental Health AI Safety Testing Authors: Kiana Jafari, Paul Ulrich Nikolaus Rust, Duncan Eddy, Robbie Fraser, Nina Vasan, Darja Djordjevic, Akanksha Dadlani, Max Lamparth, Eugenia Kim, Mykel Kochenderfer Published: 2026-01-26 Categories: cs.AI, cs.HC arXiv: https://arxiv.org/abs/2601.18061v3 Learning from human feedback~(LHF) assumes that expert judgments, appropriately aggregated, yield valid ground truth for training and evaluating AI systems. We tested this assumption in mental health, where high safety stakes make expert consensus essential. Three certified psychiatrists independently evaluated LLM-generated responses using a calibrated rubric. Despite similar training and shared instructions, inter-rater reliability was consistently poor ($ICC$ $0.087$--$0.295$), falling below thresholds considered acceptable for consequential assessment. Disagreement was highest on the most safety-critical items. Suicide and self-harm responses produced greater divergence than any other category, and was systematic rather than random. One factor yielded negative reliability (Krippendorff's $α= -0.203$), indicating structured disagreement worse than chance. Qualitative interviews revealed that disagreement reflects coherent but incompatible individual clinical frameworks, safety-first, engagement-centered, and culturally-informed orientations, rather than measurement error. By demonstrating that experts rely on holistic risk heuristics rather than granular factor discrimination, these findings suggest that aggregated labels function as arithmetic compromises that effectively erase grounded professional philosophies. Our results characterize expert disagreement in safety-critical AI as a sociotechnical phenomenon where professional experience introduces sophisticated layers of principled divergence. We discuss implications for reward modeling, safety classification, and evaluation benchmarks, recommending that practitioners shift from consensus-based aggregation to alignment methods that preserve and learn from expert disagreement. --- ### Differentiable Architecture Search for Adversarially Robust Quantum Computer Vision Authors: Mohamed Afane, Quanjiang Long, Haoting Shen, Ying Mao, Junaid Farooq, Ying Wang, Juntao Chen Published: 2026-01-26 Categories: quant-ph, cs.CV arXiv: https://arxiv.org/abs/2601.18058v1 Current quantum neural networks suffer from extreme sensitivity to both adversarial perturbations and hardware noise, creating a significant barrier to real-world deployment. Existing robustness techniques typically sacrifice clean accuracy or require prohibitive computational resources. We propose a hybrid quantum-classical Differentiable Quantum Architecture Search (DQAS) framework that addresses these limitations by jointly optimizing circuit structure and robustness through gradient-based methods. Our approach enhances traditional DQAS with a lightweight Classical Noise Layer applied before quantum processing, enabling simultaneous optimization of gate selection and noise parameters. This design preserves the quantum circuit's integrity while introducing trainable perturbations that enhance robustness without compromising standard performance. Experimental validation on MNIST, FashionMNIST, and CIFAR datasets shows consistent improvements in both clean and adversarial accuracy compared to existing quantum architecture search methods. Under various attack scenarios, including Fast Gradient Sign Method (FGSM), Projected Gradient Descent (PGD), Basic Iterative Method (BIM), and Momentum Iterative Method (MIM), and under realistic quantum noise conditions, our hybrid framework maintains superior performance. Testing on actual quantum hardware confirms the practical viability of discovered architectures. These results demonstrate that strategic classical preprocessing combined with differentiable quantum architecture optimization can significantly enhance quantum neural network robustness while maintaining computational efficiency. --- ### Neurocomputational Mechanisms of Syntactic Transfer in Bilingual Sentence Production Authors: Ahmet Yavuz Uluslu, Elliot Murphy Published: 2026-01-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.18056v2 We discuss the benefits of incorporating oscillatory neural mechanisms into the study of bilingual production errors and their traditionally documented timing signatures (e.g., event-related potentials), which can offer new implementational-level constraints for theories of bilingualism. We argue that a recent neural model of language, ROSE, can offer a neurocomputational account of syntactic transfer in bilingual production, capturing some of its formal properties and the scope of morphosyntactic sequencing failure modes. We take as a case study cross-linguistic influence (CLI) and attendant theories of functional inhibition/competition, and present these as being driven by specific oscillatory failure modes during L2 sentence planning. We argue that modeling CLI in this way not only offers the kinds of linking hypotheses ROSE was built to encourage, but also licenses the exploration of more spatiotemporally complex biomarkers of non-native processing than more commonly discussed neural signatures. --- ### Addressing LLM Diversity by Infusing Random Concepts Authors: Pulin Agrawal, Prasoon Goyal Published: 2026-01-26 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.18053v1 Large language models (LLMs) are known to produce outputs with limited diversity. In this work, we study whether infusing random concepts in the prompts can improve the diversity of the generated outputs. To benchmark the approach, we design a systematic evaluation protocol which involves prompting an LLM with questions of the form "Name 10 Hollywood actors", and analyzing diversity measures of the resulting LLM outputs. Our experiments on multiple LLMs show that prepending random words/sentences unrelated to the prompt result in greater diversity in the outputs of LLMs. We believe that this promising result and the evaluation protocol opens up interesting avenues for future work, such as how infusing randomness into LLMs could be applied to other domains. Further, the evaluation protocol could also inspire research into benchmarking LLM diversity more systematically. --- ### Semi-Supervised Hyperspectral Image Classification with Edge-Aware Superpixel Label Propagation and Adaptive Pseudo-Labeling Authors: Yunfei Qiu, Qiqiong Ma, Tianhua Lv, Li Fang, Shudong Zhou, Wei Yao Published: 2026-01-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18049v2 Significant progress has been made in semi-supervised hyperspectral image (HSI) classification regarding feature extraction and classification performance. However, due to high annotation costs and limited sample availability, semi-supervised learning still faces challenges such as boundary label diffusion and pseudo-label instability. To address these issues, this paper proposes a novel semi-supervised hyperspectral classification framework integrating spatial prior information with a dynamic learning mechanism. First, we design an Edge-Aware Superpixel Label Propagation (EASLP) module. By integrating edge intensity penalty with neighborhood correction strategy, it mitigates label diffusion from superpixel segmentation while enhancing classification robustness in boundary regions. Second, we introduce a Dynamic History-Fused Prediction (DHP) method. By maintaining historical predictions and dynamically weighting them with current results, DHP smoothens pseudo-label fluctuations and improves temporal consistency and noise resistance. Concurrently, incorporating condifence and consistency measures, the Adaptive Tripartite Sample Categorization (ATSC) strategy implements hierarchical utilization of easy, ambiguous, and hard samples, leading to enhanced pseudo-label quality and learning efficiency. The Dynamic Reliability-Enhanced Pseudo-Label Framework (DREPL), composed of DHP and ATSC, strengthens pseudo-label stability across temporal and sample domains. Through synergizes operation with EASLP, it achieves spatio-temporal consistency optimization. Evaluations on four benchmark datasets demonstrate its capability to maintain superior classification performance. --- ### Laser interferometry as a robust neuromorphic platform for machine learning Authors: Amanuel Anteneh, Kyungeun Kim, J. M. Schwarz, Israel Klich, Olivier Pfister Published: 2026-01-26 Categories: physics.optics, cs.ET, cs.LG arXiv: https://arxiv.org/abs/2601.18047v4 We present a method for implementing an optical neural network using only linear optical resources, namely field displacement and interferometry applied to coherent states of light. The nonlinearity required for learning in a neural network is realized via an encoding of the input into phase shifts allowing for far more straightforward experimental implementation compared to previous proposals for, and demonstrations of, $\textit{in situ}$ inference. Beyond $\textit{in situ}$ inference, the method enables $\textit{in situ}$ training by utilizing established techniques like parameter shift methods or physical backpropagation to extract gradients directly from measurements of the linear optical circuit. We also investigate the effect of photon losses and find the model to be very resilient to these. --- ### Results-Actionability Gap: Understanding How Practitioners Evaluate LLM Products in the Wild Authors: Willem van der Maden, Malak Sadek, Ziang Xiao, Aske Mottelson, Q. Vera Liao, Jichen Zhu Published: 2026-01-25 Categories: cs.SE, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2604.16304v1 How do product teams evaluate LLM-powered products? As organizations integrate large language models (LLMs) into digital products, their unpredictable nature makes traditional evaluation approaches inadequate, yet little is known about how practitioners navigate this challenge. Through interviews with nineteen practitioners across diverse sectors, we identify ten evaluation practices spanning informal 'vibe checks' to organizational meta-work. Beyond confirming four documented challenges, we introduce a novel fifth we call the results-actionability gap, in which practitioners gather evaluation data but cannot translate findings into concrete improvements. Drawing on patterns from successful teams, we contribute strategies to bridge this gap, supporting practitioners' formalization journey from ad-hoc interpretive practices (e.g., vibe checks) toward systematic evaluation. Our analysis suggests these interpretive practices are necessary adaptations to LLM characteristics rather than methodological failures. For HCI researchers, this presents a research opportunity to support practitioners in systematizing emerging practices rather than developing new evaluation frameworks. --- ### State Design Matters: How Representations Shape Dynamic Reasoning in Large Language Models Authors: Annie Wong, Aske Plaat, Thomas Bäck, Niki van Stein, Anna V. Kononova Published: 2026-01-25 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.15858v1 As large language models (LLMs) move from static reasoning tasks toward dynamic environments, their success depends on the ability to navigate and respond to an environment that changes as they interact at inference time. An underexplored factor in these settings is the representation of the state. Holding model parameters fixed, we systematically vary three key aspects: (1) state granularity (long form versus summary), (2) structure (natural language versus symbolic), and (3) spatial grounding (text-only versus images or textual map encodings) across sequential decision-making benchmarks. We find that trajectory summarisation improves performance by reducing noise and stabilising long-horizon reasoning. Second, natural language representations are the most robust across models, whereas structured encodings help mainly for models with strong code or structured output priors, such as JSON schemas. Third, while image-inputs show some benefit, text-based spatial encodings prove most effective. This advantage stems not from the spatial information itself, but from the act of construction, which compels the model to perform the spatial reasoning that static input does not elicit. Overall, we demonstrate that design choices for representing state are a decisive factor in performance, distinct from the availability of information itself. We note, however, that even with improved representations, current LLMs and VLMs remain brittle over long horizons, particularly when they must synthesise information to manage multiple subtasks to reach a goal. --- ### Multi-source Heterogeneous Public Opinion Analysis via Collaborative Reasoning and Adaptive Fusion: A Systematically Integrated Approach Authors: Yi Liu Published: 2026-01-25 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.15857v1 The analysis of public opinion from multiple heterogeneous sources presents significant challenges due to structural differences, semantic variations, and platform-specific biases. This paper introduces a novel Collaborative Reasoning and Adaptive Fusion (CRAF) framework that systematically integrates traditional feature-based methods with large language models (LLMs) through a structured multi-stage reasoning mechanism. Our approach features four key innovations: (1) a cross-platform collaborative attention module that aligns semantic representations while preserving source-specific characteristics, (2) a hierarchical adaptive fusion mechanism that dynamically weights features based on both data quality and task requirements, (3) a joint optimization strategy that simultaneously learns topic representations and sentiment distributions through shared latent spaces, and (4) a novel multimodal extraction capability that processes video content from platforms like Douyin and Kuaishou by integrating OCR, ASR, and visual sentiment analysis. Theoretical analysis demonstrates that CRAF achieves a tighter generalization bound with a reduction of O(sqrt(d log K / m)) compared to independent source modeling, where d is feature dimensionality, K is the number of sources, and m is sample size. Comprehensive experiments on three multi-platform datasets (Weibo-12, CrossPlatform-15, NewsForum-8) show that CRAF achieves an average topic clustering ARI of 0.76 (4.1% improvement over best baseline) and sentiment analysis F1-score of 0.84 (3.8% improvement). The framework exhibits strong cross-platform adaptability, reducing the labeled data requirement for new platforms by 75%. --- ### Rethinking Soft Compression in Retrieval-Augmented Generation: A Query-Conditioned Selector Perspective Authors: Yunhao Liu, Zian Jia, Xinyu Gao, Kanjun Xu, Yun Xiong Published: 2026-01-25 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2602.15856v2 Retrieval-Augmented Generation (RAG) effectively grounds Large Language Models (LLMs) with external knowledge and is widely applied to Web-related tasks. However, its scalability is hindered by excessive context length and redundant retrievals. Recent research on soft context compression aims to address this by encoding long documents into compact embeddings, yet they often underperform non-compressed RAG due to their reliance on auto-encoder-like full-compression that forces the encoder to compress all document information regardless of relevance to the input query. In this work, we conduct an analysis on this paradigm and reveal two fundamental limitations: (I) Infeasibility, full-compression conflicts with the LLM's downstream generation behavior; and (II) Non-necessity: full-compression is unnecessary and dilutes task-relevant information density. Motivated by these insights, we introduce SeleCom, a selector-based soft compression framework for RAG that redefines the encoder's role as query-conditioned information selector. The selector is decoder-only and is trained with a massive, diverse and difficulty-graded synthetic QA dataset with curriculum learning. Extensive experiments show that SeleCom significantly outperforms existing soft compression approaches and achieves competitive or superior performance to non-compression baselines, while reducing computation and latency by 33.8%~84.6%. --- ### Energy-Aware Reinforcement Learning for Robotic Manipulation of Articulated Components in Infrastructure Operation and Maintenance Authors: Xiaowen Tao, Yinuo Wang, Haitao Ding, Yuanyang Qi, Ziyu Song Published: 2026-01-25 Categories: eess.SY, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2602.12288v3 With the growth of intelligent civil infrastructure and smart cities, operation and maintenance (O&M) increasingly requires safe, efficient, and energy-conscious robotic manipulation of articulated components, including access doors, service drawers, and pipeline valves. However, existing robotic approaches either focus primarily on grasping or target object-specific articulated manipulation, and they rarely incorporate explicit actuation energy into multi-objective optimisation, which limits their scalability and suitability for long-term deployment in real O&M settings. Therefore, this paper proposes an articulation-agnostic and energy-aware reinforcement learning framework for robotic manipulation in intelligent infrastructure O&M. The method combines part-guided 3D perception, weighted point sampling, and PointNet-based encoding to obtain a compact geometric representation that generalises across heterogeneous articulated objects. Manipulation is formulated as a Constrained Markov Decision Process (CMDP), in which actuation energy is explicitly modelled and regulated via a Lagrangian-based constrained Soft Actor-Critic scheme. The policy is trained end-to-end under this CMDP formulation, enabling effective articulated-object operation while satisfying a long-horizon energy budget. Experiments on representative O&M tasks demonstrate 16%-30% reductions in energy consumption, 16%-32% fewer steps to success, and consistently high success rates, indicating a scalable and sustainable solution for infrastructure O&M manipulation. --- ### TabularMath: Evaluating Computational Extrapolation in Tabular Learning via Program-Verified Synthesis Authors: Zerui Cheng, Jiashuo Liu, Jianzhu Yao, Pramod Viswanath, Ge Zhang, Wenhao Huang Published: 2026-01-25 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02523v1 Standard tabular benchmarks mainly focus on the evaluation of a model's capability to interpolate values inside a data manifold, where models good at performing local statistical smoothing are rewarded. However, there exists a very large category of high-value tabular data, including financial modeling and physical simulations, which are generated based upon deterministic computational processes, as opposed to stochastic and noisy relationships. Therefore, we investigate if tabular models can provide an extension from statistical interpolation to computational extrapolation. We propose TabularMath, a diagnostic benchmark of 114 deterministic problems (233,472 rows) generated from verified programs based on GSM8K and AIME. We evaluate 9 tabular architectures and in-context learning (ICL) with GPT-OSS-120B. On standard regression metrics, TabPFN v2.5 performs remarkably well, achieving R^2=0.998 in-distribution and maintaining positive R^2 even under distribution shift, which is unique among the tabular models we tested. When we measure rounded consistency (exact integer match), a different picture emerges: TabPFN v2.5 drops below 10% on out-of-distribution data, while ICL maintains around 40%. This gap between R^2 and exact-match accuracy suggests that tabular models learn smooth function approximations but struggle to recover precise computational outputs under extrapolation. The two paradigms appear complementary: TabPFN scales efficiently with data; ICL achieves exact computation from few examples. We release all code and data to support further investigation. --- ### IMU-1: Sample-Efficient Pre-training of Small Language Models Authors: George Grigorev Published: 2026-01-25 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02522v1 We present IMU-1, a 430M-parameter language model trained on 72B tokens that approaches the benchmark performance of models trained on 56x more data. We describe a validated training recipe combining recent architectural interventions (QK-norm attention, per-head gating, value residuals, LayerNorm scaling) with optimization advances (NorMuon with cautious weight decay, muP parametrization) and a three-stage training schedule with post-hoc checkpoint EMA. We provide ablations for each component and release code, weights and data to enable reproduction: https://huggingface.co/thepowerfuldeez/imu1_base --- ### Scaled Dot-Product Attention implements projection of inputs onto a common surface Authors: Terence D Sanger Published: 2026-01-25 Categories: cs.LG, cs.AI, eess.SP arXiv: https://arxiv.org/abs/2602.02521v1 Scaled dot-product attention (SDPA) is a fundamental component responsible for the success of large-language models and other nonlinear signal processing applications. The rationale for SDPA has been based upon "query, key, value" concepts borrowed from database theory, but these concepts are difficult to reconcile with standard methods in mathematical signal processing. We show that SDPA can be rewritten in a different but mathematically equivalent form as a projection of the input vectors onto a common surface determined by the inputs themselves. Therefore SDPA discovers nonlinear dependencies in the input that are time-dependent and context-dependent. The rewritten form of SDPA permits increased speed of both feedforward and learning algorithms, but more importantly suggests potential extensions. In the context of language, we re-interpret the role of SDPA as finding a time-dependent contextual meaning determined by the surface on which the set of input vectors lies. Input token embeddings are then modified by the local context surface. This interpretation differs substantially from the concept of "self-attention", and provides a strong justification for the use of SDPA for time-series data with time-varying local nonlinear dependencies. --- ### HYPE-EDIT-1: Benchmark for Measuring Reliability in Frontier Image Editing Models Authors: Wing Chan, Richard Allen Published: 2026-01-25 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.00105v1 Public demos of image editing models are typically best-case samples; real workflows pay for retries and review time. We introduce HYPE-EDIT-1, a 100-task benchmark of reference-based marketing/design edits with binary pass/fail judging. For each task we generate 10 independent outputs to estimate per-attempt pass rate, pass@10, expected attempts under a retry cap, and an effective cost per successful edit that combines model price with human review time. We release 50 public tasks and maintain a 50-task held-out private split for server-side evaluation, plus a standardized JSON schema and tooling for VLM and human-based judging. Across the evaluated models, per-attempt pass rates span 34-83 percent and effective cost per success spans USD 0.66-1.42. Models that have low per-image pricing are more expensive when you consider the total effective cost of retries and human reviews. --- ### R3G: A Reasoning-Retrieval-Reranking Framework for Vision-Centric Answer Generation Authors: Zhuohong Chen, Zhengxian Wu, Zirui Liao, Shenao Jiang, Hangrui Xu, Yang Chen, Chaokui Su, Xiaoyu Liu, Haoqian Wang Published: 2026-01-25 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.00104v4 Vision-centric retrieval for VQA requires retrieving images to supply missing visual cues and integrating them into the reasoning process. However, selecting the right images and integrating them effectively into the model's reasoning remains challenging. To address this challenge, we propose R3G, a modular Reasoning-Retrieval-Reranking framework. It first produces a brief reasoning plan that specifies the required visual cues, then adopts a two-stage strategy, with coarse retrieval followed by fine-grained reranking, to select evidence images. On MRAG-Bench, R3G improves accuracy across six MLLM backbones and nine sub-scenarios, achieving state-of-the-art overall performance. Ablations show that sufficiency-aware reranking and reasoning steps are complementary, helping the model both choose the right images and use them well. We release code and data at https://github.com/czh24/R3G. --- ### Autonomous Multi-Agent AI for High-Throughput Polymer Informatics: From Property Prediction to Generative Design Across Synthetic and Bio-Polymers Authors: Mahule Roy, Adib Bazgir, Arthur da Silva Sousa Santos, Yuwen Zhang Published: 2026-01-25 Categories: cond-mat.soft, cond-mat.mtrl-sci, cs.AI arXiv: https://arxiv.org/abs/2602.00103v1 We present an integrated multiagent AI ecosystem for polymer discovery that unifies high-throughput materials workflows, artificial intelligence, and computational modeling within a single Polymer Research Lifecycle (PRL) pipeline. The system orchestrates specialized agents powered by state-of-the-art large language models (DeepSeek-V2 and DeepSeek-Coder) to retrieve and reason over scientific resources, invoke external tools, execute domain-specific code, and perform metacognitive self-assessment for robust end-to-end task execution. We demonstrate three practical capabilities: a high-fidelity polymer property prediction and generative design pipeline, a fully automated multimodal workflow for biopolymer structure characterization, and a metacognitive agent framework that can monitor performance and improve execution strategies over time. On a held-out test set of 1,251 polymers, our PolyGNN agent achieves strong predictive accuracy, reaching R2 = 0.89 for glass-transition temperature (Tg ), R2 = 0.82 for tensile strength, R2 = 0.75 for elongation, and R2 = 0.91 for density. The framework also provides uncertainty estimates via multiagent consensus and scales with linear complexity to at least 10,000 polymers, enabling high-throughput screening at low computational cost. For a representative workload, the system completes inference in 16.3 s using about 2 GB of memory and 0.1 GPU hours, at an estimated cost of about $0.08. On a dedicated Tg benchmark, our approach attains R2 = 0.78, outperforming strong baselines including single-LLM prediction (R2 = 0.67), group-contribution methods (R2 = 0.71), and ChemCrow (R2 = 0.66). We further demonstrate metacognitive control in a polystyrene case study, where the system not only produces domain-level scientific outputs but continually monitors and optimizes its own behavior through tactical, strategic, and meta-strategic self-assessment. --- ### RIR-Mega-Speech: A Reverberant Speech Corpus with Comprehensive Acoustic Metadata and Reproducible Evaluation Authors: Mandip Goswami Published: 2026-01-25 Categories: eess.AS, cs.CL, cs.SD, eess.SP arXiv: https://arxiv.org/abs/2601.19949v1 Despite decades of research on reverberant speech, comparing methods remains difficult because most corpora lack per-file acoustic annotations or provide limited documentation for reproduction. We present RIR-Mega-Speech, a corpus of approximately 117.5 hours created by convolving LibriSpeech utterances with roughly 5,000 simulated room impulse responses from the RIR-Mega collection. Every file includes RT60, direct-to-reverberant ratio (DRR), and clarity index ($C_{50}$) computed from the source RIR using clearly defined, reproducible procedures. We also provide scripts to rebuild the dataset and reproduce all evaluation results. Using Whisper small on 1,500 paired utterances, we measure 5.20% WER (95% CI: 4.69--5.78) on clean speech and 7.70% (7.04--8.35) on reverberant versions, corresponding to a paired increase of 2.50 percentage points (2.06--2.98). This represents a 48% relative degradation. WER increases monotonically with RT60 and decreases with DRR, consistent with prior perceptual studies. While the core finding that reverberation harms recognition is well established, we aim to provide the community with a standardized resource where acoustic conditions are transparent and results can be verified independently. The repository includes one-command rebuild instructions for both Windows and Linux environments. --- ### Time series forecasting with Hahn Kolmogorov-Arnold networks Authors: Md Zahidul Hasan, A. Ben Hamza, Nizar Bouguila Published: 2026-01-25 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2601.18837v2 Recent Transformer- and MLP-based models have demonstrated strong performance in long-term time series forecasting, yet Transformers remain limited by their quadratic complexity and permutation-equivariant attention, while MLPs exhibit spectral bias. We propose HaKAN, a versatile model based on Kolmogorov-Arnold Networks (KANs), leveraging Hahn polynomial-based learnable activation functions and providing a lightweight and interpretable alternative for multivariate time series forecasting. Our model integrates channel independence, patching, a stack of Hahn-KAN blocks with residual connections, and a bottleneck structure comprised of two fully connected layers. The Hahn-KAN block consists of inter- and intra-patch KAN layers to effectively capture both global and local temporal patterns. Extensive experiments on various forecasting benchmarks demonstrate that our model consistently outperforms recent state-of-the-art methods, with ablation studies validating the effectiveness of its core components. --- ### CanaryBench: Stress Testing Privacy Leakage in Cluster-Level Conversation Summaries Authors: Deep Mehta Published: 2026-01-25 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2601.18834v1 Aggregate analytics over conversational data are increasingly used for safety monitoring, governance, and product analysis in large language model systems. A common practice is to embed conversations, cluster them, and publish short textual summaries describing each cluster. While raw conversations may never be exposed, these derived summaries can still pose privacy risks if they contain personally identifying information (PII) or uniquely traceable strings copied from individual conversations. We introduce CanaryBench, a simple and reproducible stress test for privacy leakage in cluster-level conversation summaries. CanaryBench generates synthetic conversations with planted secret strings ("canaries") that simulate sensitive identifiers. Because canaries are known a priori, any appearance of these strings in published summaries constitutes a measurable leak. Using TF-IDF embeddings and k-means clustering on 3,000 synthetic conversations (24 topics) with a canary injection rate of 0.60, we evaluate an intentionally extractive example snippet summarizer that models quote-like reporting. In this configuration, we observe canary leakage in 50 of 52 canary-containing clusters (cluster-level leakage rate 0.961538), along with nonzero regex-based PII indicator counts. A minimal defense combining a minimum cluster-size publication threshold (k-min = 25) and regex-based redaction eliminates measured canary leakage and PII indicator hits in the reported run while maintaining a similar cluster-coherence proxy. We position this work as a societal impacts contribution centered on privacy risk measurement for published analytics artifacts rather than raw user data. --- ### Agentic Business Process Management Systems Authors: Marlon Dumas, Fredrik Milani, David Chapela-Campa Published: 2026-01-25 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2601.18833v1 Since the early 90s, the evolution of the Business Process Management (BPM) discipline has been punctuated by successive waves of automation technologies. Some of these technologies enable the automation of individual tasks, while others focus on orchestrating the execution of end-to-end processes. The rise of Generative and Agentic Artificial Intelligence (AI) is opening the way for another such wave. However, this wave is poised to be different because it shifts the focus from automation to autonomy and from design-driven management of business processes to data-driven management, leveraging process mining techniques. This position paper, based on a keynote talk at the 2025 Workshop on AI for BPM, outlines how process mining has laid the foundations on top of which agents can sense process states, reason about improvement opportunities, and act to maintain and optimize performance. The paper proposes an architectural vision for Agentic Business Process Management Systems (A-BPMS): a new class of platforms that integrate autonomy, reasoning, and learning into process management and execution. The paper contends that such systems must support a continuum of processes, spanning from human-driven to fully autonomous, thus redefining the boundaries of process automation and governance. --- ### The Geometric Reasoner: Manifold-Informed Latent Foresight Search for Long-Context Reasoning Authors: Ren Zhuang, Ben Wang, Shuifa Sun Published: 2026-01-25 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.18832v3 Scaling test-time compute enhances long chain-of-thought (CoT) reasoning, yet existing approaches face a fundamental trade-off between computational cost and coverage quality: either incurring high training expense or yielding redundant trajectories. We introduce The Geometric Reasoner (TGR), a training-free framework that performs manifold-informed latent foresight search under strict memory bounds. At each chunk boundary, TGR scores candidate latent anchors via a lightweight look-ahead estimate combined with soft geometric regularizers that encourage smooth trajectories and diverse exploration. Chunk-wise KV cache resets keep memory linear in chunk length. On challenging math and code benchmarks, TGR improves robust trajectory coverage, measured by the area under the Pass@k curve (AUC), by up to 13 points on Qwen3-8B, with negligible overhead of about 1.1--1.3 times. --- ### How Much Temporal Modeling is Enough? A Systematic Study of Hybrid CNN-RNN Architectures for Multi-Label ECG Classification Authors: Alireza Jafari, Fatemeh Jafari Published: 2026-01-25 Categories: cs.LG, eess.SP arXiv: https://arxiv.org/abs/2601.18830v1 Accurate multi-label classification of electrocardiogram (ECG) signals remains challenging due to the coexistence of multiple cardiac conditions, pronounced class imbalance, and long-range temporal dependencies in multi-lead recordings. Although recent studies increasingly rely on deep and stacked recurrent architectures, the necessity and clinical justification of such architectural complexity have not been rigorously examined. In this work, we perform a systematic comparative evaluation of convolutional neural networks (CNNs) combined with multiple recurrent configurations, including LSTM, GRU, Bidirectional LSTM (BiLSTM), and their stacked variants, for multi-label ECG classification on the PTB-XL dataset comprising 23 diagnostic categories. The CNN component serves as a morphology-driven baseline, while recurrent layers are progressively integrated to assess their contribution to temporal modeling and generalization performance. Experimental results indicate that a CNN integrated with a single BiLSTM layer achieves the most favorable trade-off between predictive performance and model complexity. This configuration attains superior Hamming loss (0.0338), macro-AUPRC (0.4715), micro-F1 score (0.6979), and subset accuracy (0.5723) compared with deeper recurrent combinations. Although stacked recurrent models occasionally improve recall for specific rare classes, our results provide empirical evidence that increasing recurrent depth yields diminishing returns and may degrade generalization due to reduced precision and overfitting. These findings suggest that architectural alignment with the intrinsic temporal structure of ECG signals, rather than increased recurrent depth, is a key determinant of robust performance and clinically relevant deployment. --- ### CP Loss: Channel-wise Perceptual Loss for Time Series Forecasting Authors: Yaohua Zha, Chunlin Fan, Peiyuan Liu, Yong Jiang, Tao Dai, Hai Wu, Shu-Tao Xia Published: 2026-01-25 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.18829v1 Multi-channel time-series data, prevalent across diverse applications, is characterized by significant heterogeneity in its different channels. However, existing forecasting models are typically guided by channel-agnostic loss functions like MSE, which apply a uniform metric across all channels. This often leads to fail to capture channel-specific dynamics such as sharp fluctuations or trend shifts. To address this, we propose a Channel-wise Perceptual Loss (CP Loss). Its core idea is to learn a unique perceptual space for each channel that is adapted to its characteristics, and to compute the loss within this space. Specifically, we first design a learnable channel-wise filter that decomposes the raw signal into disentangled multi-scale representations, which form the basis of our perceptual space. Crucially, the filter is optimized jointly with the main forecasting model, ensuring that the learned perceptual space is explicitly oriented towards the prediction task. Finally, losses are calculated within these perception spaces to optimize the model. Code is available at https://github.com/zyh16143998882/CP_Loss. --- ### IPBC: An Interactive Projection-Based Framework for Human-in-the-Loop Semi-Supervised Clustering of High-Dimensional Data Authors: Mohammad Zare Published: 2026-01-25 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.18828v1 High-dimensional datasets are increasingly common across scientific and industrial domains, yet they remain difficult to cluster effectively due to the diminishing usefulness of distance metrics and the tendency of clusters to collapse or overlap when projected into lower dimensions. Traditional dimensionality reduction techniques generate static 2D or 3D embeddings that provide limited interpretability and do not offer a mechanism to leverage the analyst's intuition during exploration. To address this gap, we propose Interactive Project-Based Clustering (IPBC), a framework that reframes clustering as an iterative human-guided visual analysis process. IPBC integrates a nonlinear projection module with a feedback loop that allows users to modify the embedding by adjusting viewing angles and supplying simple constraints such as must-link or cannot-link relationships. These constraints reshape the objective of the projection model, gradually pulling semantically related points closer together and pushing unrelated points further apart. As the projection becomes more structured and expressive through user interaction, a conventional clustering algorithm operating on the optimized 2D layout can more reliably identify distinct groups. An additional explainability component then maps each discovered cluster back to the original feature space, producing interpretable rules or feature rankings that highlight what distinguishes each cluster. Experiments on various benchmark datasets show that only a small number of interactive refinement steps can substantially improve cluster quality. Overall, IPBC turns clustering into a collaborative discovery process in which machine representation and human insight reinforce one another. --- ### Automated structural testing of LLM-based agents: methods, framework, and case studies Authors: Jens Kohl, Otto Kruse, Youssef Mostafa, Andre Luckow, Karsten Schroer, Thomas Riedl, Ryan French, David Katz, Manuel P. Luitz, Tanrajbir Takher, Ken E. Friedl, Céline Laurent-Winter Published: 2026-01-25 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2601.18827v1 LLM-based agents are rapidly being adopted across diverse domains. Since they interact with users without supervision, they must be tested extensively. Current testing approaches focus on acceptance-level evaluation from the user's perspective. While intuitive, these tests require manual evaluation, are difficult to automate, do not facilitate root cause analysis, and incur expensive test environments. In this paper, we present methods to enable structural testing of LLM-based agents. Our approach utilizes traces (based on OpenTelemetry) to capture agent trajectories, employs mocking to enforce reproducible LLM behavior, and adds assertions to automate test verification. This enables testing agent components and interactions at a deeper technical level within automated workflows. We demonstrate how structural testing enables the adaptation of software engineering best practices to agents, including the test automation pyramid, regression testing, test-driven development, and multi-language testing. In representative case studies, we demonstrate automated execution and faster root-cause analysis. Collectively, these methods reduce testing costs and improve agent quality through higher coverage, reusability, and earlier defect detection. We provide an open source reference implementation on GitHub. --- ### Differential Voting: Loss Functions For Axiomatically Diverse Aggregation of Heterogeneous Preferences Authors: Zhiyu An, Duaa Nakshbandi, Wan Du Published: 2026-01-25 Categories: cs.GT, cs.AI arXiv: https://arxiv.org/abs/2601.18824v1 Reinforcement learning from human feedback (RLHF) implicitly aggregates heterogeneous human preferences into a single utility function, even though the underlying utilities of the participants are in practice diverse. Hence, RLHF can be viewed as a form of voting, where the aggregation mechanism is defined by the loss function. Although Arrow's Impossibility Theorem suggests that different mechanisms satisfy different sets of desirable axioms, most existing methods rely on a single aggregation principle, typically the Bradley-Terry-Luce (BTL) model, which corresponds to Borda count voting. This restricts the axiomatic properties of the learned reward and obscures the normative assumptions embedded in optimization. In this work, we introduce Differential Voting, a unifying framework that constructs instance-wise, differentiable loss functions whose population-level optima provably correspond to distinct classical voting rules. We develop differentiable surrogates for majority-based aggregation (BTL), Copeland, and Kemeny rules, and formally analyze their calibration properties, gradient fields, and limiting behavior as smoothing parameters vanish. For each loss, we establish consistency with the corresponding social choice rule and characterize the axioms it satisfies or violates. Our analysis shows how design choices in loss geometry-such as margin sensitivity and boundary concentration-directly translate into normative aggregation behavior. Differential Voting makes preference aggregation an explicit and controllable design choice in RLHF, enabling principled trade-offs between axiomatic guarantees and optimization stability. Code to reproduce our experiments is open-sourced. --- ### VAE with Hyperspherical Coordinates: Improving Anomaly Detection from Hypervolume-Compressed Latent Space Authors: Alejandro Ascarate, Leo Lebrat, Rodrigo Santa Cruz, Clinton Fookes, Olivier Salvado Published: 2026-01-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18823v4 Variational autoencoders (VAE) encode data into lower-dimensional latent vectors before decoding those vectors back to data. Once trained, one can hope to detect out-of-distribution (abnormal) latent vectors, but several issues arise when the latent space is high dimensional. This includes an exponential growth of the hypervolume with the dimension, which severely affects the generative capacity of the VAE. In this paper, we draw insights from high dimensional statistics: in these regimes, the latent vectors of a standard VAE are distributed on the `equators' of a hypersphere, challenging the detection of anomalies. We propose to formulate the latent variables of a VAE using hyperspherical coordinates, which allows compressing the latent vectors towards a given direction on the hypersphere, thereby allowing for a more expressive approximate posterior. We show that this improves both the fully unconditional-OOD and conditional-OOD anomaly detection ability of the VAE, achieving the best performance on the datasets we considered, outperforming existing methods. For the unconditional-OOD and conditional-OOD modalities, respectively, these are: i) detecting unusual landscape from the Mars Rover camera and unusual Galaxies from ground based imagery (complex, real world datasets); ii) standard benchmarks like Cifar10 and subsets of ImageNet as the in-distribution (ID) class. --- ### Leveraging Persistence Image to Enhance Robustness and Performance in Curvilinear Structure Segmentation Authors: Zhuangzhi Gao, Feixiang Zhou, He Zhao, Xiuju Chen, Xiaoxin Li, Qinkai Yu, Yitian Zhao, Alena Shantsila, Gregory Y. H. Lip, Eduard Shantsila, Yalin Zheng Published: 2026-01-25 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2601.18045v1 Segmenting curvilinear structures in medical images is essential for analyzing morphological patterns in clinical applications. Integrating topological properties, such as connectivity, improves segmentation accuracy and consistency. However, extracting and embedding such properties - especially from Persistence Diagrams (PD) - is challenging due to their non-differentiability and computational cost. Existing approaches mostly encode topology through handcrafted loss functions, which generalize poorly across tasks. In this paper, we propose PIs-Regressor, a simple yet effective module that learns persistence image (PI) - finite, differentiable representations of topological features - directly from data. Together with Topology SegNet, which fuses these features in both downsampling and upsampling stages, our framework integrates topology into the network architecture itself rather than auxiliary losses. Unlike existing methods that depend heavily on handcrafted loss functions, our approach directly incorporates topological information into the network structure, leading to more robust segmentation. Our design is flexible and can be seamlessly combined with other topology-based methods to further enhance segmentation performance. Experimental results show that integrating topological features enhances model robustness, effectively handling challenges like overexposure and blurring in medical imaging. Our approach on three curvilinear benchmarks demonstrate state-of-the-art performance in both pixel-level accuracy and topological fidelity. --- ### SpatialEmb: Extract and Encode Spatial Information for 1-Stage Multi-channel Multi-speaker ASR on Arbitrary Microphone Arrays Authors: Yiwen Shao, Yong Xu, Sanjeev Khudanpur, Dong Yu Published: 2026-01-25 Categories: eess.AS, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2601.18037v1 Spatial information is a critical clue for multi-channel multi-speaker target speech recognition. Most state-of-the-art multi-channel Automatic Speech Recognition (ASR) systems extract spatial features only during the speech separation stage, followed by standard single-channel ASR on the separated speech. This approach results in an inefficient, lengthy pipeline and sub-optimal ASR performance due to the accumulated errors from preprocessing modules. Furthermore, most spatial feature extraction methods depend on the knowledge of speaker positions and microphone topology, making the systems reliant on specific settings and challenging to adapt to new equipment. In this work, we propose a solution to these issues with a lightweight embedding module named SpatialEmb, which extracts and encodes spatial information directly for the ASR model, supporting both fixed and arbitrary microphone topology. We conduct comprehensive experiments on AliMeeting, a real meeting corpus, to determine the optimal model design for SpatialEmb in terms of both performance and efficiency. Our best model trained with 105 hours Train-Ali-far achieves 17.04% and 20.32% character error rates (CER) on the Eval and Test sets, establishing a new state-of-the-art result with the same training data. --- ### An Experimental Comparison of Cognitive Forcing Functions for Execution Plans in AI-Assisted Writing: Effects On Trust, Overreliance, and Perceived Critical Thinking Authors: Ahana Ghosh, Advait Sarkar, Siân Lindley, Christian Poelitz Published: 2026-01-25 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2601.18033v1 Generative AI (GenAI) tools improve productivity in knowledge workflows such as writing, but also risk overreliance and reduced critical thinking. Cognitive forcing functions (CFFs) mitigate these risks by requiring active engagement with AI output. As GenAI workflows grow more complex, systems increasingly present execution plans for user review. However, these plans are themselves AI-generated and prone to overreliance, and the effectiveness of applying CFFs to AI plans remains underexplored. We conduct a controlled experiment in which participants completed AI-assisted writing tasks while reviewing AI-generated plans under four CFF conditions: Assumption (argument analysis), WhatIf (hypothesis testing), Both, and a no-CFF control. A follow-up think-aloud and interview study qualitatively compared these conditions. Results show that the Assumption CFF most effectively reduced overreliance without increasing cognitive load, while participants perceived the WhatIf CFF as most helpful. These findings highlight the value of plan-focused CFFs for supporting critical reflection in GenAI-assisted knowledge work. --- ### Multimodal Machine Learning for Soft High-k Elastomers under Data Scarcity Authors: Brijesh FNU, Viet Thanh Duy Nguyen, Ashima Sharma, Md Harun Rashid Molla, Chengyi Xu, Truong-Son Hy Published: 2026-01-25 Categories: cs.LG, cond-mat.mtrl-sci arXiv: https://arxiv.org/abs/2601.18032v2 Dielectric materials are critical building blocks for modern electronics such as sensors, actuators, and transistors. With rapid advances in soft and stretchable electronics for emerging human- and robot-interfacing applications, there is a growing need for high-performance dielectric elastomers. However, developing soft elastomers that simultaneously exhibit high dielectric constants (k) and low Young's moduli (E) remains a major challenge. Although individual elastomer designs have been reported, structured datasets that systematically integrate molecular sequence, dielectric, and mechanical properties are largely unavailable. To address this gap, we curate a compact, high-quality dataset of acrylate-based dielectric elastomers by aggregating experimental results from the past decade. Building on this dataset, we propose a multimodal learning framework leveraging large-scale pretrained polymer representations. These pretrained embeddings transfer chemical and structural knowledge from vast polymer corpora, enabling accurate few-shot prediction of dielectric and mechanical properties and accelerating data-efficient discovery of soft high-$k$ dielectric elastomers. Our data and implementation are publicly available at: https://github.com/HySonLab/Polymers --- ### Spelling Bee Embeddings for Language Modeling Authors: Markus N. Rabe, Judith Clymo, Zheren Dong Published: 2026-01-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.18030v1 We introduce a simple modification to the embedding layer. The key change is to infuse token embeddings with information about their spelling. Models trained with these embeddings improve not only on spelling, but also across standard benchmarks. We conduct scaling studies for models with 40M to 800M parameters, which suggest that the improvements are equivalent to needing about 8% less compute and data to achieve the same test loss. --- ### Sentipolis: Emotion-Aware Agents for Social Simulations Authors: Chiyuan Fu, Lyuhao Chen, Yunze Xiao, Weihao Xuan, Carlos Busso, Mona Diab Published: 2026-01-25 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.18027v2 LLM agents are increasingly used for social simulation, yet emotion is often treated as a transient cue, causing emotional amnesia and weak long-horizon continuity. We present Sentipolis, a framework for emotionally stateful agents that integrates continuous Pleasure-Arousal-Dominance (PAD) representation, dual-speed emotion dynamics, and emotion--memory coupling. Across thousands of interactions over multiple base models and evaluators, Sentipolis improves emotionally grounded behavior, boosting communication, and emotional continuity. Gains are model-dependent: believability increases for higher-capacity models but can drop for smaller ones, and emotion-awareness can mildly reduce adherence to social norms, reflecting a human-like tension between emotion-driven behavior and rule compliance in social simulation. Network-level diagnostics show reciprocal, moderately clustered, and temporally stable relationship structures, supporting the study of cumulative social dynamics such as alliance formation and gradual relationship change. --- ### CommonLID: Re-evaluating State-of-the-Art Language Identification Performance on Web Data Authors: Pedro Ortiz Suarez, Laurie Burchell, Catherine Arnett, Rafael Mosquera-Gómez, Sara Hincapie-Monsalve, Thom Vaughan, Damian Stewart, Malte Ostendorff, Idris Abdulmumin, Vukosi Marivate, Shamsuddeen Hassan Muhammad, Atnafu Lambebo Tonja et al. Published: 2026-01-25 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.18026v2 Language identification (LID) is a fundamental step in curating multilingual corpora. However, LID models still perform poorly for many languages, especially on the noisy and heterogeneous web data often used to train multilingual language models. In this paper, we introduce CommonLID, a community-driven, human-annotated LID benchmark for the web domain, covering 109 languages. Many of the included languages have been previously under-served, making CommonLID a key resource for developing more representative high-quality text corpora. We show CommonLID's value by using it, alongside five other common evaluation sets, to test eight popular LID models. We analyse our results to situate our contribution and to provide an overview of the state of the art. In particular, we highlight that existing evaluations overestimate LID accuracy for many languages in the web domain. We make CommonLID and the code used to create it available under an open, permissive license. --- ### MlPET: A Localized Neural Network Approach for Probabilistic Post-Reconstruction PET Image Analysis Using Informed Priors Authors: Thomas Mejer Hansen, Nana Christensen, Mikkel Vendelbo Published: 2026-01-25 Categories: physics.med-ph, cs.LG arXiv: https://arxiv.org/abs/2601.18021v1 We develop and evaluate MlPET, a fast localized machine learning approach for probabilistic PET image analysis addressing the noise-resolution trade-off in conventional reconstructions. MlPET replaces computationally demanding Markov chain Monte Carlo sampling with a localized neural network trained to estimate posterior mean voxel activity from small image neighborhoods. The method incorporates scanner-specific point spread functions, spatially correlated noise modeling, and flexible priors. Performance was evaluated on NEMA IEC phantom data from three PET systems (GE Discovery MI, Siemens Biograph Vision 600, and Quadra) under varying reconstruction settings and acquisition times. On phantom data, MlPET achieved contrast recovery coefficients consistently higher than standard PET and close to 1.0 (including 10 mm spheres), while reducing background noise and improving spatial definition. Effective pointspread function full width at half maximum decreased from approximately 2 mm in standard PET to below 1 mm with MlPET, a 2.5 fold reduction in blur. Comparable image quality was obtained at 40-80 s acquisition time with MlPET versus 900 s with conventional PET. MlPET provides an efficient approach for quantitative probabilistic post-reconstruction PET analysis. By combining informed priors with neural network speed, it achieves noise suppression and resolution enhancement without altering reconstruction algorithms. The method shows promise for improved small-lesion detectability and quantitative reliability in clinical PET imaging. Future studies will evaluate performance on patient data. --- ### A System for Name and Address Parsing with Large Language Models Authors: Adeeba Tarannum, Muzakkiruddin Ahmed Mohammed, Mert Can Cakmak, Shames Al Mandalawi, John Talburt Published: 2026-01-25 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.18014v1 Reliable transformation of unstructured person and address text into structured data remains a key challenge in large-scale information systems. Traditional rule-based and probabilistic approaches perform well on clean inputs but fail under noisy or multilingual conditions, while neural and large language models (LLMs) often lack deterministic control and reproducibility. This paper introduces a prompt-driven, validation-centered framework that converts free-text records into a consistent 17-field schema without fine-tuning. The method integrates input normalisation, structured prompting, constrained decoding, and strict rule-based validation under fixed experimental settings to ensure reproducibility. Evaluations on heterogeneous real-world address data show high field-level accuracy, strong schema adherence, and stable confidence calibration. The results demonstrate that combining deterministic validation with generative prompting provides a robust, interpretable, and scalable solution for structured information extraction, offering a practical alternative to training-heavy or domain-specific models. --- ### Evaluating Semantic and Syntactic Understanding in Large Language Models for Payroll Systems Authors: Hendrika Maclean, Mert Can Cakmak, Muzakkiruddin Ahmed Mohammed, Shames Al Mandalawi, John Talburt Published: 2026-01-25 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.18012v2 Large language models are now used daily for writing, search, and analysis, and their natural language understanding continues to improve. However, they remain unreliable on exact numerical calculation and on producing outputs that are straightforward to audit. We study synthetic payroll system as a focused, high-stakes example and evaluate whether models can understand a payroll schema, apply rules in the right order, and deliver cent-accurate results. Our experiments span a tiered dataset from basic to complex cases, a spectrum of prompts from minimal baselines to schema-guided and reasoning variants, and multiple model families including GPT, Claude, Perplexity, Grok and Gemini. Results indicate clear regimes where careful prompting is sufficient and regimes where explicit computation is required. The work offers a compact, reproducible framework and practical guidance for deploying LLMs in settings that demand both accuracy and assurance. --- ### Strip-Fusion: Spatiotemporal Fusion for Multispectral Pedestrian Detection Authors: Asiegbu Miracle Kanu-Asiegbu, Nitin Jotwani, Xiaoxiao Du Published: 2026-01-25 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.18008v1 Pedestrian detection is a critical task in robot perception. Multispectral modalities (visible light and thermal) can boost pedestrian detection performance by providing complementary visual information. Several gaps remain with multispectral pedestrian detection methods. First, existing approaches primarily focus on spatial fusion and often neglect temporal information. Second, RGB and thermal image pairs in multispectral benchmarks may not always be perfectly aligned. Pedestrians are also challenging to detect due to varying lighting conditions, occlusion, etc. This work proposes Strip-Fusion, a spatial-temporal fusion network that is robust to misalignment in input images, as well as varying lighting conditions and heavy occlusions. The Strip-Fusion pipeline integrates temporally adaptive convolutions to dynamically weigh spatial-temporal features, enabling our model to better capture pedestrian motion and context over time. A novel Kullback-Leibler divergence loss was designed to mitigate modality imbalance between visible and thermal inputs, guiding feature alignment toward the more informative modality during training. Furthermore, a novel post-processing algorithm was developed to reduce false positives. Extensive experimental results show that our method performs competitively for both the KAIST and the CVC-14 benchmarks. We also observed significant improvements compared to previous state-of-the-art on challenging conditions such as heavy occlusion and misalignment. ---