# Research Papers — 2025-10 · bundle page 116 of 217

> Full-text markdown of papers #5751–5800 of 10843 published in 2025-10, newest first. AI-readable mirrors of the full 10843-paper 2025-10 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2025-10/page/116
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/115.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/117.md)

### Deliberate Lab: A Platform for Real-Time Human-AI Social Experiments

Authors: Crystal Qian, Vivian Tsai, Michael Behr, Nada Hussein, Léo Laugier, Nithum Thain, Lucas Dixon
Published: 2025-10-14
Categories: cs.HC, cs.AI
arXiv: https://arxiv.org/abs/2510.13011v1

Social and behavioral scientists increasingly aim to study how humans interact, collaborate, and make decisions alongside artificial intelligence. However, the experimental infrastructure for such work remains underdeveloped: (1) few platforms support real-time, multi-party studies at scale; (2) most deployments require bespoke engineering, limiting replicability and accessibility, and (3) existing tools do not treat AI agents as first-class participants. We present Deliberate Lab, an open-source platform for large-scale, real-time behavioral experiments that supports both human participants and large language model (LLM)-based agents. We report on a 12-month public deployment of the platform (N=88 experimenters, N=9195 experiment participants), analyzing usage patterns and workflows. Case studies and usage scenarios are aggregated from platform users, complemented by in-depth interviews with select experimenters. By lowering technical barriers and standardizing support for hybrid human-AI experimentation, Deliberate Lab expands the methodological repertoire for studying collective decision-making and human-centered AI.

---

### Developing and Validating the Arabic Version of the Attitudes Toward Large Language Models Scale

Authors: Basad Barajeeh, Ala Yankouskaya, Sameha AlShakhsi, Chun Sing Maxwell Ho, Guandong Xu, Raian Ali
Published: 2025-10-14
Categories: cs.HC, cs.AI
arXiv: https://arxiv.org/abs/2510.13009v1

As the use of large language models (LLMs) becomes increasingly global, understanding public attitudes toward these systems requires tools that are adapted to local contexts and languages. In the Arab world, LLM adoption has grown rapidly with both globally dominant platforms and regional ones like Fanar and Jais offering Arabic-specific solutions. This highlights the need for culturally and linguistically relevant scales to accurately measure attitudes toward LLMs in the region. Tools assessing attitudes toward artificial intelligence (AI) can provide a base for measuring attitudes specific to LLMs. The 5-item Attitudes Toward Artificial Intelligence (ATAI) scale, which measures two dimensions, the AI Fear and the AI Acceptance, has been recently adopted and adapted to develop new instruments in English using a sample from the UK: the Attitudes Toward General LLMs (AT-GLLM) and Attitudes Toward Primary LLM (AT-PLLM) scales. In this paper, we translate the two scales, AT-GLLM and AT-PLLM, and validate them using a sample of 249 Arabic-speaking adults. The results show that the scale, translated into Arabic, is a reliable and valid tool that can be used for the Arab population and language. Psychometric analyses confirmed a two-factor structure, strong measurement invariance across genders, and good internal reliability. The scales also demonstrated strong convergent and discriminant validity. Our scales will support research in a non-Western context, a much-needed effort to help draw a global picture of LLM perceptions, and will also facilitate localized research and policy-making in the Arab region.

---

### CurLL: A Developmental Framework to Evaluate Continual Learning in Language Models

Authors: Pavan Kalyan, Shubhra Mishra, Satya Lokam, Navin Goyal
Published: 2025-10-14
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.13008v1

We introduce a comprehensive continual learning dataset and benchmark (CurlL) grounded in human developmental trajectories from ages 5-10, enabling systematic and fine-grained assessment of models' ability to progressively acquire new skills. CurlL spans five developmental stages (0-4) covering ages 5-10, supported by a skill graph that breaks down broad skills into smaller abilities, concrete goals, and measurable indicators, while also capturing which abilities build on others. We generate a 23.4B-token synthetic dataset with controlled skill progression, vocabulary complexity, and format diversity, comprising paragraphs, comprehension-based QA (CQA), skill-testing QA (CSQA), and instruction-response (IR) pairs. Stage-wise token counts range from 2.12B to 6.78B tokens, supporting precise analysis of forgetting, forward transfer, and backward transfer. Using a 135M-parameter transformer trained under independent, joint, and sequential (continual) setups, we show trade-offs in skill retention and transfer efficiency. By mirroring human learning patterns and providing fine-grained control over skill dependencies, this work advances continual learning evaluations for language models.

---

### What is Implementation Science; and Why It Matters for Bridging the Artificial Intelligence Innovation-to-Application Gap in Medical Imaging

Authors: Ahmad Fayaz-Bakhsh, Janice Tania, Syaheerah Lebai Lutfi, Abhinav K. Jha, Arman Rahmim
Published: 2025-10-14
Categories: physics.med-ph, cs.AI
arXiv: https://arxiv.org/abs/2510.13006v3

The transformative potential of artificial intelligence (AI) in medical Imaging (MI) is well recognized. Yet despite promising reports in research settings, many AI tools fail to achieve clinical adoption in practice. In fact, more generally, there is a documented 17-year average delay between evidence generation and implementation of a technology. Implementation science (IS) may provide a practical, evidence-based framework to bridge the gap between AI development and real-world clinical imaging use, to shorten this lag through systematic frameworks, strategies, and hybrid research designs. We outline challenges specific to AI adoption in MI workflows, including infrastructural, educational, and cultural barriers. We highlight the complementary roles of effectiveness research and implementation research, emphasizing hybrid study designs and the role of integrated KT (iKT), stakeholder engagement, and equity-focused co-creation in designing sustainable and generalizable solutions. We discuss integration of Human-Computer Interaction (HCI) frameworks in MI towards usable AI. Adopting IS is not only a methodological advancement; it is a strategic imperative for accelerating translation of innovation into improved patient outcomes.

---

### OPLoRA: Orthogonal Projection LoRA Prevents Catastrophic Forgetting during Parameter-Efficient Fine-Tuning

Authors: Yifeng Xiong, Xiaohui Xie
Published: 2025-10-14
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.13003v2

Low-Rank Adaptation (LoRA) enables efficient fine-tuning of large language models but suffers from catastrophic forgetting when learned updates interfere with the dominant singular directions that encode essential pre-trained knowledge. We propose Orthogonal Projection LoRA (OPLoRA), a theoretically grounded approach that prevents this interference through double-sided orthogonal projections. By decomposing frozen weights via SVD, OPLoRA constrains LoRA updates to lie entirely within the orthogonal complement of the top-$k$ singular subspace using projections $P_L = I - U_k U_k^\top$ and $P_R = I - V_k V_k^\top$. We prove that this construction exactly preserves the top-$k$ singular triples, providing mathematical guarantees for knowledge retention. To quantify subspace interference, we introduce $ρ_k$, a metric measuring update alignment with dominant directions. Extensive experiments across commonsense reasoning, mathematics, and code generation demonstrate that OPLoRA significantly reduces forgetting while maintaining competitive task-specific performance on LLaMA-2 7B and Qwen2.5 7B, establishing orthogonal projection as an effective mechanism for knowledge preservation in parameter-efficient fine-tuning.

---

### From Narratives to Probabilistic Reasoning: Predicting and Interpreting Drivers' Hazardous Actions in Crashes Using Large Language Model

Authors: Boyou Chen, Gerui Xu, Zifei Wang, Huizhong Guo, Ananna Ahmed, Zhaonan Sun, Zhen Hu, Kaihan Zhang, Shan Bao
Published: 2025-10-14
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.13002v1

Vehicle crashes involve complex interactions between road users, split-second decisions, and challenging environmental conditions. Among these, two-vehicle crashes are the most prevalent, accounting for approximately 70% of roadway crashes and posing a significant challenge to traffic safety. Identifying Driver Hazardous Action (DHA) is essential for understanding crash causation, yet the reliability of DHA data in large-scale databases is limited by inconsistent and labor-intensive manual coding practices. Here, we present an innovative framework that leverages a fine-tuned large language model to automatically infer DHAs from textual crash narratives, thereby improving the validity and interpretability of DHA classifications. Using five years of two-vehicle crash data from MTCF, we fine-tuned the Llama 3.2 1B model on detailed crash narratives and benchmarked its performance against conventional machine learning classifiers, including Random Forest, XGBoost, CatBoost, and a neural network. The fine-tuned LLM achieved an overall accuracy of 80%, surpassing all baseline models and demonstrating pronounced improvements in scenarios with imbalanced data. To increase interpretability, we developed a probabilistic reasoning approach, analyzing model output shifts across original test sets and three targeted counterfactual scenarios: variations in driver distraction and age. Our analysis revealed that introducing distraction for one driver substantially increased the likelihood of "General Unsafe Driving"; distraction for both drivers maximized the probability of "Both Drivers Took Hazardous Actions"; and assigning a teen driver markedly elevated the probability of "Speed and Stopping Violations." Our framework and analytical methods provide a robust and interpretable solution for large-scale automated DHA detection, offering new opportunities for traffic safety analysis and intervention.

---

### Max It or Miss It: Benchmarking LLM On Solving Extremal Problems

Authors: Binxin Gao, Jingjun Han
Published: 2025-10-14
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.12997v2

Test-time scaling has enabled Large Language Models (LLMs) with remarkable reasoning capabilities, particularly in mathematical domains, through intermediate chain-of-thought (CoT) reasoning before generating final answers. However, the specific sources and mechanisms underlying these reasoning capabilities remain insufficiently understood. Optimization reasoning, i.e. finding extrema under constraints, represents a fundamental abstraction that underpins critical applications in planning, control, resource allocation, and prompt search. To systematically evaluate this capability, we introduce ExtremBench, a benchmark dataset for solving mathematical extremal problems, curated from inequality exercises used for Chinese Mathematical Olympiad and transformed into $93$ standardized extrema-finding problems. We conduct extensive evaluations across various state-of-the-art open-source model families, including the Qwen3, GPT-OSS, and DeepSeek. Our results reveal that LLMs' extremal-solving reasoning capabilities do not always align with those of current mathematical benchmarks such as AIME25 and MATH-500, with some models showing strong general mathematical reasoning but poor extremal-solving skills, and vice versa. This discrepancy highlights a critical gap in current evaluation practices and suggests that existing benchmarks may not comprehensively capture the full spectrum of mathematical reasoning abilities.

---

### CSI-4CAST: A Hybrid Deep Learning Model for CSI Prediction with Comprehensive Robustness and Generalization Testing

Authors: Sikai Cheng, Reza Zandehshahvar, Haoruo Zhao, Daniel A. Garcia-Ulloa, Alejandro Villena-Rodriguez, Carles Navarro Manchón, Pascal Van Hentenryck
Published: 2025-10-14
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.12996v2

Channel state information (CSI) prediction is a promising strategy for ensuring reliable and efficient operation of massive multiple-input multiple-output (mMIMO) systems by providing timely downlink (DL) CSI. While deep learning-based methods have advanced beyond conventional model-driven and statistical approaches, they remain limited in robustness to practical non-Gaussian noise, generalization across diverse channel conditions, and computational efficiency. This paper introduces CSI-4CAST, a hybrid deep learning architecture that integrates 4 key components, i.e., Convolutional neural network residuals, Adaptive correction layers, ShuffleNet blocks, and Transformers, to efficiently capture both local and long-range dependencies in CSI prediction. To enable rigorous evaluation, this work further presents a comprehensive benchmark, CSI-RRG for Regular, Robustness and Generalization testing, which includes more than 300,000 samples across 3,060 realistic scenarios for both TDD and FDD systems. The dataset spans multiple channel models, a wide range of delay spreads and user velocities, and diverse noise types and intensity degrees. Experimental results show that CSI-4CAST achieves superior prediction accuracy with substantially lower computational cost, outperforming baselines in 81.5% of TDD scenarios and 44.4% of FDD scenario, the best performance among all evaluated models, while reducing FLOPs by 5x and 3x compared to LLM4CP, the strongest baseline. In addition, evaluation over CSI-RRG provides valuable insights into how different channel factors affect the performance and generalization capability of deep learning models. Both the dataset (https://huggingface.co/CSI-4CAST) and evaluation protocols (https://github.com/AI4OPT/CSI-4CAST) are publicly released to establish a standardized benchmark and to encourage further research on robust and efficient CSI prediction.

---

### Deep Learning-Based Visual Fatigue Detection Using Eye Gaze Patterns in VR

Authors: Numan Zafar, Johnathan Locke, Shafique Ahmad Chaudhry
Published: 2025-10-14
Categories: cs.HC, cs.LG
arXiv: https://arxiv.org/abs/2510.12994v1

Prolonged exposure to virtual reality (VR) systems leads to visual fatigue, impairs user comfort, performance, and safety, particularly in high-stakes or long-duration applications. Existing fatigue detection approaches rely on subjective questionnaires or intrusive physiological signals, such as EEG, heart rate, or eye-blink count, which limit their scalability and real-time applicability. This paper introduces a deep learning-based study for detecting visual fatigue using continuous eye-gaze trajectories recorded in VR. We use the GazeBaseVR dataset comprising binocular eye-tracking data from 407 participants across five immersive tasks, extract cyclopean eye-gaze angles, and evaluate six deep classifiers. Our results demonstrate that EKYT achieves up to 94% accuracy, particularly in tasks demanding high visual attention, such as video viewing and text reading. We further analyze gaze variance and subjective fatigue measures, indicating significant behavioral differences between fatigued and non-fatigued conditions. These findings establish eye-gaze dynamics as a reliable and nonintrusive modality for continuous fatigue detection in immersive VR, offering practical implications for adaptive human-computer interactions.

---

### Tailored untruths: How personalisation challenges LLM safeguards

Authors: João A. Leite, João Luz, Silvia Gargova, Arnav Arora, Gustavo Sampaio, Ian Roberts, Carolina Scarton, Kalina Bontcheva
Published: 2025-10-14
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.12993v3

Large Language Models (LLMs) can generate highly persuasive disinformation, yet little is known about how effectively they personalise it across languages and demographic groups. We present the first large-scale multilingual study of persona-targeted disinformation generation by LLMs. Using a red-teaming methodology, we prompted eight leading models with 324 false narratives and 150 demographic personas in four languages (English, Russian, Portuguese, and Hindi), creating AI-TRAITS, a dataset of 1.6 million personalised disinformation texts. We treat safeguards as compromised whenever a model generates the requested falsehood, whether directly or accompanied by a safety disclaimer. Across models, safeguards failed for 80% of non-personalised prompts and 77.7% of personalised ones, with Grok producing disinformation in over 94% of cases. All models effectively tailored outputs to target personas, employing substantially more persuasive techniques than in non-personalised content. Additional analyses reveal persona-specific linguistic and psychological patterns and show that safeguard effectiveness varies markedly across languages. Together, these findings expose significant weaknesses in current LLM safety mechanisms and highlight the need for more robust, multilingual safeguards against personalised AI-generated disinformation.

---

### UNCAP: Uncertainty-Guided Neurosymbolic Planning Using Natural Language Communication for Cooperative Autonomous Vehicles

Authors: Neel P. Bhatt, Po-han Li, Kushagra Gupta, Rohan Siva, Daniel Milan, Alexander T. Hogue, Sandeep P. Chinchali, David Fridovich-Keil, Zhangyang Wang, Ufuk Topcu
Published: 2025-10-14
Categories: cs.RO, cs.CL, cs.CV, cs.MA
arXiv: https://arxiv.org/abs/2510.12992v2

Safe large-scale coordination of multiple cooperative connected autonomous vehicles (CAVs) hinges on communication that is both efficient and interpretable. Existing approaches either rely on transmitting high-bandwidth raw sensor data streams or neglect perception and planning uncertainties inherent in shared data, resulting in systems that are neither scalable nor safe. To address these limitations, we propose Uncertainty-Guided Natural Language Cooperative Autonomous Planning (UNCAP), a vision-language model-based planning approach that enables CAVs to communicate via lightweight natural language messages while explicitly accounting for perception uncertainty in decision-making. UNCAP features a two-stage communication protocol: (i) an ego CAV first identifies the subset of vehicles most relevant for information exchange, and (ii) the selected CAVs then transmit messages that quantitatively express their perception uncertainty. By selectively fusing messages that maximize mutual information, this strategy allows the ego vehicle to integrate only the most relevant signals into its decision-making, improving both the scalability and reliability of cooperative planning. Experiments across diverse driving scenarios show a 63% reduction in communication bandwidth with a 31% increase in driving safety score, a 61% reduction in decision uncertainty, and a four-fold increase in collision distance margin during near-miss events. Project website: https://uncap-project.github.io/

---

### Behavioral Biometrics for Automatic Detection of User Familiarity in VR

Authors: Numan Zafar, Priyo Ranjan Kundu Prosun, Shafique Ahmad Chaudhry
Published: 2025-10-14
Categories: cs.HC, cs.LG
arXiv: https://arxiv.org/abs/2510.12988v2

As virtual reality (VR) devices become increasingly integrated into everyday settings, a growing number of users without prior experience will engage with VR systems. Automatically detecting a user's familiarity with VR as an interaction medium enables real-time, adaptive training and interface adjustments, minimizing user frustration and improving task performance. In this study, we explore the automatic detection of VR familiarity by analyzing hand movement patterns during a passcode-based door-opening task, which is a well-known interaction in collaborative virtual environments such as meeting rooms, offices, and healthcare spaces. While novice users may lack prior VR experience, they are likely to be familiar with analogous real-world tasks involving keypad entry. We conducted a pilot study with 26 participants, evenly split between experienced and inexperienced VR users, who performed tasks using both controller-based and hand-tracking interactions. Our approach uses state-of-the-art deep classifiers for automatic VR familiarity detection, achieving the highest accuracies of 92.05% and 83.42% for hand-tracking and controller-based interactions, respectively. In the cross-device evaluation, where classifiers trained on controller data were tested using hand-tracking data, the model achieved an accuracy of 78.89%. The integration of both modalities in the mixed-device evaluation obtained an accuracy of 94.19%. Our results underline the promise of using hand movement biometrics for the real-time detection of user familiarity in critical VR applications, paving the way for personalized and adaptive VR experiences.

---

### SENTINEL: A Multi-Level Formal Framework for Safety Evaluation of Foundation Model-based Embodied Agents

Authors: Simon Sinong Zhan, Philip Wang, Yao Liu, Yiyan Peng, Zinan Wang, Qineng Wang, Zhian Ruan, Xiangyu Shi, Xinyu Cao, Frank Yang, Zhenyang Ni, Kangrui Wang et al.
Published: 2025-10-14
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.12985v3

We present SENTINEL, a framework for formally evaluating the physical safety of foundation model (FM)-based embodied agents. SENTINEL is the first to provide multi-level safety evaluation across semantic interpretation, plan generation, and physical execution within a unified formal framework. Unlike prior methods that rely on heuristic rules or subjective FM judgments, SENTINEL grounds practical safety requirements in formal temporal logic (TL) semantics that can precisely specify state invariants, temporal dependencies, and timing constraints. It employs a multi-level verification pipeline where (i) at the semantic level, intuitive natural language safety requirements are formalized into TL formulas and the agent's understanding of these requirements is probed for alignment with the TL formulas; (ii) at the plan level, high-level action plans and subgoals generated by the agent are verified against the TL formulas to detect unsafe plans before execution; and (iii) at the trajectory level, multiple execution trajectories are merged into a computation tree and efficiently verified against physically-detailed TL specifications for a final safety check. We apply SENTINEL in VirtualHome and AI2-THOR, and formally evaluate multiple FM-based embodied agents against diverse safety requirements. Our experiments show that by grounding physical safety in temporal logic and applying verification methods across multiple levels, SENTINEL provides a rigorous foundation for systematically evaluating the safety of FM-based embodied agents in simulation-based physical environments, and can effectively expose potential safety violations in interpreting, planning, and executing the tasks.

---

### Simplicial Gaussian Models: Representation and Inference

Authors: Lorenzo Marinucci, Gabriele D'Acunto, Paolo Di Lorenzo, Sergio Barbarossa
Published: 2025-10-14
Categories: stat.ML, cs.LG, eess.SP, stat.ME
arXiv: https://arxiv.org/abs/2510.12983v1

Probabilistic graphical models (PGMs) are powerful tools for representing statistical dependencies through graphs in high-dimensional systems. However, they are limited to pairwise interactions. In this work, we propose the simplicial Gaussian model (SGM), which extends Gaussian PGM to simplicial complexes. SGM jointly models random variables supported on vertices, edges, and triangles, within a single parametrized Gaussian distribution. Our model builds upon discrete Hodge theory and incorporates uncertainty at every topological level through independent random components. Motivated by applications, we focus on the marginal edge-level distribution while treating node- and triangle-level variables as latent. We then develop a maximum-likelihood inference algorithm to recover the parameters of the full SGM and the induced conditional dependence structure. Numerical experiments on synthetic simplicial complexes with varying size and sparsity confirm the effectiveness of our algorithm.

---

### Reference-Specific Unlearning Metrics Can Hide the Truth: A Reality Check

Authors: Sungjun Cho, Dasol Hwang, Frederic Sala, Sangheum Hwang, Kyunghyun Cho, Sungmin Cha
Published: 2025-10-14
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.12981v1

Current unlearning metrics for generative models evaluate success based on reference responses or classifier outputs rather than assessing the core objective: whether the unlearned model behaves indistinguishably from a model that never saw the unwanted data. This reference-specific approach creates systematic blind spots, allowing models to appear successful while retaining unwanted knowledge accessible through alternative prompts or attacks. We address these limitations by proposing Functional Alignment for Distributional Equivalence (FADE), a novel metric that measures distributional similarity between unlearned and reference models by comparing bidirectional likelihood assignments over generated samples. Unlike existing approaches that rely on predetermined references, FADE captures functional alignment across the entire output distribution, providing a principled assessment of genuine unlearning. Our experiments on the TOFU benchmark for LLM unlearning and the UnlearnCanvas benchmark for text-to-image diffusion model unlearning reveal that methods achieving near-optimal scores on traditional metrics fail to achieve distributional equivalence, with many becoming more distant from the gold standard than before unlearning. These findings expose fundamental gaps in current evaluation practices and demonstrate that FADE provides a more robust foundation for developing and assessing truly effective unlearning methods.

---

### DeepPlanner: Scaling Planning Capability for Deep Research Agents via Advantage Shaping

Authors: Wei Fan, Wenlin Yao, Zheng Li, Feng Yao, Xin Liu, Liang Qiu, Qingyu Yin, Yangqiu Song, Bing Yin
Published: 2025-10-14
Categories: cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.12979v2

Large language models (LLMs) augmented with multi-step reasoning and action generation abilities have shown promise in leveraging external tools to tackle complex tasks that require long-horizon planning. However, existing approaches either rely on implicit planning in the reasoning stage or introduce explicit planners without systematically addressing how to optimize the planning stage. As evidence, we observe that under vanilla reinforcement learning (RL), planning tokens exhibit significantly higher entropy than other action tokens, revealing uncertain decision points that remain under-optimized. To address this, we propose DeepPlanner, an end-to-end RL framework that effectively enhances the planning capabilities of deep research agents. Our approach shapes token-level advantage with an entropy-based term to allocate larger updates to high entropy tokens, and selectively upweights sample-level advantages for planning-intensive rollouts. Extensive experiments across seven deep research benchmarks demonstrate that DeepPlanner improves planning quality and achieves state-of-the-art results under a substantially lower training budget.

---

### A Connection Between Score Matching and Local Intrinsic Dimension

Authors: Eric Yeats, Aaron Jacobson, Darryl Hannan, Yiran Jia, Timothy Doster, Henry Kvinge, Scott Mahan
Published: 2025-10-14
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2510.12975v2

The local intrinsic dimension (LID) of data is a fundamental quantity in signal processing and learning theory, but quantifying the LID of high-dimensional, complex data has been a historically challenging task. Recent works have discovered that diffusion models capture the LID of data through the spectra of their score estimates and through the rate of change of their density estimates under various noise perturbations. While these methods can accurately quantify LID, they require either many forward passes of the diffusion model or use of gradient computation, limiting their applicability in compute- and memory-constrained scenarios. We show that the LID is a lower bound on the denoising score matching loss, motivating use of the denoising score matching loss as a LID estimator. Moreover, we show that the equivalent implicit score matching loss also approximates LID via the normal dimension and is closely related to a recent LID estimator, FLIPD. Our experiments on a manifold benchmark and with Stable Diffusion 3.5 indicate that the denoising score matching loss is a highly competitive and scalable LID estimator, achieving superior accuracy and memory footprint under increasing problem size and quantization level.

---

### Scope: Selective Cross-modal Orchestration of Visual Perception Experts

Authors: Tianyu Zhang, Suyuchen Wang, Chao Wang, Juan Rodriguez, Ahmed Masry, Xiangru Jian, Yoshua Bengio, Perouz Taslakian
Published: 2025-10-14
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.12974v2

Vision-language models (VLMs) benefit from multiple vision encoders, but naively stacking them yields diminishing returns while multiplying inference costs. We propose SCOPE, a Mixture-of-Encoders (MoEnc) framework that dynamically selects one specialized encoder per image-text pair via instance-level routing, unlike token-level routing in traditional MoE. SCOPE maintains a shared encoder and a pool of routed encoders. A lightweight router uses cross-attention between text prompts and shared visual features to select the optimal encoder from the routed encoders. To train this router, we introduce dual entropy regularization with auxiliary losses to balance dataset-level load distribution with instance-level routing confidence. Remarkably, SCOPE with one shared plus one routed encoder outperforms models using all four extra encoders simultaneously, while reducing compute by 24-49\%. This demonstrates that intelligent encoder selection beats brute-force aggregation, challenging the prevailing paradigm in multi-encoder VLMs.

---

### Balancing Performance and Reject Inclusion: A Novel Confident Inlier Extrapolation Framework for Credit Scoring

Authors: Athyrson Machado Ribeiro, Marcos Medeiros Raimundo
Published: 2025-10-14
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.12967v1

Reject Inference (RI) methods aim to address sample bias by inferring missing repayment data for rejected credit applicants. Traditional approaches often assume that the behavior of rejected clients can be extrapolated from accepted clients, despite potential distributional differences between the two populations. To mitigate this blind extrapolation, we propose a novel Confident Inlier Extrapolation framework (CI-EX). CI-EX iteratively identifies the distribution of rejected client samples using an outlier detection model and assigns labels to rejected individuals closest to the distribution of the accepted population based on probabilities derived from a supervised classification model. The effectiveness of our proposed framework is validated through experiments on two large real-world credit datasets. Performance is evaluated using the Area Under the Curve (AUC) as well as RI-specific metrics such as Kickout and a novel metric introduced in this work, denoted as Area under the Kickout. Our findings reveal that RI methods, including the proposed framework, generally involve a trade-off between AUC and RI-specific metrics. However, the proposed CI-EX framework consistently outperforms existing RI models from the credit literature in terms of RI-specific metrics while maintaining competitive performance in AUC across most experiments.

---

### 3-Model Speculative Decoding

Authors: Sanghyun Byun, Mohanad Odema, Jung Ick Guack, Baisub Lee, Jacob Song, Woo Seong Chung
Published: 2025-10-14
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.12966v1

Speculative Decoding (SD) accelerates inference in large language models by using a smaller draft model to propose tokens, which are then verified by a larger target model. However, the throughput gains of SD are fundamentally limited by a trade-off between draft model size and token acceptance: smaller draft models generate tokens more quickly but exhibit greater divergence from the target model, resulting in lower acceptance rates and reduced speedups. We introduce Pyramid Speculative Decoding (PyramidSD), an extension of SD that inserts an intermediate qualifier model between the draft and target to bridge the distributional gap in output predictions, allowing smaller model to be used for drafting. This hierarchical decoding strategy improves alignment across models, enabling higher acceptance rates and allowing the use of significantly smaller draft models without sacrificing overall performance. PyramidSD builds on fuzzy acceptance criteria to support relaxed divergence thresholds at each stage, improving throughput. In experiments, PyramidSD achieves up to 1.91x generation speed over standard SD, reaching 124 tokens per second on a consumer GPU (RTX 4090). In small-memory settings with a 1B-parameter draft model and an 8B target model, PyramidSD minimally trades target model quality for improved throughput. Overall, PyramidSD offers a practical approach to enhancing speculative decoding efficiency and can be readily applied to existing inference pipelines.

---

### Simulation-Based Pretraining and Domain Adaptation for Astronomical Time Series with Minimal Labeled Data

Authors: Rithwik Gupta, Daniel Muthukrishna, Jeroen Audenaert
Published: 2025-10-14
Categories: astro-ph.IM, astro-ph.HE, astro-ph.SR, cs.LG
arXiv: https://arxiv.org/abs/2510.12958v1

Astronomical time-series analysis faces a critical limitation: the scarcity of labeled observational data. We present a pre-training approach that leverages simulations, significantly reducing the need for labeled examples from real observations. Our models, trained on simulated data from multiple astronomical surveys (ZTF and LSST), learn generalizable representations that transfer effectively to downstream tasks. Using classifier-based architectures enhanced with contrastive and adversarial objectives, we create domain-agnostic models that demonstrate substantial performance improvements over baseline methods in classification, redshift estimation, and anomaly detection when fine-tuned with minimal real data. Remarkably, our models exhibit effective zero-shot transfer capabilities, achieving comparable performance on future telescope (LSST) simulations when trained solely on existing telescope (ZTF) data. Furthermore, they generalize to very different astronomical phenomena (namely variable stars from NASA's \textit{Kepler} telescope) despite being trained on transient events, demonstrating cross-domain capabilities. Our approach provides a practical solution for building general models when labeled data is scarce, but domain knowledge can be encoded in simulations.

---

### Attribution Graphs and Causal Probing for Mechanistic Discovery and Bias Repair in Multimodal Generative Learning

Authors: Noor Islam S. Mohammad, Uluğ Bayazıt
Published: 2025-10-14
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.12957v4

We treat the internals of generative models as mechanistic objects rather than black boxes. We introduce \textbf{Attribution Graphs} (AGs), which extend GradCAM++ to circuit-level representations, and \textbf{Causal Probing}, a do-calculus intervention method for identifying causal latent structures, enabling detection and correction of spurious correlations, demographic biases, and misaligned decision circuits during training. We further propose the \textbf{Cognitive Alignment Score (CAS)}, quantifying agreement between model-internal representations and human concepts, a \textbf{saliency-first privacy mechanism} sharing only thresholded attribution nodes, a bias-aware regularizer aligning subgroup statistics, and a Reveal-to-Revise loop integrating attribution signals into parameter updates without separate fine-tuning. Evaluated on CelebA, FairFace, Jigsaw, and HateXplain, our method achieves \textbf{94.1\%} accuracy, \textbf{92.3\%} macro F1, \textbf{79.4\%} IoU-XAI, and \textbf{12.7} FID at 72--76\% adversarial robustness, while reducing subgroup disparity $Δ_{\mathrm{bias}}$ by \textbf{41\%}, demonstrating that mechanistic interpretability, fairness, and generative performance can be jointly optimized.

---

### CADE 2.5 - ZeResFDG: Frequency-Decoupled, Rescaled and Zero-Projected Guidance for SD/SDXL Latent Diffusion Models

Authors: Denis Rychkovskiy
Published: 2025-10-14
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.12954v2

We introduce CADE 2.5 (Comfy Adaptive Detail Enhancer), a sampler-level guidance stack for SD/SDXL latent diffusion models. The central module, ZeResFDG, unifies (i) frequency-decoupled guidance that reweights low- and high-frequency components of the guidance signal, (ii) energy rescaling that matches the per-sample magnitude of the guided prediction to the positive branch, and (iii) zero-projection that removes the component parallel to the unconditional direction. A lightweight spectral EMA with hysteresis switches between a conservative and a detail-seeking mode as structure crystallizes during sampling. Across SD/SDXL samplers, ZeResFDG improves sharpness, prompt adherence, and artifact control at moderate guidance scales without any retraining. In addition, we employ a training-free inference-time stabilizer, QSilk Micrograin Stabilizer (quantile clamp + depth/edge-gated micro-detail injection), which improves robustness and yields natural high-frequency micro-texture at high resolutions with negligible overhead. For completeness we note that the same rule is compatible with alternative parameterizations (e.g., velocity), which we briefly discuss in the Appendix; however, this paper focuses on SD/SDXL latent diffusion models.

---

### Epistemic-aware Vision-Language Foundation Model for Fetal Ultrasound Interpretation

Authors: Xiao He, Huangxuan Zhao, Guojia Wan, Jiancheng Pan, Yanxing Liu, Yong Luo, Juhua Liu, Yongchao Xu, Wei Zhou, Dacheng Tao, Bo Du
Published: 2025-10-14
Categories: cs.CV, cs.AI, cs.IR, cs.MM
arXiv: https://arxiv.org/abs/2510.12953v5

Recent medical vision-language models have shown promise on tasks such as VQA, report generation, and anomaly detection. However, most are adapted to structured adult imaging and underperform in fetal ultrasound, which poses challenges of multi-view image reasoning, numerous diseases, and image diversity. To bridge this gap, we introduce FetalMind, a medical AI system tailored to fetal ultrasound for both report generation and diagnosis. Guided by clinical workflow, we propose Salient Epistemic Disentanglement (SED), which injects an expert-curated bipartite graph into the model to decouple view-disease associations and to steer preference selection along clinically faithful steps via reinforcement learning. This design mitigates variability across diseases and heterogeneity across views, reducing learning bottlenecks while aligning the model's inference with obstetric practice. To train FetalMind at scale, we curate FetalSigma-1M dataset, the first large-scale fetal ultrasound report corpus, comprising 20K reports from twelve medical centers, addressing the scarcity of domain data. Extensive experiments show that FetalMind outperforms open- and closed-source baselines across all gestational stages, achieving +14% average gains and +61.2% higher accuracy on critical conditions while remaining efficient, stable, and scalable. Project Page: https://hexiao0275.github.io/FetalMind.

---

### An Investigation of Memorization Risk in Healthcare Foundation Models

Authors: Sana Tonekaboni, Lena Stempfle, Adibvafa Fallahpour, Walter Gerych, Marzyeh Ghassemi
Published: 2025-10-14
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.12950v1

Foundation models trained on large-scale de-identified electronic health records (EHRs) hold promise for clinical applications. However, their capacity to memorize patient information raises important privacy concerns. In this work, we introduce a suite of black-box evaluation tests to assess privacy-related memorization risks in foundation models trained on structured EHR data. Our framework includes methods for probing memorization at both the embedding and generative levels, and aims to distinguish between model generalization and harmful memorization in clinically relevant settings. We contextualize memorization in terms of its potential to compromise patient privacy, particularly for vulnerable subgroups. We validate our approach on a publicly available EHR foundation model and release an open-source toolkit to facilitate reproducible and collaborative privacy assessments in healthcare AI.

---

### SpareCodeSearch: Searching for Code Context When You Have No Spare GPU

Authors: Minh Nguyen
Published: 2025-10-14
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2510.12948v1

Retrieval-Augmented Generation (RAG) frameworks aim to enhance Code Language Models (CLMs) by including another module for retrieving relevant context to construct the input prompt. However, these retrieval modules commonly use semantic search, requiring substantial computational resources for training and hosting these embedded models, making them infeasible to integrate into lightweight applications such as in-IDE AI-based code completion. In this solution paper, we prove that using keyword-search is sufficient to retrieve relevant and useful code context inside large codebases, without the need for extensive GPU resources. The usefulness of code contexts found by our solution is demonstrated through their completion results on the Code Context Competition's benchmark, reaching 0.748 and 0.725 chRF scores on Kotlin and Python tracks, respectively.

---

### HyWA: Architecture-Preserving Personalized Voice Activity Detection for Full-Duplex Voice Assistants

Authors: Hamed Jafarzadeh Asl, Amin Edraki, Mahsa Ghazvini Nejad, Masoud Asgharian, Mohammadreza Sadeghi, Yuanhao Yu, Vahid Partovi Nia
Published: 2025-10-14
Categories: eess.AS, cs.AI, cs.LG, cs.SD
arXiv: https://arxiv.org/abs/2510.12947v3

Voice activity detection (VAD) serves as an early gate in voice-assistant pipelines for smart devices. Because conventional VADs respond to speech from any speaker, nearby conversations and residual assistant playback lead to unwanted triggers, degrade the user experience, and waste computational resources. Personalized voice activity detection (PVAD) addresses this limitation by detecting speech only from an enrolled target speaker. Existing PVAD methods typically incorporate speaker information into model inputs or hidden representations. These approaches require VAD architectural changes that increase engineering and requalification costs in deployment. We present HyWA, a hypernetwork-based weight-adaptation method that converts an established VAD into a PVAD, while preserving its acoustic interface and inference topology. HyWA generates speaker-conditioned weights once at enrollment and requires no per-user optimization. We apply HyWA to pretrained Alibaba's FSMN and NVIDIA's MarbleNet VAD models. Evaluations on synthetic and real-user test data show consistent improvements in target-speaker discrimination and false-positive suppression. In an integrated full-duplex barge-in pipeline, replacing generic VAD with HyWA-based PVAD reduces observed false-interruption detections from 88.9% to only 9.9%.

---

### The Curious Case of Curiosity across Human Cultures and LLMs

Authors: Angana Borah, Zhijing Jin, Rada Mihalcea
Published: 2025-10-14
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.12943v2

Recent advances in Large Language Models (LLMs) have expanded their role in human interaction, yet curiosity -- a central driver of inquiry -- remains underexplored in these systems, particularly across cultural contexts. In this work, we investigate cultural variation in curiosity using Yahoo! Answers, a real-world multi-country dataset spanning diverse topics. We introduce CUEST (CUriosity Evaluation across SocieTies), an evaluation framework that measures human-model alignment in curiosity through linguistic (style), topic preference (content) analysis and grounding insights in social science constructs. Across open- and closed-source models, we find that LLMs flatten cross-cultural diversity, aligning more closely with how curiosity is expressed in Western countries. We then explore fine-tuning strategies to induce curiosity in LLMs, narrowing the human-model alignment gap by up to 50%. Finally, we demonstrate the practical value of curiosity for LLM adaptability across cultures, showing its importance for future NLP research.

---

### Pruning Cannot Hurt Robustness: Certified Trade-offs in Reinforcement Learning

Authors: James Pedley, Benjamin Etheridge, Stephen J. Roberts, Francesco Quinzan
Published: 2025-10-14
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.12939v1

Reinforcement learning (RL) policies deployed in real-world environments must remain reliable under adversarial perturbations. At the same time, modern deep RL agents are heavily over-parameterized, raising costs and fragility concerns. While pruning has been shown to improve robustness in supervised learning, its role in adversarial RL remains poorly understood. We develop the first theoretical framework for certified robustness under pruning in state-adversarial Markov decision processes (SA-MDPs). For Gaussian and categorical policies with Lipschitz networks, we prove that element-wise pruning can only tighten certified robustness bounds; pruning never makes the policy less robust. Building on this, we derive a novel three-term regret decomposition that disentangles clean-task performance, pruning-induced performance loss, and robustness gains, exposing a fundamental performance--robustness frontier. Empirically, we evaluate magnitude and micro-pruning schedules on continuous-control benchmarks with strong policy-aware adversaries. Across tasks, pruning consistently uncovers reproducible ``sweet spots'' at moderate sparsity levels, where robustness improves substantially without harming - and sometimes even enhancing - clean performance. These results position pruning not merely as a compression tool but as a structural intervention for robust RL.

---

### Learning at the Speed of Physics: Equilibrium Propagation on Oscillator Ising Machines

Authors: Alex Gower
Published: 2025-10-14
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.12934v2

Physical systems that naturally perform energy descent offer a direct route to accelerating machine learning. Oscillator Ising Machines (OIMs) exemplify this idea: their GHz-frequency dynamics mirror both the optimization of energy-based models (EBMs) and gradient descent on loss landscapes, while intrinsic noise corresponds to Langevin dynamics - supporting sampling as well as optimization. Equilibrium Propagation (EP) unifies these processes into descent on a single total energy landscape, enabling local learning rules without global backpropagation. We show that EP on OIMs achieves competitive accuracy ($\sim 97.2 \pm 0.1 \%$ on MNIST, $\sim 88.0 \pm 0.1 \%$ on Fashion-MNIST), while maintaining robustness under realistic hardware constraints such as parameter quantization and phase noise. These results establish OIMs as a fast, energy-efficient substrate for neuromorphic learning, and suggest that EBMs - often bottlenecked by conventional processors - may find practical realization on physical hardware whose dynamics directly perform their optimization.

---

### Unifying Vision-Language Latents for Zero-label Image Caption Enhancement

Authors: Sanghyun Byun, Jung Ick Guack, Mohanad Odema, Baisub Lee, Jacob Song, Woo Seong Chung
Published: 2025-10-14
Categories: cs.CV, cs.CL
arXiv: https://arxiv.org/abs/2510.12931v1

Vision-language models (VLMs) achieve remarkable performance through large-scale image-text pretraining. However, their reliance on labeled image datasets limits scalability and leaves vast amounts of unlabeled image data underutilized. To address this, we propose Unified Vision-Language Alignment for Zero-Label Enhancement (ViZer), an enhancement training framework that enables zero-label learning in image captioning, providing a practical starting point for broader zero-label adaptation in vision-language tasks. Unlike prior approaches that rely on human or synthetically annotated datasets, ViZer actively aligns vision and language representation features during training, enabling existing VLMs to generate improved captions without requiring text labels or full retraining. We demonstrate ViZer's advantage in qualitative evaluation, as automated caption metrics such as CIDEr and BERTScore often penalize details that are absent in reference captions. Applying ViZer on SmolVLM-Base and Qwen2-VL, we observe consistent qualitative improvements, producing captions that are more grounded and descriptive than their baseline.

---

### FedGTEA: Federated Class-Incremental Learning with Gaussian Task Embedding and Alignment

Authors: Haolin Li, Hoda Bidkhori
Published: 2025-10-14
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2510.12927v1

We introduce a novel framework for Federated Class Incremental Learning, called Federated Gaussian Task Embedding and Alignment (FedGTEA). FedGTEA is designed to capture task-specific knowledge and model uncertainty in a scalable and communication-efficient manner. At the client side, the Cardinality-Agnostic Task Encoder (CATE) produces Gaussian-distributed task embeddings that encode task knowledge, address statistical heterogeneity, and quantify data uncertainty. Importantly, CATE maintains a fixed parameter size regardless of the number of tasks, which ensures scalability across long task sequences. On the server side, FedGTEA utilizes the 2-Wasserstein distance to measure inter-task gaps between Gaussian embeddings. We formulate the Wasserstein loss to enforce inter-task separation. This probabilistic formulation not only enhances representation learning but also preserves task-level privacy by avoiding the direct transmission of latent embeddings, aligning with the privacy constraints in federated learning. Extensive empirical evaluations on popular datasets demonstrate that FedGTEA achieves superior classification performance and significantly mitigates forgetting, consistently outperforming strong existing baselines.

---

### Who's Asking? Evaluating LLM Robustness to Inquiry Personas in Factual Question Answering

Authors: Nil-Jana Akpinar, Chia-Jung Lee, Vanessa Murdock, Pietro Perona
Published: 2025-10-14
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2510.12925v1

Large Language Models (LLMs) should answer factual questions truthfully, grounded in objective knowledge, regardless of user context such as self-disclosed personal information, or system personalization. In this paper, we present the first systematic evaluation of LLM robustness to inquiry personas, i.e. user profiles that convey attributes like identity, expertise, or belief. While prior work has primarily focused on adversarial inputs or distractors for robustness testing, we evaluate plausible, human-centered inquiry persona cues that users disclose in real-world interactions. We find that such cues can meaningfully alter QA accuracy and trigger failure modes such as refusals, hallucinated limitations, and role confusion. These effects highlight how model sensitivity to user framing can compromise factual reliability, and position inquiry persona testing as an effective tool for robustness evaluation.

---

### InferA: A Smart Assistant for Cosmological Ensemble Data

Authors: Justin Z. Tam, Pascal Grosset, Divya Banesh, Nesar Ramachandra, Terece L. Turton, James Ahrens
Published: 2025-10-14
Categories: astro-ph.IM, cs.AI
arXiv: https://arxiv.org/abs/2510.12920v1

Analyzing large-scale scientific datasets presents substantial challenges due to their sheer volume, structural complexity, and the need for specialized domain knowledge. Automation tools, such as PandasAI, typically require full data ingestion and lack context of the full data structure, making them impractical as intelligent data analysis assistants for datasets at the terabyte scale. To overcome these limitations, we propose InferA, a multi-agent system that leverages large language models to enable scalable and efficient scientific data analysis. At the core of the architecture is a supervisor agent that orchestrates a team of specialized agents responsible for distinct phases of the data retrieval and analysis. The system engages interactively with users to elicit their analytical intent and confirm query objectives, ensuring alignment between user goals and system actions. To demonstrate the framework's usability, we evaluate the system using ensemble runs from the HACC cosmology simulation which comprises several terabytes.

---

### Efficient Inference for Coupled Hidden Markov Models in Continuous Time and Discrete Space

Authors: Giosue Migliorini, Padhraic Smyth
Published: 2025-10-14
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2510.12916v2

Systems of interacting continuous-time Markov chains are a powerful model class, but inference is typically intractable in high dimensional settings. Auxiliary information, such as noisy observations, is typically only available at discrete times, and incorporating it via a Doob's $h$-transform gives rise to an intractable posterior process that requires approximation. We introduce Latent Interacting Particle Systems, a model class parameterizing the generator of each Markov chain in the system. Our inference method involves estimating look-ahead functions (twist potentials) that anticipate future information, for which we introduce an efficient parameterization. We incorporate this approximation in a twisted Sequential Monte Carlo sampling scheme. We demonstrate the effectiveness of our approach on a challenging posterior inference task for a latent SIRS model on a graph, and on a neural model for wildfire spread dynamics trained on real data.

---

### Toward LLM-Supported Automated Assessment of Critical Thinking Subskills

Authors: Marisa C. Peczuh, Nischal Ashok Kumar, Ryan Baker, Blair Lehman, Danielle Eisenberg, Caitlin Mills, Payu Wittawatolarn, Kushaan Naskar, Keerthi Chebrolu, Sudhip Nashi, Cadence Young, Brayden Liu et al.
Published: 2025-10-14
Categories: cs.CY, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2510.12915v2

As the world becomes increasingly saturated with AI-generated content, disinformation, and algorithmic persuasion, critical thinking - the capacity to evaluate evidence, detect unreliable claims, and exercise independent judgment - is becoming a defining human skill. Developing critical thinking skills through timely assessment and feedback is crucial; however, there has not been extensive work in educational data mining on defining, measuring, and supporting critical thinking. In this paper, we investigate the feasibility of measuring "subskills" that underlie critical thinking. We ground our work in an authentic task where students operationalize critical thinking by writing argumentative essays. We developed a coding rubric based on an established skills progression and completed human coding for a corpus of student essays. We then evaluated three distinct approaches to automated scoring: zero-shot prompting, few-shot prompting, and supervised fine-tuning, implemented across three large language models (GPT-5, Llama 3.1 8B, and ModernBERT). Fine-tuning Llama 3.1 8B achieved the best results and demonstrated particular strength on subskills with highly separable proficiency levels with balanced labels across levels, while lower performance was observed for subskills that required detection of subtle distinctions between proficiency levels or imbalanced labels. Our exploratory work represents an initial step toward scalable assessment of critical thinking skills across authentic educational contexts. Future research should continue to combine automated critical thinking assessment with human validation to more accurately detect and measure dynamic, higher-order thinking skills.

---

### Robust Plant Disease Diagnosis with Few Target-Domain Samples

Authors: Takafumi Nogami, Satoshi Kagiwada, Hitoshi Iyatomi
Published: 2025-10-14
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.12909v1

Various deep learning-based systems have been proposed for accurate and convenient plant disease diagnosis, achieving impressive performance. However, recent studies show that these systems often fail to maintain diagnostic accuracy on images captured under different conditions from the training environment -- an essential criterion for model robustness. Many deep learning methods have shown high accuracy in plant disease diagnosis. However, they often struggle to generalize to images taken in conditions that differ from the training setting. This drop in performance stems from the subtle variability of disease symptoms and domain gaps -- differences in image context and environment. The root cause is the limited diversity of training data relative to task complexity, making even advanced models vulnerable in unseen domains. To tackle this challenge, we propose a simple yet highly adaptable learning framework called Target-Aware Metric Learning with Prioritized Sampling (TMPS), grounded in metric learning. TMPS operates under the assumption of access to a limited number of labeled samples from the target (deployment) domain and leverages these samples effectively to improve diagnostic robustness. We assess TMPS on a large-scale automated plant disease diagnostic task using a dataset comprising 223,073 leaf images sourced from 23 agricultural fields, spanning 21 diseases and healthy instances across three crop species. By incorporating just 10 target domain samples per disease into training, TMPS surpasses models trained using the same combined source and target samples, and those fine-tuned with these target samples after pre-training on source data. It achieves average macro F1 score improvements of 7.3 and 3.6 points, respectively, and a remarkable 18.7 and 17.1 point improvement over the baseline and conventional metric learning.

---

### State-Change Learning for Prediction of Future Events in Endoscopic Videos

Authors: Saurav Sharma, Chinedu Innocent Nwoye, Didier Mutter, Nicolas Padoy
Published: 2025-10-14
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.12904v2

Surgical future prediction, driven by real-time AI analysis of surgical video, is critical for operating room safety and efficiency. It provides actionable insights into upcoming events, their timing, and risks-enabling better resource allocation, timely instrument readiness, and early warnings for complications (e.g., bleeding, bile duct injury). Despite this need, current surgical AI research focuses on understanding what is happening rather than predicting future events. Existing methods target specific tasks in isolation, lacking unified approaches that span both short-term (action triplets, events) and long-term horizons (remaining surgery duration, phase transitions). These methods rely on coarse-grained supervision while fine-grained surgical action triplets and steps remain underexplored. Furthermore, methods based only on future feature prediction struggle to generalize across different surgical contexts and procedures. We address these limits by reframing surgical future prediction as state-change learning. Rather than forecasting raw observations, our approach classifies state transitions between current and future timesteps. We introduce SurgFUTR, implementing this through a teacher-student architecture. Video clips are compressed into state representations via Sinkhorn-Knopp clustering; the teacher network learns from both current and future clips, while the student network predicts future states from current videos alone, guided by our Action Dynamics (ActDyn) module. We establish SFPBench with five prediction tasks spanning short-term (triplets, events) and long-term (remaining surgery duration, phase and step transitions) horizons. Experiments across four datasets and three procedures show consistent improvements. Cross-procedure transfer validates generalizability.

---

### SimULi: Real-Time LiDAR and Camera Simulation with Unscented Transforms

Authors: Haithem Turki, Qi Wu, Xin Kang, Janick Martinez Esturo, Shengyu Huang, Ruilong Li, Zan Gojcic, Riccardo de Lutio
Published: 2025-10-14
Categories: cs.CV, cs.GR, cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2510.12901v3

Rigorous testing of autonomous robots, such as self-driving vehicles, is essential to ensure their safety in real-world deployments. This requires building high-fidelity simulators to test scenarios beyond those that can be safely or exhaustively collected in the real-world. Existing neural rendering methods based on NeRF and 3DGS hold promise but suffer from low rendering speeds or can only render pinhole camera models, hindering their suitability to applications that commonly require high-distortion lenses and LiDAR data. Multi-sensor simulation poses additional challenges as existing methods handle cross-sensor inconsistencies by favoring the quality of one modality at the expense of others. To overcome these limitations, we propose SimULi, the first method capable of rendering arbitrary camera models and LiDAR data in real-time. Our method extends 3DGUT, which natively supports complex camera models, with LiDAR support, via an automated tiling strategy for arbitrary spinning LiDAR models and ray-based culling. To address cross-sensor inconsistencies, we design a factorized 3D Gaussian representation and anchoring strategy that reduces mean camera and depth error by up to 40% compared to existing methods. SimULi renders 10-20x faster than ray tracing approaches and 1.5-10x faster than prior rasterization-based work (and handles a wider range of camera models). When evaluated on two widely benchmarked autonomous driving datasets, SimULi matches or exceeds the fidelity of existing state-of-the-art methods across numerous camera and LiDAR metrics.

---

### EduDial: Constructing a Large-scale Multi-turn Teacher-Student Dialogue Corpus

Authors: Shouang Wei, Min Zhang, Xin Lin, Bo Jiang, Zhongxiang Dai, Kun Kuang
Published: 2025-10-14
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.12899v3

Recently, several multi-turn dialogue benchmarks have been proposed to evaluate the conversational abilities of large language models (LLMs). As LLMs are increasingly recognized as a key technology for advancing intelligent education, owing to their ability to deeply understand instructional contexts and provide personalized guidance, the construction of dedicated teacher-student dialogue benchmarks has become particularly important. To this end, we present EduDial, a comprehensive multi-turn teacher-student dialogue dataset. EduDial covers 345 core knowledge points and consists of 34,250 dialogue sessions generated through interactions between teacher and student agents. Its design is guided by Bloom's taxonomy of educational objectives and incorporates ten questioning strategies, including situational questioning, zone of proximal development (ZPD) questioning, and metacognitive questioning-thus better capturing authentic classroom interactions. Furthermore, we design differentiated teaching strategies for students at different cognitive levels, thereby providing more targeted teaching guidance. Building on EduDial, we further develop EduDial-LLM 32B via training and propose an 11-dimensional evaluation framework that systematically measures the teaching abilities of LLMs, encompassing both overall teaching quality and content quality. Experiments on 17 mainstream LLMs reveal that most models struggle in student-centered teaching scenarios, whereas our EduDial-LLM achieves significant gains, consistently outperforming all baselines across all metrics. The code is available at https://github.com/Mind-Lab-ECNU/EduDial/tree/main.

---

### KVCOMM: Online Cross-context KV-cache Communication for Efficient LLM-based Multi-agent Systems

Authors: Hancheng Ye, Zhengqi Gao, Mingyuan Ma, Qinsi Wang, Yuzhe Fu, Ming-Yu Chung, Yueqian Lin, Zhijian Liu, Jianyi Zhang, Danyang Zhuo, Yiran Chen
Published: 2025-10-14
Categories: cs.MA, cs.AI, stat.ML
arXiv: https://arxiv.org/abs/2510.12872v2

Multi-agent large language model (LLM) systems are increasingly adopted for complex language processing tasks that require communication and coordination among agents. However, these systems often suffer substantial overhead from repeated reprocessing of overlapping contexts across agents. In typical pipelines, once an agent receives a message from its predecessor, the full context-including prior turns-must be reprocessed from scratch, leading to inefficient processing. While key-value (KV) caching is an effective solution for avoiding redundant computation in single-agent settings where prefixes remain unchanged, it cannot be directly reused in multi-agent scenarios due to diverging prefixes introduced by agent-specific context extensions. We identify that the core challenge lies in the offset variance of KV-caches across agents. To address this, we propose KVCOMM, a training-free framework that enables efficient prefilling in multi-agent inference by reusing KV-caches and aligning cache offsets of overlapping contexts under diverse prefix contexts. KVCOMM estimates and adjusts KV-caches for shared content by referencing a pool of cached examples-termed anchors-that store observed cache deviations under varying prefixes. The anchor pool is maintained and updated online, allowing dynamic adaptation to distinct user requests and context structures. KVCOMM achieves over 70% reuse rate across diverse multi-agent workloads, including retrieval-augmented generation, math reasoning, and collaborative coding tasks, all without quality degradation. Particularly, when each fully-connected agent receives 1K input tokens with 512 prefix tokens and 512 output tokens under a five-agent setting, KVCOMM achieves up to 7.8x speedup compared to the standard prefill pipeline, reducing TTFT from ~430 ms to ~55 ms.

---

### Learning to Grasp Anything by Playing with Random Toys

Authors: Dantong Niu, Yuvan Sharma, Baifeng Shi, Rachel Ding, Matteo Gioia, Haoru Xue, Henry Tsai, Konstantinos Kallidromitis, Anirudh Pai, Caitlin Regan, Shankar Sastry, Trevor Darrell et al.
Published: 2025-10-14
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2510.12866v2

Robotic manipulation policies often struggle to generalize to novel objects, limiting their real-world utility. In contrast, cognitive science suggests that children develop generalizable dexterous manipulation skills by mastering a small set of simple toys and then applying that knowledge to more complex items. Inspired by this, we study if similar generalization capabilities can also be achieved by robots. Our results indicate robots can learn generalizable grasping using randomly assembled objects that are composed from just four shape primitives: spheres, cuboids, cylinders, and rings. We show that training on these "toys" enables robust generalization to real-world objects, yielding strong zero-shot performance. Crucially, we find the key to this generalization is an object-centric visual representation induced by our proposed detection pooling mechanism. Evaluated in both simulation and on physical robots, our model achieves a 67% real-world grasping success rate on the YCB dataset, outperforming state-of-the-art approaches that rely on substantially more in-domain data. We further study how zero-shot generalization performance scales by varying the number and diversity of training toys and the demonstrations per toy. We believe this work offers a promising path to scalable and generalizable learning in robotic manipulation. Demonstration videos, code, checkpoints and our dataset are available on our project page: https://lego-grasp.github.io/ .

---

### From Literal to Liberal: A Meta-Prompting Framework for Eliciting Human-Aligned Exception Handling in Large Language Models

Authors: Imran Khan
Published: 2025-10-14
Categories: cs.AI, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2510.12864v1

Large Language Models (LLMs) are increasingly being deployed as the reasoning engines for agentic AI systems, yet they exhibit a critical flaw: a rigid adherence to explicit rules that leads to decisions misaligned with human common sense and intent. This "rule-rigidity" is a significant barrier to building trustworthy autonomous agents. While prior work has shown that supervised fine-tuning (SFT) with human explanations can mitigate this issue, SFT is computationally expensive and inaccessible to many practitioners. To address this gap, we introduce the Rule-Intent Distinction (RID) Framework, a novel, low-compute meta-prompting technique designed to elicit human-aligned exception handling in LLMs in a zero-shot manner. The RID framework provides the model with a structured cognitive schema for deconstructing tasks, classifying rules, weighing conflicting outcomes, and justifying its final decision. We evaluated the RID framework against baseline and Chain-of-Thought (CoT) prompting on a custom benchmark of 20 scenarios requiring nuanced judgment across diverse domains. Our human-verified results demonstrate that the RID framework significantly improves performance, achieving a 95% Human Alignment Score (HAS), compared to 80% for the baseline and 75% for CoT. Furthermore, it consistently produces higher-quality, intent-driven reasoning. This work presents a practical, accessible, and effective method for steering LLMs from literal instruction-following to liberal, goal-oriented reasoning, paving the way for more reliable and pragmatic AI agents.

---

### Three Lenses on the AI Revolution: Risk, Transformation, Continuity

Authors: Masoud Makrehchi
Published: 2025-10-14
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2510.12859v2

Artificial Intelligence (AI) has emerged as both a continuation of historical technological revolutions and a potential rupture with them. This paper argues that AI must be viewed simultaneously through three lenses: \textit{risk}, where it resembles nuclear technology in its irreversible and global externalities; \textit{transformation}, where it parallels the Industrial Revolution as a general-purpose technology driving productivity and reorganization of labor; and \textit{continuity}, where it extends the fifty-year arc of computing revolutions from personal computing to the internet to mobile. Drawing on historical analogies, we emphasize that no past transition constituted a strict singularity: disruptive shifts eventually became governable through new norms and institutions. We examine recurring patterns across revolutions -- democratization at the usage layer, concentration at the production layer, falling costs, and deepening personalization -- and show how these dynamics are intensifying in the AI era. Sectoral analysis illustrates how accounting, law, education, translation, advertising, and software engineering are being reshaped as routine cognition is commoditized and human value shifts to judgment, trust, and ethical responsibility. At the frontier, the challenge of designing moral AI agents highlights the need for robust guardrails, mechanisms for moral generalization, and governance of emergent multi-agent dynamics. We conclude that AI is neither a singular break nor merely incremental progress. It is both evolutionary and revolutionary: predictable in its median effects yet carrying singularity-class tail risks. Good outcomes are not automatic; they require coupling pro-innovation strategies with safety governance, ensuring equitable access, and embedding AI within a human order of responsibility.

---

### A Critical Review of the Need for Knowledge-Centric Evaluation of Quranic Recitation

Authors: Mohammed Hilal Al-Kharusi, Khizar Hayat, Khalil Bader Al Ruqeishi, Haroon Rashid Lone
Published: 2025-10-14
Categories: cs.CL, cs.AI, cs.SD
arXiv: https://arxiv.org/abs/2510.12858v2

The art and science of Quranic recitation (Tajweed), a discipline governed by meticulous phonetic, rhythmic, and theological principles, confronts substantial educational challenges in today's digital age. Although modern technology offers unparalleled opportunities for learning, existing automated systems for evaluating recitation have struggled to gain broad acceptance or demonstrate educational effectiveness. This literature review examines this crucial disparity, offering a thorough analysis of scholarly research, digital platforms, and commercial tools developed over the past twenty years. Our analysis uncovers a fundamental flaw in current approaches that adapt Automatic Speech Recognition (ASR) systems, which emphasize word identification over qualitative acoustic evaluation. These systems suffer from limitations such as reliance on biased datasets, demographic disparities, and an inability to deliver meaningful feedback for improvement. Challenging these data-centric methodologies, we advocate for a paradigm shift toward a knowledge-based computational framework. By leveraging the unchanging nature of the Quranic text and the well-defined rules of Tajweed, we propose that an effective evaluation system should be built upon rule-based acoustic modeling centered on canonical pronunciation principles and articulation points (Makhraj), rather than depending on statistical patterns derived from flawed or biased data. The review concludes that the future of automated Quranic recitation assessment lies in hybrid systems that combine linguistic expertise with advanced audio processing. Such an approach paves the way for developing reliable, fair, and pedagogically effective tools that can authentically assist learners across the globe.

---

### Adaptive Generation of Bias-Eliciting Questions for LLMs

Authors: Robin Staab, Jasper Dekoninck, Maximilian Baader, Martin Vechev
Published: 2025-10-14
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2510.12857v2

Large language models (LLMs) are now widely deployed in user-facing applications, reaching hundreds of millions of users worldwide. Despite their widespread adoption, growing reliance on their outputs raises significant concerns, particularly as users may be exposed to model-inherent biases that disadvantage or stereotype certain groups. However, existing bias benchmarks commonly rely on simple templated prompts or restrictive multiple-choice questions that fail to capture the complexity of real-world user interactions. In this work, we address this gap by introducing a counterfactual framework that automatically generates realistic, open-ended questions for LLM bias evaluation. Through iterative question mutation, our approach systematically explores areas where models are most likely to exhibit biased behavior. Beyond just detecting harmful biases, we also capture increasingly relevant response dimensions, such as asymmetric refusals and explicit bias acknowledgment. Building on this, we construct CAB, a diverse and human-verified benchmark for realistic and nuanced bias evaluations on current frontier LLMs. Our evaluation using CAB highlights the continued need for fairness research by showing that all examined models exhibit persistent biases across certain scenarios.

---

### Efficient Adaptive Transformer: An Empirical Study and Reproducible Framework

Authors: Jan Miller
Published: 2025-10-14
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.12856v1

The Efficient Adaptive Transformer (EAT) framework unifies three adaptive efficiency techniques - progressive token pruning, sparse attention, and dynamic early exiting - into a single, reproducible architecture for input-adaptive inference. EAT provides an open-source benchmarking pipeline that automates data processing, timing, and ablation across GLUE tasks (SST-2, QQP, MNLI). Although this empirical study finds that combining these mechanisms can increase latency in shallow six-layer models, it demonstrates that EAT achieves slightly higher accuracy than the optimized DistilBERT baseline on SST-2, illustrating the potential of dynamic computation for latency-sensitive NLP. The main contribution is the open, end-to-end reproducible framework - complete with scripts, CSV logging, and analysis utilities - intended to serve as a community tool for further research on adaptive transformers.

---

### Silence is Golden: Mitigating Hallucinations in Large Audio-Language Models via Layer-Weighted Vector Steering

Authors: Tsung-En Lin, Kuan-Yi Lee, Hung-Yi Lee
Published: 2025-10-14
Categories: cs.SD, cs.LG, eess.AS
arXiv: https://arxiv.org/abs/2510.12851v2

Large Audio-Language Models (LALMs) excel in Audio QA but often suffer from hallucinations ungrounded in the audio. To our knowledge, we are the first to propose applying vector steering to the audio domain to mitigate this. Unlike text-based steering, our silence-anchored contrastive approach steers the model away from hallucinations by contrasting active audio against a silent baseline. Probing internal states reveals a strong correlation between specific layer representations and output correctness. Leveraging this, we introduce Layer-Weighted Vector Steering (LWVS), a training-free intervention that increases steering strength at influential layers. On the Audio Hallucination QA dataset, LWVS significantly outperforms baselines, boosting Recall on the Gemma model by 15.6% (53.4% to 69.0%). Crucially, MMAU benchmark tests confirm LWVS preserves and even enhances general audio understanding, achieving an 8% relative accuracy increase on the Qwen model (54.8% to 59.2%).

---

### Ethic-BERT: An Enhanced Deep Learning Model for Ethical and Non-Ethical Content Classification

Authors: Mahamodul Hasan Mahadi, Md. Nasif Safwan, Souhardo Rahman, Shahnaj Parvin, Aminun Nahar, Kamruddin Nur
Published: 2025-10-14
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2510.12850v1

Developing AI systems capable of nuanced ethical reasoning is critical as they increasingly influence human decisions, yet existing models often rely on superficial correlations rather than principled moral understanding. This paper introduces Ethic-BERT, a BERT-based model for ethical content classification across four domains: Commonsense, Justice, Virtue, and Deontology. Leveraging the ETHICS dataset, our approach integrates robust preprocessing to address vocabulary sparsity and contextual ambiguities, alongside advanced fine-tuning strategies like full model unfreezing, gradient accumulation, and adaptive learning rate scheduling. To evaluate robustness, we employ an adversarially filtered "Hard Test" split, isolating complex ethical dilemmas. Experimental results demonstrate Ethic-BERT's superiority over baseline models, achieving 82.32% average accuracy on the standard test, with notable improvements in Justice and Virtue. In addition, the proposed Ethic-BERT attains 15.28% average accuracy improvement in the HardTest. These findings contribute to performance improvement and reliable decision-making using bias-aware preprocessing and proposed enhanced AI model.

---

### Lifting Manifolds to Mitigate Pseudo-Alignment in LLM4TS

Authors: Liangwei Nathan Zheng, Wenhao Liang, Wei Emma Zhang, Miao Xu, Olaf Maennel, Weitong Chen
Published: 2025-10-14
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.12847v1

Pseudo-Alignment is a pervasive challenge in many large language models for time series (LLM4TS) models, often causing them to underperform compared to linear models or randomly initialised backbones. However, there is limited discussion in the community for the reasons that pseudo-alignment occurs. In this work, we conduct a thorough investigation into the root causes of pseudo-alignment in LLM4TS and build a connection of pseudo-alignment to the cone effect in LLM. We demonstrate that pseudo-alignment arises from the interplay of cone effect within pretrained LLM components and the intrinsically low-dimensional manifold of time-series data. In addition, we also introduce \textit{\textbf{TimeSUP}}, a novel technique designed to mitigate this issue and improve forecast performance in existing LLM4TS approaches. TimeSUP addresses this by increasing the time series manifold to more closely match the intrinsic dimension of language embeddings, allowing the model to distinguish temporal signals clearly while still capturing shared structures across modalities. As a result, representations for time and language tokens remain distinct yet exhibit high cosine similarity, signifying that the model preserves each modality unique features while learning their commonalities in a unified embedding space. Empirically, TimeSUP consistently outperforms state-of-the-art LLM4TS methods and other lightweight baselines on long-term forecasting performance. Furthermore, it can be seamlessly integrated into four existing LLM4TS pipelines and delivers significant improvements in forecasting performance.

---