# Research Papers — 2025-10 · bundle page 190 of 217 > Full-text markdown of papers #9451–9500 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/190 [← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/189.md) [Next bundle →](https://aipod.dev/papers/archive/2025-10/page/191.md) ### From Filters to VLMs: Benchmarking Defogging Methods through Object Detection and Segmentation Performance Authors: Ardalan Aryashad, Parsa Razmara, Amin Mahjoub, Seyedarmin Azizi, Mahdi Salmani, Arad Firouzkouhi Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03906v2 Autonomous driving perception systems are particularly vulnerable in foggy conditions, where light scattering reduces contrast and obscures fine details critical for safe operation. While numerous defogging methods exist, from handcrafted filters to learned restoration models, improvements in image fidelity do not consistently translate into better downstream detection and segmentation. Moreover, prior evaluations often rely on synthetic data, raising concerns about real-world transferability. We present a structured empirical study that benchmarks a comprehensive set of defogging pipelines, including classical dehazing filters, modern defogging networks, chained variants combining filters and models, and prompt-driven visual language image editing models applied directly to foggy images. To bridge the gap between simulated and physical environments, we evaluate these pipelines on both the synthetic Foggy Cityscapes dataset and the real-world Adverse Conditions Dataset with Correspondences (ACDC). We examine generalization by evaluating performance on synthetic fog and real-world conditions, assessing both image quality and downstream perception in terms of object detection mean average precision and segmentation panoptic quality. Our analysis identifies when defogging is effective, the impact of combining models, and how visual language models compare to traditional approaches. We additionally report qualitative rubric-based evaluations from both human and visual language model judges and analyze their alignment with downstream task metrics. Together, these results establish a transparent, task-oriented benchmark for defogging methods and identify the conditions under which pre-processing meaningfully improves autonomous perception in adverse weather. Project page: https://aradfir.github.io/filters-to-vlms-defogging-page/ --- ### LLM as an Algorithmist: Enhancing Anomaly Detectors via Programmatic Synthesis Authors: Hangting Ye, Jinmeng Li, He Zhao, Mingchen Zhuge, Dandan Guo, Yi Chang, Hongyuan Zha Published: 2025-10-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.03904v2 Existing anomaly detection (AD) methods for tabular data usually rely on some assumptions about anomaly patterns, leading to inconsistent performance in real-world scenarios. While Large Language Models (LLMs) show remarkable reasoning capabilities, their direct application to tabular AD is impeded by fundamental challenges, including difficulties in processing heterogeneous data and significant privacy risks. To address these limitations, we propose LLM-DAS, a novel framework that repositions the LLM from a ``data processor'' to an ``algorithmist''. Instead of being exposed to raw data, our framework leverages the LLM's ability to reason about algorithms. It analyzes a high-level description of a given detector to understand its intrinsic weaknesses and then generates detector-specific, data-agnostic Python code to synthesize ``hard-to-detect'' anomalies that exploit these vulnerabilities. This generated synthesis program, which is reusable across diverse datasets, is then instantiated to augment training data, systematically enhancing the detector's robustness by transforming the problem into a more discriminative two-class classification task. Extensive experiments on 36 TAD benchmarks show that LLM-DAS consistently boosts the performance of mainstream detectors. By bridging LLM reasoning with classic AD algorithms via programmatic synthesis, LLM-DAS offers a scalable, effective, and privacy-preserving approach to patching the logical blind spots of existing detectors. --- ### Zero-Shot Fine-Grained Image Classification Using Large Vision-Language Models Authors: Md. Atabuzzaman, Andrew Zhang, Chris Thomas Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03903v1 Large Vision-Language Models (LVLMs) have demonstrated impressive performance on vision-language reasoning tasks. However, their potential for zero-shot fine-grained image classification, a challenging task requiring precise differentiation between visually similar categories, remains underexplored. We present a novel method that transforms zero-shot fine-grained image classification into a visual question-answering framework, leveraging LVLMs' comprehensive understanding capabilities rather than relying on direct class name generation. We enhance model performance through a novel attention intervention technique. We also address a key limitation in existing datasets by developing more comprehensive and precise class description benchmarks. We validate the effectiveness of our method through extensive experimentation across multiple fine-grained image classification benchmarks. Our proposed method consistently outperforms the current state-of-the-art (SOTA) approach, demonstrating both the effectiveness of our method and the broader potential of LVLMs for zero-shot fine-grained classification tasks. Code and Datasets: https://github.com/Atabuzzaman/Fine-grained-classification --- ### Optimal Computation from Fluctuation Responses Authors: Jinghao Lyu, Kyle J. Ray, James P. Crutchfield Published: 2025-10-04 Categories: cond-mat.stat-mech, cs.LG arXiv: https://arxiv.org/abs/2510.03900v1 The energy cost of computation has emerged as a central challenge at the intersection of physics and computer science. Recent advances in statistical physics -- particularly in stochastic thermodynamics -- enable precise characterizations of work, heat, and entropy production in information-processing systems driven far from equilibrium by time-dependent control protocols. A key open question is then how to design protocols that minimize thermodynamic cost while ensur- ing correct outcomes. To this end, we develop a unified framework to identify optimal protocols using fluctuation response relations (FRR) and machine learning. Unlike previous approaches that optimize either distributions or protocols separately, our method unifies both using FRR-derived gradients. Moreover, our method is based primarily on iteratively learning from sampled noisy trajectories, which is generally much easier than solving for the optimal protocol directly from a set of governing equations. We apply the framework to canonical examples -- bit erasure in a double-well potential and translating harmonic traps -- demonstrating how to construct loss functions that trade-off energy cost against task error. The framework extends trivially to underdamped systems, and we show this by optimizing a bit-flip in an underdamped system. In all computations we test, the framework achieves the theoretically optimal protocol or achieves work costs comparable to relevant finite time bounds. In short, the results provide principled strategies for designing thermodynamically efficient protocols in physical information-processing systems. Applications range from quantum gates robust under noise to energy-efficient control of chemical and synthetic biological networks. --- ### Fair Minimum Labeling: Efficient Temporal Network Activations for Reachability and Equity Authors: Lutz Oettershagen, Othon Michail Published: 2025-10-04 Categories: cs.SI, cs.DS, cs.LG arXiv: https://arxiv.org/abs/2510.03899v3 Balancing resource efficiency and fairness is critical in networked systems that support modern learning applications. We introduce the \emph{Fair Minimum Labeling} (FML) problem: the task of designing a minimum-cost temporal edge activation plan that ensures each group of nodes in a network has sufficient access to a designated target set, according to specified coverage requirements. FML captures key trade-offs in systems where edge activations incur resource costs and equitable access is essential, such as distributed data collection, update dissemination in edge-cloud systems, and fair service restoration in critical infrastructure. We first give a structural characterisation of the single-terminal case, showing that it is equivalent to the rooted Covering Steiner problem. We prove that FML is NP-hard and admits no $((1-ε)\ln |\mathcal{C}|)$-approximation for $|\mathcal{C}|$ groups, already on a star, while for any fixed number of groups it inherits a constant-factor approximation and remains APX-hard. We then present probabilistic approximation algorithms for the two-group, single-terminal case: an algorithm whose tree subroutine is exact, hence optimal on tree-structured networks and $\mathcal{O}(\log |V|)$ in expectation on general graphs, together with a faster bicriteria variant whose coverage violation degrades gracefully with the merge depth of the tree computation. For practical scalability, we additionally introduce a graph-native variant based on a shortest-path-tree reduction. Empirical results show that FML enforces group-level fairness, while the graph-native variant substantially improves scalability and achieves competitive activation cost. --- ### Read Between the Lines: A Benchmark for Uncovering Political Bias in Bangla News Articles Authors: Nusrat Jahan Lia, Shubhashis Roy Dipta, Abdullah Khan Zehady, Naymul Islam, Madhusodan Chakraborty, Abdullah Al Wasif Published: 2025-10-04 Categories: cs.CL arXiv: https://arxiv.org/abs/2510.03898v2 Detecting media bias is crucial, specifically in the South Asian region. Despite this, annotated datasets and computational studies for Bangla political bias research remain scarce. Crucially because, political stance detection in Bangla news requires understanding of linguistic cues, cultural context, subtle biases, rhetorical strategies, code-switching, implicit sentiment, and socio-political background. To address this, we introduce the first benchmark dataset of 200 politically significant and highly debated Bangla news articles, labeled for government-leaning, government-critique, and neutral stances, alongside diagnostic analyses for evaluating large language models (LLMs). Our comprehensive evaluation of 28 proprietary and open-source LLMs shows strong performance in detecting government-critique content (F1 up to 0.83) but substantial difficulty with neutral articles (F1 as low as 0.00). Models also tend to over-predict government-leaning stances, often misinterpreting ambiguous narratives. This dataset and its associated diagnostics provide a foundation for advancing stance detection in Bangla media research and offer insights for improving LLM performance in low-resource languages. --- ### GAE: Unleashing Physical Potential of VLM with Generalizable Action Expert Authors: Mingyu Liu, Zheng Huang, Xiaoyi Lin, Muzhi Zhu, Canyu Zhao, Yating Wang, Haoyi Zhu, Hao Chen, Chunhua Shen Published: 2025-10-04 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2510.03896v2 Vision-language models demonstrate strong reasoning and planning abilities, yet grounding these predictions into precise robot actions remains a central challenge. Existing Vision-Language-Action methods typically entangle reasoning and action generation, leading to limited generalization. We propose Generalizable Action Expert (GAE), a task-agnostic model that converts sparse geometric plans into dense robot actions. Our approach introduces a sparse geometric interface: the VLM predicts sparse 3D waypoints representing high-level intention, while GAE maps these waypoints together with real-time point cloud observations to continuous action trajectories. GAE is pretrained on a large-scale pointcloud-trajectory dataset comprising 150k trajectories from both simulation and real-world robots. To further improve efficiency and generalization, we introduce an Action Pre-training, Pointcloud Fine-tuning (APPF) scheme that decouples learning action dynamics from geometry grounding. After pretraining, GAE is frozen and reused across downstream tasks, requiring only lightweight fine-tuning of the VLM to produce the sparse interface. Experiments show that our method achieves strong performance and generalization across diverse visual domains, camera viewpoints, and natural language instructions. --- ### NoTVLA: Semantics-Preserving Robot Adaptation via Narrative Action Interfaces Authors: Zheng Huang, Mingyu Liu, Xiaoyi Lin, Muzhi Zhu, Canyu Zhao, Zongze Du, Ye Lin, Xiaoman Li, Yiduo Jia, Hao Zhong, Hao Chen, Chunhua Shen Published: 2025-10-04 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2510.03895v2 Vision-Language-Action (VLA) models represent a pivotal advance in embodied intelligence, yet they confront critical barriers to real-world deployment, most notably catastrophic forgetting. This issue stems from their overreliance on continuous action sequences or action chunks, which inadvertently create isolated data silos that disrupt knowledge retention across tasks. To tackle these challenges, we propose the Narrowing of Trajectory VLA (NoTVLA) framework: a novel approach that narrows its focus to sparse trajectories, thereby avoiding the catastrophic forgetting associated with dense trajectory fine-tuning. A key innovation of NoTVLA lies in its trajectory planning strategy: instead of centering on the target object's trajectory, it leverages temporal compression and spatial reasoning pruning specifically for the robot end effector's trajectory. Furthermore, training is conducted using these sparse trajectories rather than dense action trajectories, an optimization that delivers remarkable practical advantages with better performance in zero-shot. In multi-task evaluation scenarios, NoTVLA achieves superior performance and generalization compared to pi0 while operating under two critical constraints: it uses over an order of magnitude less computing power than pi0 and requires no wrist-mounted camera. This design ensures that NoTVLA's operational accuracy closely approximates that of single-task expert models. Crucially, it also preserves the model's inherent language capabilities, enabling zero-shot generalization in specific scenarios, supporting unified model deployment across multiple robot platforms, and fostering a degree of generalization even when perceiving tasks from novel perspectives. --- ### BONSAI: Structure-exploiting robust Bayesian optimization for networked black-box systems under uncertainty Authors: Akshay Kudva, Joel A. Paulson Published: 2025-10-04 Categories: cs.LG, math.OC arXiv: https://arxiv.org/abs/2510.03893v1 Optimal design under uncertainty remains a fundamental challenge in advancing reliable, next-generation process systems. Robust optimization (RO) offers a principled approach by safeguarding against worst-case scenarios across a range of uncertain parameters. However, traditional RO methods typically require known problem structure, which limits their applicability to high-fidelity simulation environments. To overcome these limitations, recent work has explored robust Bayesian optimization (RBO) as a flexible alternative that can accommodate expensive, black-box objectives. Existing RBO methods, however, generally ignore available structural information and struggle to scale to high-dimensional settings. In this work, we introduce BONSAI (Bayesian Optimization of Network Systems under uncertAInty), a new RBO framework that leverages partial structural knowledge commonly available in simulation-based models. Instead of treating the objective as a monolithic black box, BONSAI represents it as a directed graph of interconnected white- and black-box components, allowing the algorithm to utilize intermediate information within the optimization process. We further propose a scalable Thompson sampling-based acquisition function tailored to the structured RO setting, which can be efficiently optimized using gradient-based methods. We evaluate BONSAI across a diverse set of synthetic and real-world case studies, including applications in process systems engineering. Compared to existing simulation-based RO algorithms, BONSAI consistently delivers more sample-efficient and higher-quality robust solutions, highlighting its practical advantages for uncertainty-aware design in complex engineering systems. --- ### Kantian-Utilitarian XAI: Meta-Explained Authors: Zahra Atf, Peter R. Lewis Published: 2025-10-04 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2510.03892v1 We present a gamified explainable AI (XAI) system for ethically aware consumer decision-making in the coffee domain. Each session comprises six rounds with three options per round. Two symbolic engines provide real-time reasons: a Kantian module flags rule violations (e.g., child labor, deforestation risk without shade certification, opaque supply chains, unsafe decaf), and a utilitarian module scores options via multi-criteria aggregation over normalized attributes (price, carbon, water, transparency, farmer income share, taste/freshness, packaging, convenience). A meta-explainer with a regret bound (0.2) highlights Kantian--utilitarian (mis)alignment and switches to a deontically clean, near-parity option when welfare loss is small. We release a structured configuration (attribute schema, certification map, weights, rule set), a policy trace for auditability, and an interactive UI. --- ### Rare Text Semantics Were Always There in Your Diffusion Transformer Authors: Seil Kang, Woojung Han, Dayun Ju, Seong Jae Hwang Published: 2025-10-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2510.03886v1 Starting from flow- and diffusion-based transformers, Multi-modal Diffusion Transformers (MM-DiTs) have reshaped text-to-vision generation, gaining acclaim for exceptional visual fidelity. As these models advance, users continually push the boundary with imaginative or rare prompts, which advanced models still falter in generating, since their concepts are often too scarce to leave a strong imprint during pre-training. In this paper, we propose a simple yet effective intervention that surfaces rare semantics inside MM-DiTs without additional training steps, data, denoising-time optimization, or reliance on external modules (e.g., large language models). In particular, the joint-attention mechanism intrinsic to MM-DiT sequentially updates text embeddings alongside image embeddings throughout transformer blocks. We find that by mathematically expanding representational basins around text token embeddings via variance scale-up before the joint-attention blocks, rare semantics clearly emerge in MM-DiT's outputs. Furthermore, our results generalize effectively across text-to-vision tasks, including text-to-image, text-to-video, and text-driven image editing. Our work invites generative models to reveal the semantics that users intend, once hidden yet ready to surface. --- ### Exploring Instruction Data Quality for Explainable Image Quality Assessment Authors: Yunhao Li, Sijing Wu, Jun Jia, Kang Fu, Qi Jia, Yucheng Zhu, Wei Sun, Guangtao Zhai Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03880v2 In recent years, with the rapid development of large multimodal models (LMMs), explainable image quality assessment (IQA) has attracted increasing attention, aiming to understand the perceptual quality problems of images. Existing studies typically construct large-scale instruction tuning datasets to enhance the quality perception capabilities of LMMs, following the data scaling law. However, as the fundamental capabilities of LMMs continue to improve, existing instruction tuning datasets may contain redundant and less challenging samples, resulting in substantial computational costs. In this paper, we investigate whether data quantity remains the dominant factor in LMM instruction tuning for explainable IQA. Based on a strong pre-trained LMM, we observe that randomly selecting a properly sized subset of training data can outperform full-data fine-tuning, indicating substantial redundancy in existing instruction tuning datasets. Motivated by this observation, we propose Q-Selector, a clustering-based data selection framework consisting of three stages: hierarchical LMM-based clustering feature extraction, cluster quota allocation through density and transferability, and SVD-based cluster sampling strategy. Specifically, Q-Selector extracts multi-layer features from the vision encoder, text encoder, and large language model components of LMMs as clustering features. It then determines the sampling quota for each cluster based on inter-cluster transferability and intra-cluster density. Finally, Q-Selector employs a Singular Value Decomposition (SVD)-based sampling strategy to select high-quality instruction data. Experimental results demonstrate that Q-Selector achieves 102.1% and 103.7% of the performance of full-data fine-tuning using only 10% of the training data on explainable IQA and image aesthetics assessment tasks, respectively. --- ### Adversarial Agent Collaboration for Correctness Improvements of C to Safe Rust Translation Authors: Tianyu Li, Ruishi Li, Bo Wang, Brandon Paulsen, Umang Mathur, Prateek Saxena Published: 2025-10-04 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2510.03879v3 Translating C to memory-safe languages, like Rust, prevents critical memory safety vulnerabilities that are prevalent in legacy C software. Even with recent LLM-based and tool-augmented translators, the resulting Rust code frequently diverges from the C source on inputs absent from the test suite used during translation; this correctness gap on unseen inputs remains a dominant obstacle to reliable, automatic C-to-Rust translation. In this work, we present ACToR (Adversarial C To Rust), a simple LLM-agent loop that closes this gap by adversarially searching for inputs on which the translation diverges from the C source, and using them to drive subsequent refinements. Inspired by GANs, ACToR pits a translator agent against a discriminator agent that collaborate to iteratively refine the Rust translation. On each iteration, the translator agent synthesizes and refines a Rust translation to pass an existing suite of tests, and then the discriminator agent finds new failing tests by constructing and refining a differential fuzzer over the C and Rust binaries. Across 63 real-world command-line C utilities, with an average size of 473 lines of code and the longest reaching thousands of lines in size, ACToR achieves over 90% test pass rate with zero human intervention. The improvement holds across seven agent-LLM configurations on our micro-benchmark, indicating that the loop is largely independent of the choice of underlying translator and LLM. Compared to a non-adversarial, coverage-driven test-generation baseline, ACToR improves correctness by up to 36.7%. When applied on top of one recent translator, C2SaferRust, ACToR further improves the validation pass rate by 16.6%. --- ### Multi-Modal Oral Cancer Detection Using Weighted Ensemble Convolutional Neural Networks Authors: Ajo Babu George, Sreehari J R Ajo Babu George, Sreehari J R Ajo Babu George, Sreehari J R Published: 2025-10-04 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.03878v1 Aims Late diagnosis of Oral Squamous Cell Carcinoma (OSCC) contributes significantly to its high global mortality rate, with over 50\% of cases detected at advanced stages and a 5-year survival rate below 50\% according to WHO statistics. This study aims to improve early detection of OSCC by developing a multimodal deep learning framework that integrates clinical, radiological, and histopathological images using a weighted ensemble of DenseNet-121 convolutional neural networks (CNNs). Material and Methods A retrospective study was conducted using publicly available datasets representing three distinct medical imaging modalities. Each modality-specific dataset was used to train a DenseNet-121 CNN via transfer learning. Augmentation and modality-specific preprocessing were applied to increase robustness. Predictions were fused using a validation-weighted ensemble strategy. Evaluation was performed using accuracy, precision, recall, F1-score. Results High validation accuracy was achieved for radiological (100\%) and histopathological (95.12\%) modalities, with clinical images performing lower (63.10\%) due to visual heterogeneity. The ensemble model demonstrated improved diagnostic robustness with an overall accuracy of 84.58\% on a multimodal validation dataset of 55 samples. Conclusion The multimodal ensemble framework bridges gaps in the current diagnostic workflow by offering a non-invasive, AI-assisted triage tool that enhances early identification of high-risk lesions. It supports clinicians in decision-making, aligning with global oncology guidelines to reduce diagnostic delays and improve patient outcomes. --- ### Skin Lesion Classification Based on ResNet-50 Enhanced With Adaptive Spatial Feature Fusion Authors: Runhao Liu, Fengyi Zha, Fei Ding, Guangzhen Yao, Peng Zhang Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03876v2 Skin cancer classification is challenging due to high inter-class similarity, intra-class variability, and artifacts in dermoscopic images. To address these issues, we propose an improved ResNet-50 with Adaptive Spatial Feature Fusion (ASFF), which adaptively integrates multi-scale semantic and surface features to refine representations and reduce overfitting. The ResNet-50 model is enhanced with an adaptive feature fusion mechanism to achieve more effective multi-scale feature extraction and improve overall performance. Specifically, a dual-branch design fuses high-level semantic and mid-level detail features which use global average pooling and fully connected layers to produce spatial weights, and emphasizes lesion-relevant regions. Evaluated on a balanced subset of ISIC 2020 (3,297 images, randomly selected from the original dataset), the ASFF-based ResNet-50 outperforms multiple CNN baselines, achieving 93.182% accuracy with superior precision, recall, specificity, and F1. It also reaches 0.9670 AUC (P-R) and 0.9717 AUC (ROC). Grad-CAM visualizations show more accurate focus on lesion areas.The proposed model also generalizes well to ISIC 2019 external validation, outperforming the ResNet-50 baseline. These findings demonstrate that the proposed approach provides a more effective and efficient solution for computer-aided skin cancer diagnosis. The generation codes, weights and confusion matrices are open sourced in https://github.com/Grapesea/ASFF-ResNet50-enhanced. --- ### DHQA-4D: Perceptual Quality Assessment of Dynamic 4D Digital Human Authors: Yunhao Li, Sijing Wu, Yucheng Zhu, Huiyu Duan, Zicheng Zhang, Guangtao Zhai Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03874v1 With the rapid development of 3D scanning and reconstruction technologies, dynamic digital human avatars based on 4D meshes have become increasingly popular. A high-precision dynamic digital human avatar can be applied to various fields such as game production, animation generation, and remote immersive communication. However, these 4D human avatar meshes are prone to being degraded by various types of noise during the processes of collection, compression, and transmission, thereby affecting the viewing experience of users. In light of this fact, quality assessment of dynamic 4D digital humans becomes increasingly important. In this paper, we first propose a large-scale dynamic digital human quality assessment dataset, DHQA-4D, which contains 32 high-quality real-scanned 4D human mesh sequences, 1920 distorted textured 4D human meshes degraded by 11 textured distortions, as well as their corresponding textured and non-textured mean opinion scores (MOSs). Equipped with DHQA-4D dataset, we analyze the influence of different types of distortion on human perception for textured dynamic 4D meshes and non-textured dynamic 4D meshes. Additionally, we propose DynaMesh-Rater, a novel large multimodal model (LMM) based approach that is able to assess both textured 4D meshes and non-textured 4D meshes. Concretely, DynaMesh-Rater elaborately extracts multi-dimensional features, including visual features from a projected 2D video, motion features from cropped video clips, and geometry features from the 4D human mesh to provide comprehensive quality-related information. Then we utilize a LMM model to integrate the multi-dimensional features and conduct a LoRA-based instruction tuning technique to teach the LMM model to predict the quality scores. Extensive experimental results on the DHQA-4D dataset demonstrate the superiority of our DynaMesh-Rater method over previous quality assessment methods. --- ### PoseGaze-AHP: A Knowledge-Based 3D Dataset for AI-Driven Ocular and Postural Diagnosis Authors: Saja Al-Dabet, Sherzod Turaev, Nazar Zaki, Arif O. Khan, Luai Eldweik Published: 2025-10-04 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.03873v1 Diagnosing ocular-induced abnormal head posture (AHP) requires a comprehensive analysis of both head pose and ocular movements. However, existing datasets focus on these aspects separately, limiting the development of integrated diagnostic approaches and restricting AI-driven advancements in AHP analysis. To address this gap, we introduce PoseGaze-AHP, a novel 3D dataset that synchronously captures head pose and gaze movement information for ocular-induced AHP assessment. Structured clinical data were extracted from medical literature using large language models (LLMs) through an iterative process with the Claude 3.5 Sonnet model, combining stepwise, hierarchical, and complex prompting strategies. The extracted records were systematically imputed and transformed into 3D representations using the Neural Head Avatar (NHA) framework. The dataset includes 7,920 images generated from two head textures, covering a broad spectrum of ocular conditions. The extraction method achieved an overall accuracy of 91.92%, demonstrating its reliability for clinical dataset construction. PoseGaze-AHP is the first publicly available resource tailored for AI-driven ocular-induced AHP diagnosis, supporting the development of accurate and privacy-compliant diagnostic tools. --- ### Optimal Scaling Needs Optimal Norm Authors: Oleg Filatov, Jiangtao Wang, Jan Ebert, Stefan Kesselheim Published: 2025-10-04 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2510.03871v2 Despite recent progress in optimal hyperparameter transfer under model and dataset scaling, no unifying explanatory principle has been established. For Adam and Scion optimizers, we discover that joint optimal scaling across model and dataset sizes is conditioned on a single invariant: the operator norm of the output layer. Across models with up to 1.3B parameters trained on up to 138B tokens, the optimal learning rate/batch size pair $(η^{\ast}, B^{\ast})$ consistently has the same operator norm value - a phenomenon we term norm transfer. This constant norm condition is necessary but not sufficient: while for each dataset size, multiple $(η, B)$ reach the optimal norm, only a unique $(η^{\ast}, B^{\ast})$ achieves the best loss. As a sufficient condition, we provide the first measurement of $(η^{\ast}, B^{\ast})$ scaling with dataset size for Scion, and find that the scaling rules are consistent with those of Adam. Tuning per-layer-group learning rates also improves model performance, with the output layer being the most sensitive and hidden layers benefiting from lower learning rates. We provide practical insights on norm-guided optimal scaling and release our Distributed Scion (Disco) implementation with logs from over two thousand runs to support research on LLM training dynamics at scale. --- ### SDAKD: Student Discriminator Assisted Knowledge Distillation for Super-Resolution Generative Adversarial Networks Authors: Nikolaos Kaparinos, Vasileios Mezaris Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03870v1 Generative Adversarial Networks (GANs) achieve excellent performance in generative tasks, such as image super-resolution, but their computational requirements make difficult their deployment on resource-constrained devices. While knowledge distillation is a promising research direction for GAN compression, effectively training a smaller student generator is challenging due to the capacity mismatch between the student generator and the teacher discriminator. In this work, we propose Student Discriminator Assisted Knowledge Distillation (SDAKD), a novel GAN distillation methodology that introduces a student discriminator to mitigate this capacity mismatch. SDAKD follows a three-stage training strategy, and integrates an adapted feature map distillation approach in its last two training stages. We evaluated SDAKD on two well-performing super-resolution GANs, GCFSR and Real-ESRGAN. Our experiments demonstrate consistent improvements over the baselines and SOTA GAN knowledge distillation methods. The SDAKD source code will be made openly available upon acceptance of the paper. --- ### Exploring the Challenge and Value of Deep Learning in Automated Skin Disease Diagnosis Authors: Runhao Liu, Ziming Chen, Guangzhen Yao, Peng Zhang Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03869v2 Skin cancer is one of the most prevalent and deadly forms of cancer worldwide, highlighting the critical importance of early detection and diagnosis in improving patient outcomes. Deep learning (DL) has shown significant promise in enhancing the accuracy and efficiency of automated skin disease diagnosis, particularly in detecting and classifying skin lesions. However, several challenges remain for DL-based skin cancer diagnosis, including complex features, image noise, intra-class variation, inter-class similarity, and data imbalance. This review synthesizes recent research and discusses innovative approaches to address these challenges, such as data augmentation, hybrid models, and feature fusion. Furthermore, the review highlights the integration of DL models into clinical workflows, offering insights into the potential of deep learning to revolutionize skin disease diagnosis and improve clinical decision-making. This review uniquely integrates a PRISMA-based methodology with a challenge-oriented taxonomy, providing a systematic and transparent synthesis of recent deep learning advances for skin disease diagnosis. It further highlights emerging directions such as hybrid CNN-Transformer architectures and uncertainty-aware models, emphasizing its contribution to future dermatological AI research. --- ### AI Adoption Across Mission-Driven Organizations Authors: Dalia Ali, Muneeb Ahmed, Hailan Wang, Arfa Khan, Naira Paola Arnez Jordan, Sunnie S. Y. Kim, Meet Dilip Muchhala, Anne Kathrin Merkle, Orestis Papakyriakopoulos Published: 2025-10-04 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2510.03868v2 Despite AI's promise, little is known about how mission-driven organizations (MDOs) adopt AI. We conducted semi-structured interviews with 15 experts and leaders within environmental, humanitarian, and development organizations across the Global North and South. We find that MDOs adopt AI selectively, with uses in content creation and data analysis, while maintaining human oversight for mission-critical applications. In case of conflicts between the organization's values and efficiency, participants consider sustainability, neutrality, and community trust as constraints that could limit or prevent AI adoption. Individual adoption of AI is not translated into organizational practice due to fragmented data infrastructure, lack of in-house expertise, inertia, ethical challenges, and dependency on vendors. Participants expected AI use in MDOs to involve institutional control over infrastructure, support for missions, and human-centered human-AI collaboration. We contribute an analysis on how MDOs manage AI adoption, identify barriers to responsible deployment, and provide recommendations for designing mission-aligned AI systems. --- ### On Provable Benefits of Muon in Federated Learning Authors: Xinwen Zhang, Hongchang Gao Published: 2025-10-04 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2510.03866v1 The recently introduced optimizer, Muon, has gained increasing attention due to its superior performance across a wide range of applications. However, its effectiveness in federated learning remains unexplored. To address this gap, this paper investigates the performance of Muon in the federated learning setting. Specifically, we propose a new algorithm, FedMuon, and establish its convergence rate for nonconvex problems. Our theoretical analysis reveals multiple favorable properties of FedMuon. In particular, due to its orthonormalized update direction, the learning rate of FedMuon is independent of problem-specific parameters, and, importantly, it can naturally accommodate heavy-tailed noise. The extensive experiments on a variety of neural network architectures validate the effectiveness of the proposed algorithm. --- ### Unlocking Reasoning Capabilities in LLMs via Reinforcement Learning Exploration Authors: Wenhao Deng, Long Wei, Chenglei Yu, Tailin Wu Published: 2025-10-04 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2510.03865v2 Reinforcement learning with verifiable rewards (RLVR) has recently enhanced the reasoning capabilities of large language models (LLMs), particularly for mathematical problem solving. However, a fundamental limitation remains: as the sampling budget increases, the advantage of RLVR-trained models over their pretrained bases often diminishes or even vanishes, revealing a strong dependence on the base model's restricted search space. We attribute this phenomenon to the widespread use of the reverse Kullback-Leibler (KL) divergence regularizer, whose mode-seeking behavior keeps the policy trapped inside the base model's support region and hampers wider exploration. To address this issue, we propose RAPO (Rewards-Aware Policy Optimization), an algorithm to promote broader yet focused exploration. Our method (i) utilizes the forward KL penalty to replace the reverse KL penalty for out-of-distribution exploration, and (ii) reweights the reference policy to facilitate adaptive in-distribution exploration. We train Qwen2.5-3B and 7B models with RAPO on the 8K SimpleRL-Zero dataset, without supervised fine-tuning, and evaluate them on AIME2024 and AIME2025. Results show that RAPO consistently improves problem-solving performance. Notably, RAPO enables models to surpass the base model's performance ceiling and solves previously intractable problems, advancing the frontier of RLVR for challenging reasoning tasks. --- ### Spatial CAPTCHA: Generatively Benchmarking Spatial Reasoning for Human-Machine Differentiation Authors: Arina Kharlamova, Bowei He, Chen Ma, Xue Liu Published: 2025-10-04 Categories: cs.AI, cs.CR arXiv: https://arxiv.org/abs/2510.03863v1 Online services rely on CAPTCHAs as a first line of defense against automated abuse, yet recent advances in multi-modal large language models (MLLMs) have eroded the effectiveness of conventional designs that focus on text recognition or 2D image understanding. To address this challenge, we present Spatial CAPTCHA, a novel human-verification framework that leverages fundamental differences in spatial reasoning between humans and MLLMs. Unlike existing CAPTCHAs which rely on low-level perception tasks that are vulnerable to modern AI, Spatial CAPTCHA generates dynamic questions requiring geometric reasoning, perspective-taking, occlusion handling, and mental rotation. These skills are intuitive for humans but difficult for state-of-the-art (SOTA) AI systems. The system employs a procedural generation pipeline with constraint-based difficulty control, automated correctness verification, and human-in-the-loop validation to ensure scalability, robustness, and adaptability. Evaluation on a corresponding benchmark, Spatial-CAPTCHA-Bench, demonstrates that humans vastly outperform 10 state-of-the-art MLLMs, with the best model achieving only 31.0% Pass@1 accuracy. Furthermore, we compare Spatial CAPTCHA with Google reCAPTCHA, which confirms its effectiveness as both a security mechanism and a diagnostic tool for spatial reasoning in AI. --- ### Designing Empirical Studies on LLM-Based Code Generation: Towards a Reference Framework Authors: Nathalia Nascimento, Everton Guimaraes, Paulo Alencar Published: 2025-10-04 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2510.03862v1 The rise of large language models (LLMs) has introduced transformative potential in automated code generation, addressing a wide range of software engineering challenges. However, empirical evaluation of LLM-based code generation lacks standardization, with studies varying widely in goals, tasks, and metrics, which limits comparability and reproducibility. In this paper, we propose a theoretical framework for designing and reporting empirical studies on LLM-based code generation. The framework is grounded in both our prior experience conducting such experiments and a comparative analysis of key similarities and differences among recent studies. It organizes evaluation around core components such as problem sources, quality attributes, and metrics, supporting structured and systematic experimentation. We demonstrate its applicability through representative case mappings and identify opportunities for refinement. Looking forward, we plan to evolve the framework into a more robust and mature tool for standardizing LLM evaluation across software engineering contexts. --- ### Adaptive and Explainable AI Agents for Anomaly Detection in Critical IoT Infrastructure using LLM-Enhanced Contextual Reasoning Authors: Raghav Sharma, Manan Mehta Published: 2025-10-04 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2510.03859v1 Ensuring that critical IoT systems function safely and smoothly depends a lot on finding anomalies quickly. As more complex systems, like smart healthcare, energy grids and industrial automation, appear, it is easier to see the shortcomings of older methods of detection. Monitoring failures usually happen in dynamic, high dimensional situations, especially when data is incomplete, messy or always evolving. Such limits point out the requirement for adaptive, intelligent systems that always improve and think. LLMs are now capable of significantly changing how context is understood and semantic inference is done across all types of data. This proposal suggests using an LLM supported contextual reasoning method along with XAI agents to improve how anomalies are found in significant IoT environments. To discover hidden patterns and notice inconsistencies in data streams, it uses attention methods, avoids dealing with details from every time step and uses memory buffers with meaning. Because no code AI stresses transparency and interpretability, people can check and accept the AI's decisions, helping ensure AI follows company policies. The two architectures are put together in a test that compares the results of the traditional model with those of the suggested LLM enhanced model. Important measures to check are the accuracy of detection, how much inaccurate information is included in the results, how clearly the findings can be read and how fast the system responds under different test situations. The metaheuristic is tested in simulations of real world smart grid and healthcare contexts to check its adaptability and reliability. From the study, we see that the new approach performs much better than most existing models in both accuracy and interpretation, so it could be a good fit for future anomaly detection tasks in IoT --- ### Cross-View Open-Vocabulary Object Detection in Aerial Imagery Authors: Jyoti Kini, Rohit Gupta, Mubarak Shah Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03858v1 Traditional object detection models are typically trained on a fixed set of classes, limiting their flexibility and making it costly to incorporate new categories. Open-vocabulary object detection addresses this limitation by enabling models to identify unseen classes without explicit training. Leveraging pretrained models contrastively trained on abundantly available ground-view image-text classification pairs provides a strong foundation for open-vocabulary object detection in aerial imagery. Domain shifts, viewpoint variations, and extreme scale differences make direct knowledge transfer across domains ineffective, requiring specialized adaptation strategies. In this paper, we propose a novel framework for adapting open-vocabulary representations from ground-view images to solve object detection in aerial imagery through structured domain alignment. The method introduces contrastive image-to-image alignment to enhance the similarity between aerial and ground-view embeddings and employs multi-instance vocabulary associations to align aerial images with text embeddings. Extensive experiments on the xView, DOTAv2, VisDrone, DIOR, and HRRSD datasets are used to validate our approach. Our open-vocabulary model achieves improvements of +6.32 mAP on DOTAv2, +4.16 mAP on VisDrone (Images), and +3.46 mAP on HRRSD in the zero-shot setting when compared to finetuned closed-vocabulary dataset-specific model performance, thus paving the way for more flexible and scalable object detection systems in aerial applications. --- ### Optimized Minimal 4D Gaussian Splatting Authors: Minseo Lee, Byeonghyeon Lee, Lucas Yunkyu Lee, Eunsoo Lee, Sangmin Kim, Seunghyeon Song, Joo Chan Lee, Jong Hwan Ko, Jaesik Park, Eunbyung Park Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03857v1 4D Gaussian Splatting has emerged as a new paradigm for dynamic scene representation, enabling real-time rendering of scenes with complex motions. However, it faces a major challenge of storage overhead, as millions of Gaussians are required for high-fidelity reconstruction. While several studies have attempted to alleviate this memory burden, they still face limitations in compression ratio or visual quality. In this work, we present OMG4 (Optimized Minimal 4D Gaussian Splatting), a framework that constructs a compact set of salient Gaussians capable of faithfully representing 4D Gaussian models. Our method progressively prunes Gaussians in three stages: (1) Gaussian Sampling to identify primitives critical to reconstruction fidelity, (2) Gaussian Pruning to remove redundancies, and (3) Gaussian Merging to fuse primitives with similar characteristics. In addition, we integrate implicit appearance compression and generalize Sub-Vector Quantization (SVQ) to 4D representations, further reducing storage while preserving quality. Extensive experiments on standard benchmark datasets demonstrate that OMG4 significantly outperforms recent state-of-the-art methods, reducing model sizes by over 60% while maintaining reconstruction quality. These results position OMG4 as a significant step forward in compact 4D scene representation, opening new possibilities for a wide range of applications. Our source code is available at https://minshirley.github.io/OMG4/. --- ### AI-Assisted Pleural Effusion Volume Estimation from Contrast-Enhanced CT Images Authors: Sanhita Basu, Tomas Fröding, Ali Teymur Kahraman, Dimitris Toumpanakis, Tobias Sjöblom Published: 2025-10-04 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2510.03856v1 Background: Pleural Effusions (PE) is a common finding in many different clinical conditions, but accurately measuring their volume from CT scans is challenging. Purpose: To improve PE segmentation and quantification for enhanced clinical management, we have developed and trained a semi-supervised deep learning framework on contrast-enhanced CT volumes. Materials and Methods: This retrospective study collected CT Pulmonary Angiogram (CTPA) data from internal and external datasets. A subset of 100 cases was manually annotated for model training, while the remaining cases were used for testing and validation. A novel semi-supervised deep learning framework, Teacher-Teaching Assistant-Student (TTAS), was developed and used to enable efficient training in non-segmented examinations. Segmentation performance was compared to that of state-of-the-art models. Results: 100 patients (mean age, 72 years, 28 (standard deviation); 55 men) were included in the study. The TTAS model demonstrated superior segmentation performance compared to state-of-the-art models, achieving a mean Dice score of 0.82 (95% CI, 0.79 - 0.84) versus 0.73 for nnU-Net (p < 0.0001, Student's T test). Additionally, TTAS exhibited a four-fold lower mean Absolute Volume Difference (AbVD) of 6.49 mL (95% CI, 4.80 - 8.20) compared to nnU-Net's AbVD of 23.16 mL (p < 0.0001). Conclusion: The developed TTAS framework offered superior PE segmentation, aiding accurate volume determination from CT scans. --- ### UGround: Towards Unified Visual Grounding with Unrolled Transformers Authors: Rui Qian, Xin Yin, Chuanhang Deng, Zhiyuan Peng, Jian Xiong, Wei Zhai, Dejing Dou Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03853v4 We present UGround, a \textbf{U}nified visual \textbf{Ground}ing paradigm that dynamically selects intermediate layers across \textbf{U}nrolled transformers as ``mask as prompt,'' diverging from the prevailing pipeline that leverages the fixed last hidden layer as ``\texttt{<SEG>} as prompt.'' UGround addresses two primary challenges posed by the prevailing paradigm: (1) its reliance on the fixed last hidden layer, which sequentially amplifies cumulative errors arising from layer-by-layer propagation without intermediate correction, and (2) its use of \texttt{<SEG>} as a prompt, which implicitly projects textual embeddings into visual space without explicit spatial cues (e.g., coordinates). Central to UGround is Policy-Prompted Masking, which comprises two key components: Stochastic Skip Connection (SSC) and Mask as Prompt (MasP). SSC is a reinforcement learning policy that, via stochastic sampling, allows each \texttt{<SEG>} token to slide across unrolled transformer layers, enabling dynamic layer selection at which it connects to the vision model (e.g., SAM) in a skip-connection fashion. Given the selected hidden layer, MasP uses the similarity map derived from the \texttt{<SEG>} token and image tokens as a soft logit mask to prompt SAM for mask generation, offering explicit spatial cues through its activation regions. To validate the effectiveness of UGround, we, for the first time, have unified visual grounding within a single framework from an attribute perspective, spanning from traditional refer expression segmentation to newly proposed reasoning segmentation, single-target to multi-target, positive query to false premise (empty target). All code and models are publicly available at https://github.com/rui-qian/UGround. --- ### MetaMuse: Algorithm Generation via Creative Ideation Authors: Ruiying Ma, Chieh-Jan Mike Liang, Yanjie Gao, Francis Y. Yan Published: 2025-10-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2510.03851v2 Designing system algorithms remains challenging, where the discontinuous nature of the solution space often forces system engineers to rely on generic heuristics at the expense of performance. We study whether LLMs can practically drive algorithm generation, and find that they are biased towards well-known generic designs, rather than making the creative leaps needed to navigate the discontinuous solution space. To address this limitation, we introduce MetaMuse, a framework for creative ideation built on three self-reflection principles: (1) quantifying solution diversity and usefulness in measurable performance space, rather than abstract idea space, (2) steering ideation through external stimuli, rather than internal randomness, and (3) constructing executable solutions using waypoint reasoning, rather than free-form chain-of-thought. Considering two critical online problems at a global cloud provider, extensive evaluations show that MetaMuse can generate high-performing solutions: it reduces cache misses by up to 35.76% in cache replacement and reduces bin usage by up to 30.93% in online bin packing. --- ### Small Language Models for Agentic Systems: A Survey of Architectures, Capabilities, and Deployment Trade offs Authors: Raghav Sharma, Manan Mehta Published: 2025-10-04 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2510.03847v1 Small language models (SLMs; 1-12B params, sometimes up to 20B) are sufficient and often superior for agentic workloads where the objective is schema- and API-constrained accuracy rather than open-ended generation. We synthesize recent evidence across open and proprietary SLMs (Phi-4-Mini, Qwen-2.5-7B, Gemma-2-9B, Llama-3.2-1B/3B, Ministral-3B/8B, Apple on-device 3B, DeepSeek-R1-Distill) and connect it to modern evaluations (BFCL v3/v4, StableToolBench) and serving stacks (vLLM, SGLang, TensorRT-LLM) paired with guided decoding libraries (XGrammar, Outlines). We formalize SLM-default, LLM-fallback systems with uncertainty-aware routing and verifier cascades, and propose engineering metrics that reflect real production goals: cost per successful task (CPS), schema validity rate, executable call rate, p50/p95 latency, and energy per request. Guided decoding, strict JSON Schema outputs, and validator-first tool execution close much of the capability gap with larger models and often let SLMs match or surpass LLMs on tool use, function calling, and RAG at 10x-100x lower token cost with materially better latency and energy. We provide design patterns for agent stacks that prioritize SLMs: schema-first prompting, type-safe function registries, confidence scoring with verifier rollups, and lightweight adaptation via LoRA/QLoRA. We also delineate limits where fallback remains valuable (open-domain reasoning and some long-horizon planning). The result is a practical blueprint for building fast, inexpensive, and reliable agents that default to SLMs while preserving headroom with targeted LLM assistance. Keywords: small language models, agents, function calling, structured outputs, JSON Schema, guided decoding, LoRA/QLoRA, routing, energy efficiency, edge inference --- ### The Hidden Game Problem Authors: Gon Buzaglo, Noah Golowich, Elad Hazan Published: 2025-10-04 Categories: cs.AI, cs.GT, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2510.03845v1 This paper investigates a class of games with large strategy spaces, motivated by challenges in AI alignment and language games. We introduce the hidden game problem, where for each player, an unknown subset of strategies consistently yields higher rewards compared to the rest. The central question is whether efficient regret minimization algorithms can be designed to discover and exploit such hidden structures, leading to equilibrium in these subgames while maintaining rationality in general. We answer this question affirmatively by developing a composition of regret minimization techniques that achieve optimal external and swap regret bounds. Our approach ensures rapid convergence to correlated equilibria in hidden subgames, leveraging the hidden game structure for improved computational efficiency. --- ### On Using Large Language Models to Enhance Clinically-Driven Missing Data Recovery Algorithms in Electronic Health Records Authors: Sarah C. Lotspeich, Abbey Collins, Brian J. Wells, Ashish K. Khanna, Joseph Rigdon, Lucy D'Agostino McGowan Published: 2025-10-04 Categories: cs.LG, stat.AP, stat.ME arXiv: https://arxiv.org/abs/2510.03844v1 Objective: Electronic health records (EHR) data are prone to missingness and errors. Previously, we devised an "enriched" chart review protocol where a "roadmap" of auxiliary diagnoses (anchors) was used to recover missing values in EHR data (e.g., a diagnosis of impaired glycemic control might imply that a missing hemoglobin A1c value would be considered unhealthy). Still, chart reviews are expensive and time-intensive, which limits the number of patients whose data can be reviewed. Now, we investigate the accuracy and scalability of a roadmap-driven algorithm, based on ICD-10 codes (International Classification of Diseases, 10th revision), to mimic expert chart reviews and recover missing values. Materials and Methods: In addition to the clinicians' original roadmap from our previous work, we consider new versions that were iteratively refined using large language models (LLM) in conjunction with clinical expertise to expand the list of auxiliary diagnoses. Using chart reviews for 100 patients from the EHR at an extensive learning health system, we examine algorithm performance with different roadmaps. Using the larger study of $1000$ patients, we applied the final algorithm, which used a roadmap with clinician-approved additions from the LLM. Results: The algorithm recovered as much, if not more, missing data as the expert chart reviewers, depending on the roadmap. Discussion: Clinically-driven algorithms (enhanced by LLM) can recover missing EHR data with similar accuracy to chart reviews and can feasibly be applied to large samples. Extending them to monitor other dimensions of data quality (e.g., plausability) is a promising future direction. --- ### Smart Paste: Automatically Fixing Copy/Paste for Google Developers Authors: Vincent Nguyen, Guilherme Herzog, José Cambronero, Marcus Revaj, Aditya Kini, Alexander Frömmgen, Maxim Tabachnyk Published: 2025-10-04 Categories: cs.SE, cs.HC, cs.LG arXiv: https://arxiv.org/abs/2510.03843v4 Manually editing pasted code is a long-standing developer pain point. In internal software development at Google, we observe that code is pasted 4 times more often than it is manually typed. These paste actions frequently require follow-up edits, ranging from simple reformatting and renaming to more complex style adjustments and cross-language translations. Prior work has shown deep learning can be used to predict these edits. In this work, we show how to iteratively develop and scale Smart Paste, an IDE feature for post-paste edit suggestions, to Google's development environment. This experience can serve as a guide for AI practitioners on a holistic approach to feature development, covering user experience, system integration, and model capabilities. Since deployment, Smart Paste has had overwhelmingly positive feedback with a 45% acceptance rate. At Google's enterprise scale, these accepted suggestions account substantially for over 1% of all code written company-wide. --- ### Mirage: Unveiling Hidden Artifacts in Synthetic Images with Large Vision-Language Models Authors: Pranav Sharma, Shivank Garg, Durga Toshniwal Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03840v1 Recent advances in image generation models have led to models that produce synthetic images that are increasingly difficult for standard AI detectors to identify, even though they often remain distinguishable by humans. To identify this discrepancy, we introduce \textbf{Mirage}, a curated dataset comprising a diverse range of AI-generated images exhibiting visible artifacts, where current state-of-the-art detection methods largely fail. Furthermore, we investigate whether Large Vision-Language Models (LVLMs), which are increasingly employed as substitutes for human judgment in various tasks, can be leveraged for explainable AI image detection. Our experiments on both Mirage and existing benchmark datasets demonstrate that while LVLMs are highly effective at detecting AI-generated images with visible artifacts, their performance declines when confronted with images lacking such cues. --- ### Technical note on Sequential Test-Time Adaptation via Martingale-Driven Fisher Prompting Authors: Behraj Khan, Tahir Qasim Syed Published: 2025-10-04 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2510.03839v2 We present a theoretical framework for M-FISHER, a method for sequential distribution shift detection and stable adaptation in streaming data. For detection, we construct an exponential martingale from non-conformity scores and apply Ville's inequality to obtain time-uniform guarantees on false alarm control, ensuring statistical validity at any stopping time. Under sustained shifts, we further bound the expected detection delay as $\mathcal{O}(\log(1/δ)/Γ)$, where $Γ$ reflects the post-shift information gain, thereby linking detection efficiency to distributional divergence. For adaptation, we show that Fisher-preconditioned updates of prompt parameters implement natural gradient descent on the distributional manifold, yielding locally optimal updates that minimize KL divergence while preserving stability and parameterization invariance. Together, these results establish M-FISHER as a principled approach for robust, anytime-valid detection and geometrically stable adaptation in sequential decision-making under covariate shift. --- ### Technical note on Fisher Information for Robust Federated Cross-Validation Authors: Behraj Khan, Tahir Qasim Syed Published: 2025-10-04 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2510.03838v1 When training data are fragmented across batches or federated-learned across different geographic locations, trained models manifest performance degradation. That degradation partly owes to covariate shift induced by data having been fragmented across time and space and producing dissimilar empirical training distributions. Each fragment's distribution is slightly different to a hypothetical unfragmented training distribution of covariates, and to the single validation distribution. To address this problem, we propose Fisher Information for Robust fEderated validation (\textbf{FIRE}). This method accumulates fragmentation-induced covariate shift divergences from the global training distribution via an approximate Fisher information. That term, which we prove to be a more computationally-tractable estimate, is then used as a per-fragment loss penalty, enabling scalable distribution alignment. FIRE outperforms importance weighting benchmarks by $5.1\%$ at maximum and federated learning (FL) benchmarks by up to $5.3\%$ on shifted validation sets. --- ### Joint Neural SDF Reconstruction and Semantic Segmentation for CAD Models Authors: Shen Fan, Przemyslaw Musialski Published: 2025-10-04 Categories: cs.GR, cs.CV arXiv: https://arxiv.org/abs/2510.03837v1 We propose a simple, data-efficient pipeline that augments an implicit reconstruction network based on neural SDF-based CAD parts with a part-segmentation head trained under PartField-generated supervision. Unlike methods tied to fixed taxonomies, our model accepts meshes with any number of parts and produces coherent, geometry-aligned labels in a single pass. We evaluate on randomly sampled CAD meshes from the ABC dataset with intentionally varied part cardinalities, including over-segmented shapes, and report strong performance across reconstruction (CDL1/CDL2, F1-micro, NC) and segmentation (mIoU, Accuracy), together with a new Segmentation Consistency metric that captures local label smoothness. We attach a lightweight segmentation head to the Flat-CAD SDF trunk; on a paired evaluation it does not alter reconstruction while providing accurate part labels for meshes with any number of parts. Even under degraded reconstructions on thin or intricate geometries, segmentation remains accurate and label-coherent, often preserving the correct part count. Our approach therefore offers a practical route to semantically structured CAD meshes without requiring curated taxonomies or exact palette matches. We discuss limitations in boundary precision, partly due to per-face supervision, and outline paths toward boundary-aware training and higher resolution labels. --- ### Towards Robust and Generalizable Continuous Space-Time Video Super-Resolution with Events Authors: Shuoyan Wei, Feng Li, Shengeng Tang, Runmin Cong, Yao Zhao, Meng Wang, Huihui Bai Published: 2025-10-04 Categories: eess.IV, cs.CV, cs.MM arXiv: https://arxiv.org/abs/2510.03833v1 Continuous space-time video super-resolution (C-STVSR) has garnered increasing interest for its capability to reconstruct high-resolution and high-frame-rate videos at arbitrary spatial and temporal scales. However, prevailing methods often generalize poorly, producing unsatisfactory results when applied to out-of-distribution (OOD) scales. To overcome this limitation, we present EvEnhancer, a novel approach that marries the unique properties of high temporal resolution and high dynamic range encapsulated in event streams to achieve robust and generalizable C-STVSR. Our approach incorporates event-adapted synthesis that capitalizes on the spatiotemporal correlations between frames and events to capture long-term motion trajectories, enabling adaptive interpolation and fusion across space and time. This is then coupled with a local implicit video transformer that integrates local implicit video neural function with cross-scale spatiotemporal attention to learn continuous video representations and generate plausible videos at arbitrary resolutions and frame rates. We further develop EvEnhancerPlus, which builds a controllable switching mechanism that dynamically determines the reconstruction difficulty for each spatiotemporal pixel based on local event statistics. This allows the model to adaptively route reconstruction along the most suitable pathways at a fine-grained pixel level, substantially reducing computational overhead while maintaining excellent performance. Furthermore, we devise a cross-derivative training strategy that stabilizes the convergence of such a multi-pathway framework through staged cross-optimization. Extensive experiments demonstrate that our method achieves state-of-the-art performance on both synthetic and real-world datasets, while maintaining superior generalizability at OOD scales. The code is available at https://github.com/W-Shuoyan/EvEnhancerPlus. --- ### Detecting Malicious Pilot Contamination in Multiuser Massive MIMO Using Decision Trees Authors: Pedro Ivo da Cruz, Dimitri Silva, Tito Spadini, Ricardo Suyama, Murilo Bellezoni Loiola Published: 2025-10-04 Categories: cs.CR, cs.IT, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2510.03831v2 Massive multiple-input multiple-output (MMIMO) is essential to modern wireless communication systems, like 5G and 6G, but it is vulnerable to active eavesdropping attacks. One type of such attack is the pilot contamination attack (PCA), where a malicious user copies pilot signals from an authentic user during uplink, intentionally interfering with the base station's (BS) channel estimation accuracy. In this work, we propose to use a Decision Tree (DT) algorithm for PCA detection at the BS in a multi-user system. We present a methodology to generate training data for the DT classifier and select the best DT according to their depth. Then, we simulate different scenarios that could be encountered in practice and compare the DT to a classical technique based on likelihood ratio testing (LRT) submitted to the same scenarios. The results revealed that a DT with only one level of depth is sufficient to outperform the LRT. The DT shows a good performance regarding the probability of detection in noisy scenarios and when the malicious user transmits with low power, in which case the LRT fails to detect the PCA. We also show that the reason for the good performance of the DT is its ability to compute a threshold that separates PCA data from non-PCA data better than the LRT's threshold. Moreover, the DT does not necessitate prior knowledge of noise power or assumptions regarding the signal power of malicious users, prerequisites typically essential for LRT and other hypothesis testing methodologies. --- ### HOFLON: Hybrid Offline Learning and Online Optimization for Process Start-Up and Grade-Transition Control Authors: Alex Durkin, Jasper Stolte, Mehmet Mercangöz Published: 2025-10-04 Categories: cs.LG, eess.SY, stat.ML arXiv: https://arxiv.org/abs/2510.03830v1 Start-ups and product grade-changes are critical steps in continuous-process plant operation, because any misstep immediately affects product quality and drives operational losses. These transitions have long relied on manual operation by a handful of expert operators, but the progressive retirement of that workforce is leaving plant owners without the tacit know-how needed to execute them consistently. In the absence of a process model, offline reinforcement learning (RL) promises to capture and even surpass human expertise by mining historical start-up and grade-change logs, yet standard offline RL struggles with distribution shift and value-overestimation whenever a learned policy ventures outside the data envelope. We introduce HOFLON (Hybrid Offline Learning + Online Optimization) to overcome those limitations. Offline, HOFLON learns (i) a latent data manifold that represents the feasible region spanned by past transitions and (ii) a long-horizon Q-critic that predicts the cumulative reward from state-action pairs. Online, it solves a one-step optimization problem that maximizes the Q-critic while penalizing deviations from the learned manifold and excessive rates of change in the manipulated variables. We test HOFLON on two industrial case studies: a polymerization reactor start-up and a paper-machine grade-change problem, and benchmark it against Implicit Q-Learning (IQL), a leading offline-RL algorithm. In both plants HOFLON not only surpasses IQL but also delivers, on average, better cumulative rewards than the best start-up or grade-change observed in the historical data, demonstrating its potential to automate transition operations beyond current expert capability. --- ### A4FN: an Agentic AI Architecture for Autonomous Flying Networks Authors: André Coelho, Pedro Ribeiro, Helder Fontes, Rui Campos Published: 2025-10-04 Categories: cs.NI, cs.AI arXiv: https://arxiv.org/abs/2510.03829v1 This position paper presents A4FN, an Agentic Artificial Intelligence (AI) architecture for intent-driven automation in Flying Networks (FNs) using Unmanned Aerial Vehicles (UAVs) as access nodes. A4FN leverages Generative AI and Large Language Models (LLMs) to enable real-time, context-aware network control via a distributed agentic system. It comprises two components: the Perception Agent (PA), which semantically interprets multimodal input -- including imagery, audio, and telemetry data -- from UAV-mounted sensors to derive Service Level Specifications (SLSs); and the Decision-and-Action Agent (DAA), which reconfigures the network based on inferred intents. A4FN embodies key properties of Agentic AI, including autonomy, goal-driven reasoning, and continuous perception-action cycles. Designed for mission-critical, infrastructure-limited scenarios such as disaster response, it supports adaptive reconfiguration, dynamic resource management, and interoperability with emerging wireless technologies. The paper details the A4FN architecture, its core innovations, and open research challenges in multi-agent coordination and Agentic AI integration in next-generation FNs. --- ### LIBERO-PRO: Towards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorization Authors: Xueyang Zhou, Yangming Xu, Guiyao Tie, Yongchao Chen, Guowen Zhang, Duanfeng Chu, Pan Zhou, Lichao Sun Published: 2025-10-04 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2510.03827v2 LIBERO has emerged as a widely adopted benchmark for evaluating Vision-Language-Action (VLA) models; however, its current training and evaluation settings are problematic, often leading to inflated performance estimates and preventing fair model comparison. To address these issues, we introduce LIBERO-PRO, an extended LIBERO benchmark that systematically evaluates model performance under reasonable perturbations across four dimensions: manipulated objects, initial states, task instructions, and environments. Experimental results reveal that, although existing models achieve over 90% accuracy under the standard LIBERO evaluation, their performance collapses to 0.0% under our generalized setting. Crucially, this discrepancy exposes the models' reliance on rote memorization of action sequences and environment layouts from the training set, rather than genuine task understanding or environmental perception. For instance, models persist in executing grasping actions when the target object is replaced with irrelevant items, and their outputs remain unchanged even when given corrupted instructions or even messy tokens. These findings expose the severe flaws in current evaluation practices, and we call on the community to abandon misleading methodologies in favor of robust assessments of model generalization and comprehension. Our code is available at: https://github.com/Zxy-MLlab/LIBERO-PRO. --- ### Proximal Diffusion Neural Sampler Authors: Wei Guo, Jaemoo Choi, Yuchen Zhu, Molei Tao, Yongxin Chen Published: 2025-10-04 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2510.03824v2 The task of learning a diffusion-based neural sampler for drawing samples from an unnormalized target distribution can be viewed as a stochastic optimal control problem on path measures. However, the training of neural samplers can be challenging when the target distribution is multimodal with significant barriers separating the modes, potentially leading to mode collapse. We propose a framework named Proximal Diffusion Neural Sampler (PDNS) that addresses these challenges by tackling the stochastic optimal control problem via proximal point method on the space of path measures. PDNS decomposes the learning process into a series of simpler subproblems that create a path gradually approaching the desired distribution. This staged procedure traces a progressively refined path to the desired distribution and promotes thorough exploration across modes. For a practical and efficient realization, we instantiate each proximal step with a proximal weighted denoising cross-entropy (WDCE) objective. We demonstrate the effectiveness and robustness of PDNS through extensive experiments on both continuous and discrete sampling tasks, including challenging scenarios in molecular dynamics and statistical physics. Our code is available at https://github.com/AlexandreGUO2001/PDNS. --- ### Distributed Area Coverage with High Altitude Balloons Using Multi-Agent Reinforcement Learning Authors: Adam Haroon, Tristan Schuler Published: 2025-10-04 Categories: cs.LG, cs.MA, cs.RO arXiv: https://arxiv.org/abs/2510.03823v1 High Altitude Balloons (HABs) can leverage stratospheric wind layers for limited horizontal control, enabling applications in reconnaissance, environmental monitoring, and communications networks. Existing multi-agent HAB coordination approaches use deterministic methods like Voronoi partitioning and extremum seeking control for large global constellations, which perform poorly for smaller teams and localized missions. While single-agent HAB control using reinforcement learning has been demonstrated on HABs, coordinated multi-agent reinforcement learning (MARL) has not yet been investigated. This work presents the first systematic application of multi-agent reinforcement learning (MARL) to HAB coordination for distributed area coverage. We extend our previously developed reinforcement learning simulation environment (RLHAB) to support cooperative multi-agent learning, enabling multiple agents to operate simultaneously in realistic atmospheric conditions. We adapt QMIX for HAB area coverage coordination, leveraging Centralized Training with Decentralized Execution to address atmospheric vehicle coordination challenges. Our approach employs specialized observation spaces providing individual state, environmental context, and teammate data, with hierarchical rewards prioritizing coverage while encouraging spatial distribution. We demonstrate that QMIX achieves similar performance to the theoretically optimal geometric deterministic method for distributed area coverage, validating the MARL approach and providing a foundation for more complex autonomous multi-HAB missions where deterministic methods become intractable. --- ### Contrastive-SDE: Guiding Stochastic Differential Equations with Contrastive Learning for Unpaired Image-to-Image Translation Authors: Venkata Narendra Kotyada, Revanth Eranki, Nagesh Bhattu Sristy Published: 2025-10-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.03821v1 Unpaired image-to-image translation involves learning mappings between source domain and target domain in the absence of aligned or corresponding samples. Score based diffusion models have demonstrated state-of-the-art performance in generative tasks. Their ability to approximate complex data distributions through stochastic differential equations (SDEs) enables them to generate high-fidelity and diverse outputs, making them particularly well-suited for unpaired I2I settings. In parallel, contrastive learning provides a powerful framework for learning semantic similarities without the need for explicit supervision or paired data. By pulling together representations of semantically similar samples and pushing apart dissimilar ones, contrastive methods are inherently aligned with the objectives of unpaired translation. Its ability to selectively enforce semantic consistency at the feature level makes contrastive learning particularly effective for guiding generation in unpaired scenarios. In this work, we propose a time-dependent contrastive learning approach where a model is trained with SimCLR by considering an image and its domain invarient feature as a positive pair, enabling the preservation of domain-invariant features and the discarding of domain-specific ones. The learned contrastive model then guides the inference of a pretrained SDE for the I2I translation task. We empirically compare Contrastive-SDE with several baselines across three common unpaired I2I tasks, using four metrics for evaluation. Constrastive-SDE achieves comparable results to the state-of-the-art on several metrics. Furthermore, we observe that our model converges significantly faster and requires no label supervision or classifier training, making it a more efficient alternative for this task. --- ### TROLL: Trust Regions improve Reinforcement Learning for Large Language Models Authors: Philipp Becker, Niklas Freymuth, Serge Thilges, Fabian Otto, Gerhard Neumann Published: 2025-10-04 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2510.03817v3 Reinforcement Learning (RL) with PPO-like clip objectives has become the standard choice for reward-based fine-tuning of large language models (LLMs). Although recent work has explored improved estimators of advantages and normalization, the clipping mechanism itself has remained untouched. Originally introduced as a proxy for principled KL-based trust regions, clipping is a crude approximation that often causes unstable updates and suboptimal performance. We replace the clip objective with a novel discrete differentiable trust region projection, which provides principled token-level KL constraints. The projection operates on a sparse subset of the model's most important token logits to balance computational cost and projection effectiveness. Our approach, Trust Region Optimization for Large Language models (TROLL), serves as a direct replacement for PPO-like clipping during training and does not alter the model's inference behavior. Across mathematical reasoning and code generation tasks, model families, as well as advantage-estimation methods, TROLL consistently outperforms PPO-like clipping in terms of training speed, stability, and final success rates. --- ### A Trustworthy Industrial Fault Diagnosis Architecture Integrating Probabilistic Models and Large Language Models Authors: Yue wu Published: 2025-10-04 Categories: eess.SY, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2510.03815v1 There are limitations of traditional methods and deep learning methods in terms of interpretability, generalization, and quantification of uncertainty in industrial fault diagnosis, and there are core problems of insufficient credibility in industrial fault diagnosis. The architecture performs preliminary analysis through a Bayesian network-based diagnostic engine and features an LLM-driven cognitive quorum module with multimodal input capabilities. The module conducts expert-level arbitration of initial diagnoses by analyzing structured features and diagnostic charts, prioritizing final decisions after conflicts are identified. To ensure the reliability of the system output, the architecture integrates a confidence calibration module based on temperature calibration and a risk assessment module, which objectively quantifies the reliability of the system using metrics such as expected calibration error (ECE). Experimental results on a dataset containing multiple fault types showed that the proposed framework improved diagnostic accuracy by more than 28 percentage points compared to the baseline model, while the calibrated ECE was reduced by more than 75%. Case studies have confirmed that HCAA effectively corrects misjudgments caused by complex feature patterns or knowledge gaps in traditional models, providing novel and practical engineering solutions for building high-trust, explainable AI diagnostic systems for industrial applications. --- ### Detecting Invariant Manifolds in ReLU-Based RNNs Authors: Lukas Eisenmann, Alena Brändle, Zahra Monfared, Daniel Durstewitz Published: 2025-10-04 Categories: cs.LG, cs.AI, math.DS arXiv: https://arxiv.org/abs/2510.03814v4 Recurrent Neural Networks (RNNs) have found widespread applications in machine learning for time series prediction and dynamical systems reconstruction, and experienced a recent renaissance with improved training algorithms and architectural designs. Understanding why and how trained RNNs produce their behavior is important for scientific and medical applications, and explainable AI more generally. An RNN's dynamical repertoire depends on the topological and geometrical properties of its state space. Stable and unstable manifolds of periodic points play a particularly important role: They dissect a dynamical system's state space into different basins of attraction, and their intersections lead to chaotic dynamics with fractal geometry. Here we introduce a novel algorithm for detecting these manifolds, with a focus on piecewise-linear RNNs (PLRNNs) employing rectified linear units (ReLUs) as their activation function. We demonstrate how the algorithm can be used to trace the boundaries between different basins of attraction, and hence to characterize multistability, a computationally important property. We further show its utility in finding so-called homoclinic points, the intersections between stable and unstable manifolds, and thus establish the existence of chaos in PLRNNs. Finally we show for an empirical example, electrophysiological recordings from a cortical neuron, how insights into the underlying dynamics could be gained through our method. ---