# Research Papers — 2025-10 · bundle page 40 of 217 > Full-text markdown of papers #1951–2000 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/40 [← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/39.md) [Next bundle →](https://aipod.dev/papers/archive/2025-10/page/41.md) ### TABL-ABM: A Hybrid Framework for Synthetic LOB Generation Authors: Ollie Olby, Rory Baggott, Namid Stillman Published: 2025-10-26 Categories: q-fin.CP, cs.AI, cs.MA, q-fin.TR arXiv: https://arxiv.org/abs/2510.22685v1 The recent application of deep learning models to financial trading has heightened the need for high fidelity financial time series data. This synthetic data can be used to supplement historical data to train large trading models. The state-of-the-art models for the generative application often rely on huge amounts of historical data and large, complicated models. These models range from autoregressive and diffusion-based models through to architecturally simpler models such as the temporal-attention bilinear layer. Agent-based approaches to modelling limit order book dynamics can also recreate trading activity through mechanistic models of trader behaviours. In this work, we demonstrate how a popular agent-based framework for simulating intraday trading activity, the Chiarella model, can be combined with one of the most performant deep learning models for forecasting multi-variate time series, the TABL model. This forecasting model is coupled to a simulation of a matching engine with a novel method for simulating deleted order flow. Our simulator gives us the ability to test the generative abilities of the forecasting model using stylised facts. Our results show that this methodology generates realistic price dynamics however, when analysing deeper, parts of the markets microstructure are not accurately recreated, highlighting the necessity for including more sophisticated agent behaviors into the modeling framework to help account for tail events. --- ### RoboSVG: A Unified Framework for Interactive SVG Generation with Multi-modal Guidance Authors: Jiuniu Wang, Gongjie Zhang, Quanhao Qian, Junlong Gao, Deli Zhao, Ran Xu Published: 2025-10-26 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2510.22684v1 Scalable Vector Graphics (SVGs) are fundamental to digital design and robot control, encoding not only visual structure but also motion paths in interactive drawings. In this work, we introduce RoboSVG, a unified multimodal framework for generating interactive SVGs guided by textual, visual, and numerical signals. Given an input query, the RoboSVG model first produces multimodal guidance, then synthesizes candidate SVGs through dedicated generation modules, and finally refines them under numerical guidance to yield high-quality outputs. To support this framework, we construct RoboDraw, a large-scale dataset of one million examples, each pairing an SVG generation condition (e.g., text, image, and partial SVG) with its corresponding ground-truth SVG code. RoboDraw dataset enables systematic study of four tasks, including basic generation (Text-to-SVG, Image-to-SVG) and interactive generation (PartialSVG-to-SVG, PartialImage-to-SVG). Extensive experiments demonstrate that RoboSVG achieves superior query compliance and visual fidelity across tasks, establishing a new state of the art in versatile SVG generation. The dataset and source code of this project will be publicly available soon. --- ### Estimation of Fireproof Structure Class and Construction Year for Disaster Risk Assessment Authors: Hibiki Ayabe, Kazushi Okamoto, Koki Karube, Atsushi Shibata, Kei Harada Published: 2025-10-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.22683v1 Structural fireproof classification is vital for disaster risk assessment and insurance pricing in Japan. However, key building metadata such as construction year and structure type are often missing or outdated, particularly in the second-hand housing market. This study proposes a multi-task learning model that predicts these attributes from facade images. The model jointly estimates the construction year, building structure, and property type, from which the structural fireproof class - defined as H (non-fireproof), T (semi-fireproof), or M (fireproof) - is derived via a rule-based mapping based on official insurance criteria. We trained and evaluated the model using a large-scale dataset of Japanese residential images, applying rigorous filtering and deduplication. The model achieved high accuracy in construction-year regression and robust classification across imbalanced categories. Qualitative analyses show that it captures visual cues related to building age and materials. Our approach demonstrates the feasibility of scalable, interpretable, image-based risk-profiling systems, offering potential applications in insurance, urban planning, and disaster preparedness. --- ### Uncertainty-Aware Autonomous Vehicles: Predicting the Road Ahead Authors: Shireen Kudukkil Manchingal, Armand Amaritei, Mihir Gohad, Maryam Sultana, Julian F. P. Kooij, Fabio Cuzzolin, Andrew Bradley Published: 2025-10-26 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2510.22680v1 Autonomous Vehicle (AV) perception systems have advanced rapidly in recent years, providing vehicles with the ability to accurately interpret their environment. Perception systems remain susceptible to errors caused by overly-confident predictions in the case of rare events or out-of-sample data. This study equips an autonomous vehicle with the ability to 'know when it is uncertain', using an uncertainty-aware image classifier as part of the AV software stack. Specifically, the study exploits the ability of Random-Set Neural Networks (RS-NNs) to explicitly quantify prediction uncertainty. Unlike traditional CNNs or Bayesian methods, RS-NNs predict belief functions over sets of classes, allowing the system to identify and signal uncertainty clearly in novel or ambiguous scenarios. The system is tested in a real-world autonomous racing vehicle software stack, with the RS-NN classifying the layout of the road ahead and providing the associated uncertainty of the prediction. Performance of the RS-NN under a range of road conditions is compared against traditional CNN and Bayesian neural networks, with the RS-NN achieving significantly higher accuracy and superior uncertainty calibration. This integration of RS-NNs into Robot Operating System (ROS)-based vehicle control pipeline demonstrates that predictive uncertainty can dynamically modulate vehicle speed, maintaining high-speed performance under confident predictions while proactively improving safety through speed reductions in uncertain scenarios. These results demonstrate the potential of uncertainty-aware neural networks - in particular RS-NNs - as a practical solution for safer and more robust autonomous driving. --- ### Do Stop Me Now: Detecting Boilerplate Responses with a Single Iteration Authors: Yuval Kainan, Shaked Zychlinski Published: 2025-10-26 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2510.22679v1 Large Language Models (LLMs) often expend significant computational resources generating boilerplate responses, such as refusals, simple acknowledgements and casual greetings, which adds unnecessary cost and latency. To address this inefficiency, we propose a simple yet highly effective method for detecting such responses after only a single generation step. We demonstrate that the log-probability distribution of the first generated token serves as a powerful signal for classifying the nature of the entire subsequent response. Our experiments, conducted across a diverse range of small, large, and reasoning-specialized models, show that the first-token log-probability vectors form distinctly separable clusters for different response types. Using a lightweight k-NN classifier, we achieve high accuracy in predicting whether a response will be a substantive answer or a form of boilerplate response, including user-specified refusals. The primary implication is a practical, computationally trivial technique, optimizing LLM inference by enabling early termination or redirection to a smaller model, thereby yielding significant savings in computational cost. This work presents a direct path toward more efficient and sustainable LLM deployment. --- ### DAMap: Distance-aware MapNet for High Quality HD Map Construction Authors: Jinpeng Dong, Chen Li, Yutong Lin, Jingwen Fu, Sanping Zhou, Nanning Zheng Published: 2025-10-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.22675v1 Predicting High-definition (HD) map elements with high quality (high classification and localization scores) is crucial to the safety of autonomous driving vehicles. However, current methods perform poorly in high quality predictions due to inherent task misalignment. Two main factors are responsible for misalignment: 1) inappropriate task labels due to one-to-many matching queries sharing the same labels, and 2) sub-optimal task features due to task-shared sampling mechanism. In this paper, we reveal two inherent defects in current methods and develop a novel HD map construction method named DAMap to address these problems. Specifically, DAMap consists of three components: Distance-aware Focal Loss (DAFL), Hybrid Loss Scheme (HLS), and Task Modulated Deformable Attention (TMDA). The DAFL is introduced to assign appropriate classification labels for one-to-many matching samples. The TMDA is proposed to obtain discriminative task-specific features. Furthermore, the HLS is proposed to better utilize the advantages of the DAFL. We perform extensive experiments and consistently achieve performance improvement on the NuScenes and Argoverse2 benchmarks under different metrics, baselines, splits, backbones, and schedules. Code will be available at https://github.com/jpdong-xjtu/DAMap. --- ### Alias-Free ViT: Fractional Shift Invariance via Linear Attention Authors: Hagay Michaeli, Daniel Soudry Published: 2025-10-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.22673v1 Transformers have emerged as a competitive alternative to convnets in vision tasks, yet they lack the architectural inductive bias of convnets, which may hinder their potential performance. Specifically, Vision Transformers (ViTs) are not translation-invariant and are more sensitive to minor image translations than standard convnets. Previous studies have shown, however, that convnets are also not perfectly shift-invariant, due to aliasing in downsampling and nonlinear layers. Consequently, anti-aliasing approaches have been proposed to certify convnets' translation robustness. Building on this line of work, we propose an Alias-Free ViT, which combines two main components. First, it uses alias-free downsampling and nonlinearities. Second, it uses linear cross-covariance attention that is shift-equivariant to both integer and fractional translations, enabling a shift-invariant global representation. Our model maintains competitive performance in image classification and outperforms similar-sized models in terms of robustness to adversarial translations. --- ### Look and Tell: A Dataset for Multimodal Grounding Across Egocentric and Exocentric Views Authors: Anna Deichler, Jonas Beskow Published: 2025-10-26 Categories: cs.CV, cs.CL, cs.RO arXiv: https://arxiv.org/abs/2510.22672v2 We introduce Look and Tell, a multimodal dataset for studying referential communication across egocentric and exocentric perspectives. Using Meta Project Aria smart glasses and stationary cameras, we recorded synchronized gaze, speech, and video as 25 participants instructed a partner to identify ingredients in a kitchen. Combined with 3D scene reconstructions, this setup provides a benchmark for evaluating how different spatial representations (2D vs. 3D; ego vs. exo) affect multimodal grounding. The dataset contains 3.67 hours of recordings, including 2,707 richly annotated referential expressions, and is designed to advance the development of embodied agents that can understand and engage in situated dialogue. --- ### LVD-GS: Gaussian Splatting SLAM for Dynamic Scenes via Hierarchical Explicit-Implicit Representation Collaboration Rendering Authors: Wenkai Zhu, Xu Li, Qimin Xu, Benwu Wang, Kun Wei, Yiming Peng, Zihang Wang Published: 2025-10-26 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.22669v1 3D Gaussian Splatting SLAM has emerged as a widely used technique for high-fidelity mapping in spatial intelligence. However, existing methods often rely on a single representation scheme, which limits their performance in large-scale dynamic outdoor scenes and leads to cumulative pose errors and scale ambiguity. To address these challenges, we propose \textbf{LVD-GS}, a novel LiDAR-Visual 3D Gaussian Splatting SLAM system. Motivated by the human chain-of-thought process for information seeking, we introduce a hierarchical collaborative representation module that facilitates mutual reinforcement for mapping optimization, effectively mitigating scale drift and enhancing reconstruction robustness. Furthermore, to effectively eliminate the influence of dynamic objects, we propose a joint dynamic modeling module that generates fine-grained dynamic masks by fusing open-world segmentation with implicit residual constraints, guided by uncertainty estimates from DINO-Depth features. Extensive evaluations on KITTI, nuScenes, and self-collected datasets demonstrate that our approach achieves state-of-the-art performance compared to existing methods. --- ### Block Coordinate Descent for Neural Networks Provably Finds Global Minima Authors: Shunta Akiyama Published: 2025-10-26 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2510.22667v1 In this paper, we consider a block coordinate descent (BCD) algorithm for training deep neural networks and provide a new global convergence guarantee under strictly monotonically increasing activation functions. While existing works demonstrate convergence to stationary points for BCD in neural networks, our contribution is the first to prove convergence to global minima, ensuring arbitrarily small loss. We show that the loss with respect to the output layer decreases exponentially while the loss with respect to the hidden layers remains well-controlled. Additionally, we derive generalization bounds using the Rademacher complexity framework, demonstrating that BCD not only achieves strong optimization guarantees but also provides favorable generalization performance. Moreover, we propose a modified BCD algorithm with skip connections and non-negative projection, extending our convergence guarantees to ReLU activation, which are not strictly monotonic. Empirical experiments confirm our theoretical findings, showing that the BCD algorithm achieves a small loss for strictly monotonic and ReLU activations. --- ### SARVLM: A Vision Language Foundation Model for Semantic Understanding in SAR Imagery Authors: Qiwei Ma, Xukun Lu, Wang Liu, Puhong Duan, Xudong Kang, Shutao Li Published: 2025-10-26 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.22665v4 Synthetic Aperture Radar (SAR) is a critical imaging modality due to its all-weather operational capability. Although recent advances in self-supervised learning and masked image modeling (MIM) have enabled SAR foundation models, these approaches primarily focus on low-level visual features and often neglect multi-modal representation. Moreover, multimodal data for SAR is scarce, limiting the development of robust cross-modal models. To address this limitation, we construct SARVLM-1M, a large-scale vision-language dataset comprising over one million image-text pairs aggregated from existing datasets. Furthermore, to mitigate the substantial differences between SAR and natural imagery, we propose a two-stage domain transfer training strategy that leverages optical remote sensing data as an intermediate bridge, facilitating effective knowledge transfer from natural images to SAR domains. Based on this strategy, we develop SARVLM, the first vision-language foundation model tailored for SAR, consisting of SARCLIP and SARCap. In addition, an ensemble strategy is utilized to improve the cross-scene generalization capability of the model. Moreover, SARDet and SARRot further validate the capability of the proposed framework in object detection. Extensive experiments on 13 benchmarks across image-text retrieval, target recognition, zero-shot classification, object detection, semantic localization, and image captioning demonstrate the superior feature extraction and interpretation capabilities of SARVLM. It consistently outperforms state-of-the-art vision-language models and advances semantic understanding in SAR imagery. Code and datasets will be released on https://github.com/KlayMa527/SARVLM.git. --- ### Conjugate Relation Modeling for Few-Shot Knowledge Graph Completion Authors: Zilong Wang, Qingtian Zeng, Hua Duan, Cheng Cheng, Minghao Zou, Ziyang Wang Published: 2025-10-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2510.22656v2 Few-shot Knowledge Graph Completion (FKGC) infers missing triples from limited support samples, tackling long-tail distribution challenges. Existing methods, however, struggle to capture complex relational patterns and mitigate data sparsity. To address these challenges, we propose a novel FKGC framework for conjugate relation modeling (CR-FKGC). Specifically, it employs a neighborhood aggregation encoder to integrate higher-order neighbor information, a conjugate relation learner combining an implicit conditional diffusion relation module with a stable relation module to capture stable semantics and uncertainty offsets, and a manifold conjugate decoder for efficient evaluation and inference of missing triples in manifold space. Experiments on three benchmarks demonstrate that our method achieves superior performance over state-of-the-art methods. --- ### Learning Without Augmenting: Unsupervised Time Series Representation Learning via Frame Projections Authors: Berken Utku Demirel, Christian Holz Published: 2025-10-26 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2510.22655v2 Self-supervised learning (SSL) has emerged as a powerful paradigm for learning representations without labeled data. Most SSL approaches rely on strong, well-established, handcrafted data augmentations to generate diverse views for representation learning. However, designing such augmentations requires domain-specific knowledge and implicitly imposes representational invariances on the model, which can limit generalization. In this work, we propose an unsupervised representation learning method that replaces augmentations by generating views using orthonormal bases and overcomplete frames. We show that embeddings learned from orthonormal and overcomplete spaces reside on distinct manifolds, shaped by the geometric biases introduced by representing samples in different spaces. By jointly leveraging the complementary geometry of these distinct manifolds, our approach achieves superior performance without artificially increasing data diversity through strong augmentations. We demonstrate the effectiveness of our method on nine datasets across five temporal sequence tasks, where signal-specific characteristics make data augmentations particularly challenging. Without relying on augmentation-induced diversity, our method achieves performance gains of up to 15--20\% over existing self-supervised approaches. Source code: https://github.com/eth-siplab/Learning-with-FrameProjections --- ### Managing Self-Learning Experts under Per-Round Budget Constraints Authors: Ilgam Latypov, Alexandra Suvorikova, Alexey Kroshnin, Alexander Gasnikov, Yuriy Dorn Published: 2025-10-26 Categories: cs.LG, cs.MA arXiv: https://arxiv.org/abs/2510.22654v3 This paper addresses the problem of sequential decision-making under learning budget constraints. Such settings naturally arise in applications like managing a portfolio of bandit or reinforcement learning (RL) algorithms. We propose a novel UCB-type algorithm, M-LCB, designed to manage a pool of $K$ self-learning experts in a stochastic environment while accounting for a limited per-round learning budget $M$. At each round, M-LCB selects one expert to make a decision and at most $M \le K$ experts to learn. For selection, M-LCB uses confidence bounds constructed from limited prior knowledge about the experts (i.e., mild assumptions) and their observed training losses. We derive anytime regret bounds for M-LCB that scale with the individual regrets of the experts. In particular, if each expert has regret $\tilde O(T^α)$ by round $T$, then M-LCB guarantees an overall regret of $\tilde O\left(\sqrt{KT/M} + (K/M)^{1-α}T^α\right)$ relative to the best expert in hindsight. Finally, we demonstrate the applicability of M-LCB using self-learning experts instantiated as (i) parametric models and (ii) bandit algorithms. --- ### If You Want to Be Robust, Be Wary of Initialization Authors: Sofiane Ennadir, Johannes F. Lutzeyer, Michalis Vazirgiannis, El Houcine Bergou Published: 2025-10-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.22652v1 Graph Neural Networks (GNNs) have demonstrated remarkable performance across a spectrum of graph-related tasks, however concerns persist regarding their vulnerability to adversarial perturbations. While prevailing defense strategies focus primarily on pre-processing techniques and adaptive message-passing schemes, this study delves into an under-explored dimension: the impact of weight initialization and associated hyper-parameters, such as training epochs, on a model's robustness. We introduce a theoretical framework bridging the connection between initialization strategies and a network's resilience to adversarial perturbations. Our analysis reveals a direct relationship between initial weights, number of training epochs and the model's vulnerability, offering new insights into adversarial robustness beyond conventional defense mechanisms. While our primary focus is on GNNs, we extend our theoretical framework, providing a general upper-bound applicable to Deep Neural Networks. Extensive experiments, spanning diverse models and real-world datasets subjected to various adversarial attacks, validate our findings. We illustrate that selecting appropriate initialization not only ensures performance on clean datasets but also enhances model robustness against adversarial perturbations, with observed gaps of up to 50\% compared to alternative initialization approaches. --- ### Variational Polya Tree Authors: Lu Xu, Tsai Hor Chan, Kwok Fai Lam, Lequan Yu, Guosheng Yin Published: 2025-10-26 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2510.22651v1 Density estimation is essential for generative modeling, particularly with the rise of modern neural networks. While existing methods capture complex data distributions, they often lack interpretability and uncertainty quantification. Bayesian nonparametric methods, especially the \polya tree, offer a robust framework that addresses these issues by accurately capturing function behavior over small intervals. Traditional techniques like Markov chain Monte Carlo (MCMC) face high computational complexity and scalability limitations, hindering the use of Bayesian nonparametric methods in deep learning. To tackle this, we introduce the variational \polya tree (VPT) model, which employs stochastic variational inference to compute posterior distributions. This model provides a flexible, nonparametric Bayesian prior that captures latent densities and works well with stochastic gradient optimization. We also leverage the joint distribution likelihood for a more precise variational posterior approximation than traditional mean-field methods. We evaluate the model performance on both real data and images, and demonstrate its competitiveness with other state-of-the-art deep density estimation methods. We also explore its ability in enhancing interpretability and uncertainty quantification. Code is available at https://github.com/howardchanth/var-polya-tree. --- ### Self-Attention Decomposition For Training Free Diffusion Editing Authors: Tharun Anand, Mohammad Hassan Vali, Arno Solin, Green Rosh, BH Pawan Prasad Published: 2025-10-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.22650v2 Diffusion models achieve remarkable fidelity in image synthesis, yet precise control over their outputs for targeted editing remains challenging. A key step toward controllability is to identify interpretable directions in the model's latent representations that correspond to semantic attributes. Existing approaches for finding interpretable directions typically rely on sampling large sets of images or training auxiliary networks, which limits efficiency. We propose an analytical method that derives semantic editing directions directly from the pretrained parameters of diffusion models, requiring neither additional data nor fine-tuning. Our insight is that self-attention weight matrices encode rich structural information about the data distribution learned during training. By computing the eigenvectors of these weight matrices, we obtain robust and interpretable editing directions. Experiments demonstrate that our method produces high-quality edits across multiple datasets while reducing editing time significantly by 60% over current benchmarks. --- ### A Critical Study on Tea Leaf Disease Detection using Deep Learning Techniques Authors: Nabajyoti Borah, Raju Moni Borah, Bandan Boruah, Purnendu Bikash Acharjee, Sajal Saha, Ripjyoti Hazarika Published: 2025-10-26 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.22647v1 The proposed solution is Deep Learning Technique that will be able classify three types of tea leaves diseases from which two diseases are caused by the pests and one due to pathogens (infectious organisms) and environmental conditions and also show the area damaged by a disease in leaves. Namely Red Rust, Helopeltis and Red spider mite respectively. In this paper we have evaluated two models namely SSD MobileNet V2 and Faster R-CNN ResNet50 V1 for the object detection. The SSD MobileNet V2 gave precision of 0.209 for IOU range of 0.50:0.95 with recall of 0.02 on IOU 0.50:0.95 and final mAP of 20.9%. While Faster R-CNN ResNet50 V1 has precision of 0.252 on IOU range of 0.50:0.95 and recall of 0.044 on IOU of 0.50:0.95 with a mAP of 25%, which is better than SSD. Also used Mask R-CNN for Object Instance Segmentation where we have implemented our custom method to calculate the damaged diseased portion of leaves. Keywords: Tea Leaf Disease, Deep Learning, Red Rust, Helopeltis and Red Spider Mite, SSD MobileNet V2, Faster R-CNN ResNet50 V1 and Mask RCNN. --- ### Enhancing Graph Classification Robustness with Singular Pooling Authors: Sofiane Ennadir, Oleg Smirnov, Yassine Abbahaddou, Lele Cao, Johannes F. Lutzeyer Published: 2025-10-26 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2510.22643v1 Graph Neural Networks (GNNs) have achieved strong performance across a range of graph representation learning tasks, yet their adversarial robustness in graph classification remains underexplored compared to node classification. While most existing defenses focus on the message-passing component, this work investigates the overlooked role of pooling operations in shaping robustness. We present a theoretical analysis of standard flat pooling methods (sum, average and max), deriving upper bounds on their adversarial risk and identifying their vulnerabilities under different attack scenarios and graph structures. Motivated by these insights, we propose \textit{Robust Singular Pooling (RS-Pool)}, a novel pooling strategy that leverages the dominant singular vector of the node embedding matrix to construct a robust graph-level representation. We theoretically investigate the robustness of RS-Pool and interpret the resulting bound leading to improved understanding of our proposed pooling operator. While our analysis centers on Graph Convolutional Networks (GCNs), RS-Pool is model-agnostic and can be implemented efficiently via power iteration. Empirical results on real-world benchmarks show that RS-Pool provides better robustness than the considered pooling methods when subject to state-of-the-art adversarial attacks while maintaining competitive clean accuracy. Our code is publicly available at:\href{https://github.com/king/rs-pool}{https://github.com/king/rs-pool}. --- ### FastVLM: Self-Speculative Decoding for Fast Vision-Language Model Inference Authors: Divya Jyoti Bajpai, Manjesh Kumar Hanawal Published: 2025-10-26 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2510.22641v1 Vision-language Models (VLMs) have made significant strides in visual understanding and query response generation, but often face challenges of high computational cost and inference latency due to autoregressive decoding. In this work, we introduce an imitation-learning-based Self-Speculative Decoding (SSD) framework, named FastVLM, to address these limitations. Our approach employs a lightweight draft model for token generation in an autoregressive manner, while a full model verifies these tokens non-autoregressively. Accepted tokens proceed seamlessly, while rejected tokens are corrected by the full model and used to guide the draft model's refinement. Through an imitation network, FastVLM enhances the draft model by integrating deeper level insights from the full model's architecture. Also, it maintains the performance integrity of the full model while training the draft model, achieving a balance between efficiency and accuracy. Our method speeds up the inference process by 1.55-1.85x as compared to the final layer with minimal loss in performance. --- ### Environment-aware Motion Matching Authors: Jose Luis Ponton, Sheldon Andrews, Carlos Andujar, Nuria Pelechano Published: 2025-10-26 Categories: cs.GR, cs.LG arXiv: https://arxiv.org/abs/2510.22632v1 Interactive applications demand believable characters that respond naturally to dynamic environments. Traditional character animation techniques often struggle to handle arbitrary situations, leading to a growing trend of dynamically selecting motion-captured animations based on predefined features. While Motion Matching has proven effective for locomotion by aligning to target trajectories, animating environment interactions and crowd behaviors remains challenging due to the need to consider surrounding elements. Existing approaches often involve manual setup or lack the naturalism of motion capture. Furthermore, in crowd animation, body animation is frequently treated as a separate process from trajectory planning, leading to inconsistencies between body pose and root motion. To address these limitations, we present Environment-aware Motion Matching, a novel real-time system for full-body character animation that dynamically adapts to obstacles and other agents, emphasizing the bidirectional relationship between pose and trajectory. In a preprocessing step, we extract shape, pose, and trajectory features from a motion capture database. At runtime, we perform an efficient search that matches user input and current pose while penalizing collisions with a dynamic environment. Our method allows characters to naturally adjust their pose and trajectory to navigate crowded scenes. --- ### Culturally Grounded Physical Commonsense Reasoning in Italian and English: A Submission to the MRL 2025 Shared Task Authors: Marco De Santis, Lisa Alazraki Published: 2025-10-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2510.22631v1 This paper presents our submission to the MRL 2025 Shared Task on Multilingual Physical Reasoning Datasets. The objective of the shared task is to create manually-annotated evaluation data in the physical commonsense reasoning domain, for languages other than English, following a format similar to PIQA. Our contribution, FormaMentis, is a novel benchmark for physical commonsense reasoning that is grounded in Italian language and culture. The data samples in FormaMentis are created by expert annotators who are native Italian speakers and are familiar with local customs and norms. The samples are additionally translated into English, while preserving the cultural elements unique to the Italian context. --- ### Robust Atypical Mitosis Classification with DenseNet121: Stain-Aware Augmentation and Hybrid Loss for Domain Generalization Authors: Adinath Dukre, Ankan Deria, Yutong Xie, Imran Razzak Published: 2025-10-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.22630v2 Atypical mitotic figures are important biomarkers of tumor aggressiveness in histopathology, yet reliable recognition remains challenging due to severe class imbalance and variability across imaging domains. We present a DenseNet-121-based framework tailored for atypical mitosis classification in the MIDOG 2025 (Track 2) setting. Our method integrates stain-aware augmentation (Macenko), geometric and intensity transformations, and imbalance-aware learning via weighted sampling with a hybrid objective combining class-weighted binary cross-entropy and focal loss. Trained end-to-end with AdamW and evaluated across multiple independent domains, the model demonstrates strong generalization under scanner and staining shifts, achieving balanced accuracy 85.0%, AUROC 0.927, sensitivity 89.2%, and specificity 80.9% on the official test set. These results indicate that combining DenseNet-121 with stain-aware augmentation and imbalance-adaptive objectives yields a robust, domain-generalizable framework for atypical mitosis classification suitable for real-world computational pathology workflows. --- ### Integrating Linguistics and AI: Morphological Analysis and Corpus development of Endangered Toto Language of West Bengal Authors: Ambalika Guha, Sajal Saha, Debanjan Ballav, Soumi Mitra, Hritwick Chakraborty Published: 2025-10-26 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2510.22629v1 Preserving linguistic diversity is necessary as every language offers a distinct perspective on the world. There have been numerous global initiatives to preserve endangered languages through documentation. This paper is a part of a project which aims to develop a trilingual (Toto-Bangla-English) language learning application to digitally archive and promote the endangered Toto language of West Bengal, India. This application, designed for both native Toto speakers and non-native learners, aims to revitalize the language by ensuring accessibility and usability through Unicode script integration and a structured language corpus. The research includes detailed linguistic documentation collected via fieldwork, followed by the creation of a morpheme-tagged, trilingual corpus used to train a Small Language Model (SLM) and a Transformer-based translation engine. The analysis covers inflectional morphology such as person-number-gender agreement, tense-aspect-mood distinctions, and case marking, alongside derivational strategies that reflect word-class changes. Script standardization and digital literacy tools were also developed to enhance script usage. The study offers a sustainable model for preserving endangered languages by incorporating traditional linguistic methodology with AI. This bridge between linguistic research with technological innovation highlights the value of interdisciplinary collaboration for community-based language revitalization. --- ### Sentra-Guard: A Real-Time Multilingual Defense Against Adversarial LLM Prompts Authors: Md. Mehedi Hasan, Sk Tanzir Mehedi, Ziaur Rahman, Rafid Mostafiz, Md. Abir Hossain Published: 2025-10-26 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2510.22628v2 This paper presents a real-time modular defense system named Sentra-Guard. The system detects and mitigates jailbreak and prompt injection attacks targeting large language models (LLMs). The framework uses a hybrid architecture with FAISS-indexed SBERT embedding representations that capture the semantic meaning of prompts, combined with fine-tuned transformer classifiers, which are machine learning models specialized for distinguishing between benign and adversarial language inputs. It identifies adversarial prompts in both direct and obfuscated attack vectors. A core innovation is the classifier-retriever fusion module, which dynamically computes context-aware risk scores that estimate how likely a prompt is to be adversarial based on its content and context. The framework ensures multilingual resilience with a language-agnostic preprocessing layer. This component automatically translates non-English prompts into English for semantic evaluation, enabling consistent detection across over 100 languages. The system includes a HITL feedback loop, where decisions made by the automated system are reviewed by human experts for continual learning and rapid adaptation under adversarial pressure. Sentra-Guard maintains an evolving dual-labeled knowledge base of benign and malicious prompts, enhancing detection reliability and reducing false positives. Evaluation results show a 99.96% detection rate (AUC = 1.00, F1 = 1.00) and an attack success rate (ASR) of only 0.004%. This outperforms leading baselines such as LlamaGuard-2 (1.3%) and OpenAI Moderation (3.7%). Unlike black-box approaches, Sentra-Guard is transparent, fine-tunable, and compatible with diverse LLM backends. Its modular design supports scalable deployment in both commercial and open-source environments. The system establishes a new state-of-the-art in adversarial LLM defense. --- ### RAMAN: Resource-efficient ApproxiMate Posit Processing for Algorithm-Hardware Co-desigN Authors: Mohd Faisal Khan, Mukul Lokhande, Santosh Kumar Vishvakarma Published: 2025-10-26 Categories: cs.AR, cs.NE arXiv: https://arxiv.org/abs/2510.22627v1 Edge-AI applications still face considerable challenges in enhancing computational efficiency in resource-constrained environments. This work presents RAMAN, a resource-efficient and approximate posit(8,2)-based Multiply-Accumulate (MAC) architecture designed to improve hardware efficiency within bandwidth limitations. The proposed REAP (Resource-Efficient Approximate Posit) MAC engine, which is at the core of RAMAN, uses approximation in the posit multiplier to achieve significant area and power reductions with an impact on accuracy. To support diverse AI workloads, this MAC unit is incorporated in a scalable Vector Execution Unit (VEU), which permits hardware reuse and parallelism among deep neural network layers. Furthermore, we propose an algorithm-hardware co-design framework incorporating approximation-aware training to evaluate the impact of hardware-level approximation on application-level performance. Empirical validation on FPGA and ASIC platforms shows that the proposed REAP MAC achieves up to 46% in LUT savings and 35.66% area, 31.28% power reduction, respectively, over the baseline Posit Dot-Product Unit (PDPU) design, while maintaining high accuracy (98.45%) for handwritten digit recognition. RAMAN demonstrates a promising trade-off between hardware efficiency and learning performance, making it suitable for next-generation edge intelligence. --- ### SwiftSolve: A Self-Iterative, Complexity-Aware Multi-Agent Framework for Competitive Programming Authors: Adhyayan Veer Singh, Aaron Shen, Brian Law, Ahmed Ismail, Jonas Rohweder, Sean O'Brien, Kevin Zhu Published: 2025-10-26 Categories: cs.AI arXiv: https://arxiv.org/abs/2510.22626v1 Correctness alone is insufficient: LLM-generated programs frequently satisfy unit tests while violating contest time or memory budgets. We present SwiftSolve, a complexity-aware multi-agent system for competitive programming that couples algorithmic planning with empirical profiling and complexity-guided repair. We frame competitive programming as a software environment where specialized agents act as programmers, each assuming roles such as planning, coding, profiling, and complexity analysis. A Planner proposes an algorithmic sketch; a deterministic Static Pruner filters high-risk plans; a Coder emits ISO C++17; a Profiler compiles and executes candidates on a fixed input-size schedule to record wall time and peak memory; and a Complexity Analyst fits log-log growth (s, R2) with an LLM fallback to assign a complexity class and dispatch targeted patches to either the Planner or Coder. Agents communicate via typed, versioned JSON; a controller enforces iteration caps and diminishing returns stopping. Evaluated on 26 problems (16 BigO, 10 Codeforces Div. 2) in a POSIX sandbox (2 s / 256-512 MB), SwiftSolve attains pass@1 = 61.54% (16/26) on the first attempt and Solved@<=3 = 80.77% with marginal latency change (mean 11.96 s to 12.66 s per attempt). Aggregate run-level success is 73.08% at 12.40 s mean. Failures are predominantly resource-bound, indicating inefficiency rather than logic errors. Against Claude Opus 4, SwiftSolve improves run-level success (73.1% vs 52.6%) at approximately 2x runtime overhead (12.4 s vs 6.8 s). Beyond correctness (pass@k), we report efficiency metrics (eff@k for runtime and memory, incidence of TLE or MLE, and complexity fit accuracy on BigO), demonstrating that profiling and complexity-guided replanning reduce inefficiency while preserving accuracy. --- ### DeepfakeBench-MM: A Comprehensive Benchmark for Multimodal Deepfake Detection Authors: Kangran Zhao, Yupeng Chen, Xiaoyu Zhang, Yize Chen, Weinan Guan, Baicheng Chen, Chengzhe Sun, Soumyya Kanti Datta, Qingshan Liu, Siwei Lyu, Baoyuan Wu Published: 2025-10-26 Categories: cs.CR, cs.CV, cs.MM arXiv: https://arxiv.org/abs/2510.22622v1 The misuse of advanced generative AI models has resulted in the widespread proliferation of falsified data, particularly forged human-centric audiovisual content, which poses substantial societal risks (e.g., financial fraud and social instability). In response to this growing threat, several works have preliminarily explored countermeasures. However, the lack of sufficient and diverse training data, along with the absence of a standardized benchmark, hinder deeper exploration. To address this challenge, we first build Mega-MMDF, a large-scale, diverse, and high-quality dataset for multimodal deepfake detection. Specifically, we employ 21 forgery pipelines through the combination of 10 audio forgery methods, 12 visual forgery methods, and 6 audio-driven face reenactment methods. Mega-MMDF currently contains 0.1 million real samples and 1.1 million forged samples, making it one of the largest and most diverse multimodal deepfake datasets, with plans for continuous expansion. Building on it, we present DeepfakeBench-MM, the first unified benchmark for multimodal deepfake detection. It establishes standardized protocols across the entire detection pipeline and serves as a versatile platform for evaluating existing methods as well as exploring novel approaches. DeepfakeBench-MM currently supports 5 datasets and 11 multimodal deepfake detectors. Furthermore, our comprehensive evaluations and in-depth analyses uncover several key findings from multiple perspectives (e.g., augmentation, stacked forgery). We believe that DeepfakeBench-MM, together with our large-scale Mega-MMDF, will serve as foundational infrastructures for advancing multimodal deepfake detection. --- ### Breaking Agent Backbones: Evaluating the Security of Backbone LLMs in AI Agents Authors: Julia Bazinska, Max Mathys, Francesco Casucci, Mateo Rojas-Carulla, Xander Davies, Alexandra Souly, Niklas Pfister Published: 2025-10-26 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2510.22620v2 AI agents powered by large language models (LLMs) are being deployed at scale, yet we lack a systematic understanding of how the choice of backbone LLM affects agent security. The non-deterministic sequential nature of AI agents complicates security modeling, while the integration of traditional software with AI components entangles novel LLM vulnerabilities with conventional security risks. Existing frameworks only partially address these challenges as they either capture specific vulnerabilities only or require modeling of complete agents. To address these limitations, we introduce threat snapshots: a framework that isolates specific states in an agent's execution flow where LLM vulnerabilities manifest, enabling the systematic identification and categorization of security risks that propagate from the LLM to the agent level. We apply this framework to construct the $b^3$ benchmark, a security benchmark based on 194,331 unique crowdsourced adversarial attacks. We then evaluate 34 popular LLMs with it, revealing, among other insights, that enhanced reasoning capabilities improve security, while model size does not correlate with security. We release our benchmark, dataset, and evaluation code to facilitate widespread adoption by LLM providers and practitioners, offering guidance for agent developers and incentivizing model developers to prioritize backbone security improvements. --- ### CLEANet: Robust and Efficient Anomaly Detection in Contaminated Multivariate Time Series Authors: Songhan Zhang, Yuanhao Lai, Pengfei Zheng, Boxi Yu, Xiaoying Tang, Qiuai Fu, Pinjia He Published: 2025-10-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.22619v1 Multivariate time series (MTS) anomaly detection is essential for maintaining the reliability of industrial systems, yet real-world deployment is hindered by two critical challenges: training data contamination (noises and hidden anomalies) and inefficient model inference. Existing unsupervised methods assume clean training data, but contamination distorts learned patterns and degrades detection accuracy. Meanwhile, complex deep models often overfit to contamination and suffer from high latency, limiting practical use. To address these challenges, we propose CLEANet, a robust and efficient anomaly detection framework in contaminated multivariate time series. CLEANet introduces a Contamination-Resilient Training Framework (CRTF) that mitigates the impact of corrupted samples through an adaptive reconstruction weighting strategy combined with clustering-guided contrastive learning, thereby enhancing robustness. To further avoid overfitting on contaminated data and improve computational efficiency, we design a lightweight conjugate MLP that disentangles temporal and cross-feature dependencies. Across five public datasets, CLEANet achieves up to 73.04% higher F1 and 81.28% lower runtime compared with ten state-of-the-art baselines. Furthermore, integrating CRTF into three advanced models yields an average 5.35% F1 gain, confirming its strong generalizability. --- ### Cross-Species Transfer Learning in Agricultural AI: Evaluating ZebraPose Adaptation for Dairy Cattle Pose Estimation Authors: Mackenzie Tapp, Sibi Chakravarthy Parivendan, Kashfia Sailunaz, Suresh Neethirajan Published: 2025-10-26 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.22618v1 Pose estimation serves as a cornerstone of computer vision for understanding animal posture, behavior, and welfare. Yet, agricultural applications remain constrained by the scarcity of large, annotated datasets for livestock, especially dairy cattle. This study evaluates the potential and limitations of cross-species transfer learning by adapting ZebraPose - a vision transformer-based model trained on synthetic zebra imagery - for 27-keypoint detection in dairy cows under real barn conditions. Using three configurations - a custom on-farm dataset (375 images, Sussex, New Brunswick, Canada), a subset of the APT-36K benchmark dataset, and their combination, we systematically assessed model accuracy and generalization across environments. While the combined model achieved promising performance (AP = 0.86, AR = 0.87, PCK 0.5 = 0.869) on in-distribution data, substantial generalization failures occurred when applied to unseen barns and cow populations. These findings expose the synthetic-to-real domain gap as a major obstacle to agricultural AI deployment and emphasize that morphological similarity between species is insufficient for cross-domain transfer. The study provides practical insights into dataset diversity, environmental variability, and computational constraints that influence real-world deployment of livestock monitoring systems. We conclude with a call for agriculture-first AI design, prioritizing farm-level realism, cross-environment robustness, and open benchmark datasets to advance trustworthy and scalable animal-centric technologies. --- ### PerCoR: Evaluating Commonsense Reasoning in Persian via Multiple-Choice Sentence Completion Authors: Morteza Alikhani, Mohammadtaha Bagherifard, Erfan Zinvandi, Mehran Sarmadi Published: 2025-10-26 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2510.22616v2 We introduced PerCoR (Persian Commonsense Reasoning), the first large-scale Persian benchmark for commonsense reasoning. PerCoR contains 106K multiple-choice sentence-completion problems drawn from more than forty news, cultural, and other web sources. We introduce a novel conjunction-based segmentation strategy to generate coherent sentence-completion pairs, enabling broad topical and structural diversity. To create challenging distractors, we propose DRESS-AF (Distractor Ranking via Embedding Similarity Scoring and Adversarial Filtering), a generation-free adversarial filtering method that selects distractors from the pool of gold continuations while maximising model confusion. Human annotators score 89% on PerCoR, while OpenAI-o3 achieves the highest performance at 92.18%, followed closely by Claude-Sonnet-3.7 (91.17%). The strongest open-source model, DeepSeek-R1, reaches 82.51%, underscoring both the dataset's difficulty and the remaining performance gap in Persian commonsense reasoning. We further show that DRESS-AF transfers to the English HellaSwag benchmark, increasing its difficulty without hurting human solvability. The dataset is available at https://huggingface.co/datasets/MCINext/PerCoR. --- ### Does In-IDE Calibration of Large Language Models work at Scale? Authors: Roham Koohestani, Agnia Sergeyuk, David Gros, Claudio Spiess, Sergey Titov, Prem Devanbu, Maliheh Izadi Published: 2025-10-26 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2510.22614v1 The introduction of large language models into integrated development environments (IDEs) is revolutionizing software engineering, yet it poses challenges to the usefulness and reliability of Artificial Intelligence-generated code. Post-hoc calibration of internal model confidences aims to align probabilities with an acceptability measure. Prior work suggests calibration can improve alignment, but at-scale evidence is limited. In this work, we investigate the feasibility of applying calibration of code models to an in-IDE context. We study two aspects of the problem: (1) the technical method for implementing confidence calibration and improving the reliability of code generation models, and (2) the human-centered design principles for effectively communicating reliability signal to developers. First, we develop a scalable and flexible calibration framework which can be used to obtain calibration weights for open-source models using any dataset, and evaluate whether calibrators improve the alignment between model confidence and developer acceptance behavior. Through a large-scale analysis of over 24 million real-world developer interactions across multiple programming languages, we find that a general, post-hoc calibration model based on Platt-scaling does not, on average, improve the reliability of model confidence signals. We also find that while dynamically personalizing calibration to individual users can be effective, its effectiveness is highly dependent on the volume of user interaction data. Second, we conduct a multi-phase design study with 3 expert designers and 153 professional developers, combining scenario-based design, semi-structured interviews, and survey validation, revealing a clear preference for presenting reliability signals via non-numerical, color-coded indicators within the in-editor code generation workflow. --- ### CLIN-LLM: A Safety-Constrained Hybrid Framework for Clinical Diagnosis and Treatment Generation Authors: Md. Mehedi Hasan, Md. Abir Hossain, Farman Hossain Sayem, Bikash Kumar Paul, Ziaur Rahman, Mohammad Shorif Uddin, Rafid Mostafiz Published: 2025-10-26 Categories: cs.AI arXiv: https://arxiv.org/abs/2510.22609v2 Accurate symptom-to-disease classification and clinically grounded treatment recommendations remain challenging, particularly in heterogeneous patient settings with high diagnostic risk. Existing large language model (LLM)-based systems often lack medical grounding and fail to quantify uncertainty, resulting in unsafe outputs. We propose CLIN-LLM, a safety-constrained hybrid pipeline that integrates multimodal patient encoding, uncertainty-calibrated disease classification, and retrieval-augmented treatment generation. The framework fine-tunes BioBERT on 1,200 clinical cases from the Symptom2Disease dataset and incorporates Focal Loss with Monte Carlo Dropout to enable confidence-aware predictions from free-text symptoms and structured vitals. Low-certainty cases (18%) are automatically flagged for expert review, ensuring human oversight. For treatment generation, CLIN-LLM employs Biomedical Sentence-BERT to retrieve top-k relevant dialogues from the 260,000-sample MedDialog corpus. The retrieved evidence and patient context are fed into a fine-tuned FLAN-T5 model for personalized treatment generation, followed by post-processing with RxNorm for antibiotic stewardship and drug-drug interaction (DDI) screening. CLIN-LLM achieves 98% accuracy and F1 score, outperforming ClinicalBERT by 7.1% (p < 0.001), with 78% top-5 retrieval precision and a clinician-rated validity of 4.2 out of 5. Unsafe antibiotic suggestions are reduced by 67% compared to GPT-5. These results demonstrate CLIN-LLM's robustness, interpretability, and clinical safety alignment. The proposed system provides a deployable, human-in-the-loop decision support framework for resource-limited healthcare environments. Future work includes integrating imaging and lab data, multilingual extensions, and clinical trial validation. --- ### SWAN: Self-supervised Wavelet Neural Network for Hyperspectral Image Unmixing Authors: Yassh Ramchandani, Vijayashekhar S S, Jignesh S. Bhatt Published: 2025-10-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.22607v1 In this article, we present SWAN: a three-stage, self-supervised wavelet neural network for joint estimation of endmembers and abundances from hyperspectral imagery. The contiguous and overlapping hyperspectral band images are first expanded to Biorthogonal wavelet basis space that provides sparse, distributed, and multi-scale representations. The idea is to exploit latent symmetries from thus obtained invariant and covariant features using a self-supervised learning paradigm. The first stage, SWANencoder maps the input wavelet coefficients to a compact lower-dimensional latent space. The second stage, SWANdecoder uses the derived latent representation to reconstruct the input wavelet coefficients. Interestingly, the third stage SWANforward learns the underlying physics of the hyperspectral image. A three-stage combined loss function is formulated in the image acquisition domain that eliminates the need for ground truth and enables self-supervised training. Adam is employed for optimizing the proposed loss function, while Sigmoid with a dropout of 0.3 is incorporated to avoid possible overfitting. Kernel regularizers bound the magnitudes and preserve spatial variations in the estimated endmember coefficients. The output of SWANencoder represents estimated abundance maps during inference, while weights of SWANdecoder are retrieved to extract endmembers. Experiments are conducted on two benchmark synthetic data sets with different signal-to-noise ratios as well as on three real benchmark hyperspectral data sets while comparing the results with several state-of-the-art neural network-based unmixing methods. The qualitative, quantitative, and ablation results show performance enhancement by learning a resilient unmixing function as well as promoting self-supervision and compact network parameters for practical applications. --- ### Projection Embedded Diffusion Bridge for CT Reconstruction from Incomplete Data Authors: Yuang Wang, Pengfei Jin, Siyeop Yoon, Matthew Tivnan, Shaoyang Zhang, Li Zhang, Quanzheng Li, Zhiqiang Chen, Dufan Wu Published: 2025-10-26 Categories: cs.CV, physics.med-ph arXiv: https://arxiv.org/abs/2510.22605v1 Reconstructing CT images from incomplete projection data remains challenging due to the ill-posed nature of the problem. Diffusion bridge models have recently shown promise in restoring clean images from their corresponding Filtered Back Projection (FBP) reconstructions, but incorporating data consistency into these models remains largely underexplored. Incorporating data consistency can improve reconstruction fidelity by aligning the reconstructed image with the observed projection data, and can enhance detail recovery by integrating structural information contained in the projections. In this work, we propose the Projection Embedded Diffusion Bridge (PEDB). PEDB introduces a novel reverse stochastic differential equation (SDE) to sample from the distribution of clean images conditioned on both the FBP reconstruction and the incomplete projection data. By explicitly conditioning on the projection data in sampling the clean images, PEDB naturally incorporates data consistency. We embed the projection data into the score function of the reverse SDE. Under certain assumptions, we derive a tractable expression for the posterior score. In addition, we introduce a free parameter to control the level of stochasticity in the reverse process. We also design a discretization scheme for the reverse SDE to mitigate discretization error. Extensive experiments demonstrate that PEDB achieves strong performance in CT reconstruction from three types of incomplete data, including sparse-view, limited-angle, and truncated projections. For each of these types, PEDB outperforms evaluated state-of-the-art diffusion bridge models across standard, noisy, and domain-shift evaluations. --- ### Mitigating Attention Sinks and Massive Activations in Audio-Visual Speech Recognition with LLMs Authors: Anand, Umberto Cappellazzo, Stavros Petridis, Maja Pantic Published: 2025-10-26 Categories: eess.AS, cs.CV, cs.SD arXiv: https://arxiv.org/abs/2510.22603v3 Large language models (LLMs) have recently advanced auditory speech recognition (ASR), visual speech recognition (VSR), and audio-visual speech recognition (AVSR). However, understanding of their internal dynamics under fine-tuning remains limited. In natural language processing, recent work has revealed attention sinks, tokens that attract disproportionately high attention, and associated massive activations in which some features of sink tokens exhibit huge activation in LLMs. In this work, we are the first to study these phenomena in multimodal speech recognition. Through a detailed analysis of audio-visual LLMs, we identify attention sinks and massive activations not only at the BOS token but also at intermediate low-semantic tokens across ASR, VSR, and AVSR. We show that massive activations originate in the MLP layers and correspond to fixed feature indices across all sink tokens. We further show that intermediate sink tokens exhibit high cosine similarity to the BOS token, thereby amplifying attention and activation. Building on these insights, we introduce a simple decorrelation loss that reduces cosine similarity between BOS and other tokens, effectively mitigating intermediate sinks and massive activations. Furthermore, our method improves word error rate (WER) under high audio-visual feature downsampling while remaining stable at lower downsampling rates. --- ### Personal Care Utility (PCU): Building the Health Infrastructure for Everyday Insight and Guidance Authors: Mahyar Abbasian, Ramesh Jain Published: 2025-10-26 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2510.22602v1 Building on decades of success in digital infrastructure and biomedical innovation, we propose the Personal Care Utility (PCU) - a cybernetic system for lifelong health guidance. PCU is conceived as a global, AI-powered utility that continuously orchestrates multimodal data, knowledge, and services to assist individuals and populations alike. Drawing on multimodal agents, event-centric modeling, and contextual inference, it offers three essential capabilities: (1) trusted health information tailored to the individual, (2) proactive health navigation and behavior guidance, and (3) ongoing interpretation of recovery and treatment response after medical events. Unlike conventional episodic care, PCU functions as an ambient, adaptive companion - observing, interpreting, and guiding health in real time across daily life. By integrating personal sensing, experiential computing, and population-level analytics, PCU promises not only improved outcomes for individuals but also a new substrate for public health and scientific discovery. We describe the architecture, design principles, and implementation challenges of this emerging paradigm. --- ### RoGER-SLAM: A Robust Gaussian Splatting SLAM System for Noisy and Low-light Environment Resilience Authors: Huilin Yin, Zhaolin Yang, Linchuan Zhang, Gerhard Rigoll, Johannes Betz Published: 2025-10-26 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2510.22600v1 The reliability of Simultaneous Localization and Mapping (SLAM) is severely constrained in environments where visual inputs suffer from noise and low illumination. Although recent 3D Gaussian Splatting (3DGS) based SLAM frameworks achieve high-fidelity mapping under clean conditions, they remain vulnerable to compounded degradations that degrade mapping and tracking performance. A key observation underlying our work is that the original 3DGS rendering pipeline inherently behaves as an implicit low-pass filter, attenuating high-frequency noise but also risking over-smoothing. Building on this insight, we propose RoGER-SLAM, a robust 3DGS SLAM system tailored for noise and low-light resilience. The framework integrates three innovations: a Structure-Preserving Robust Fusion (SP-RoFusion) mechanism that couples rendered appearance, depth, and edge cues; an adaptive tracking objective with residual balancing regularization; and a Contrastive Language-Image Pretraining (CLIP)-based enhancement module, selectively activated under compounded degradations to restore semantic and structural fidelity. Comprehensive experiments on Replica, TUM, and real-world sequences show that RoGER-SLAM consistently improves trajectory accuracy and reconstruction quality compared with other 3DGS-SLAM systems, especially under adverse imaging conditions. --- ### A roadmap for curvature-based geometric data analysis and learning Authors: Yasharth Yadav, Kelin Xia Published: 2025-10-26 Categories: cs.LG, math.DG arXiv: https://arxiv.org/abs/2510.22599v1 Geometric data analysis and learning has emerged as a distinct and rapidly developing research area, increasingly recognized for its effectiveness across diverse applications. At the heart of this field lies curvature, a powerful and interpretable concept that captures intrinsic geometric structure and underpins numerous tasks, from community detection to geometric deep learning. A wide range of discrete curvature models have been proposed for various data representations, including graphs, simplicial complexes, cubical complexes, and point clouds sampled from manifolds. These models not only provide efficient characterizations of data geometry but also constitute essential components in geometric learning frameworks. In this paper, we present the first comprehensive review of existing discrete curvature models, covering their mathematical foundations, computational formulations, and practical applications in data analysis and learning. In particular, we discuss discrete curvature from both Riemannian and metric geometry perspectives and propose a systematic pipeline for curvature-driven data analysis. We further examine the corresponding computational algorithms across different data representations, offering detailed comparisons and insights. Finally, we review state-of-the-art applications of curvature in both supervised and unsupervised learning. This survey provides a conceptual and practical roadmap for researchers to gain a better understanding of discrete curvature as a fundamental tool for geometric understanding and learning. --- ### A Framework for Quantifying How Pre-Training and Context Benefit In-Context Learning Authors: Bingqing Song, Jiaxiang Li, Rong Wang, Songtao Lu, Mingyi Hong Published: 2025-10-26 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2510.22594v1 Pre-trained large language models have demonstrated a strong ability to learn from context, known as in-context learning (ICL). Despite a surge of recent applications that leverage such capabilities, it is by no means clear, at least theoretically, how the ICL capabilities arise, and in particular, what is the precise role played by key factors such as pre-training procedure as well as context construction. In this work, we propose a new framework to analyze the ICL performance, for a class of realistic settings, which includes network architectures, data encoding, data generation, and prompt construction process. As a first step, we construct a simple example with a one-layer transformer, and show an interesting result, namely when the pre-train data distribution is different from the query task distribution, a properly constructed context can shift the output distribution towards the query task distribution, in a quantifiable manner, leading to accurate prediction on the query topic. We then extend the findings in the previous step to a more general case, and derive the precise relationship between ICL performance, context length and the KL divergence between pre-train and query task distribution. Finally, we provide experiments to validate our theoretical results. --- ### AutoBench: Automating LLM Evaluation through Reciprocal Peer Assessment Authors: Dario Loi, Elena Maria Muià, Federico Siciliano, Giovanni Trappolini, Vincenzo Crisà, Peter Kruger, Fabrizio Silvestri Published: 2025-10-26 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2510.22593v1 We present AutoBench, a fully automated and self-sustaining framework for evaluating Large Language Models (LLMs) through reciprocal peer assessment. This paper provides a rigorous scientific validation of the AutoBench methodology, originally developed as an open-source project by eZecute S.R.L.. Unlike static benchmarks that suffer from test-set contamination and limited adaptability, AutoBench dynamically generates novel evaluation tasks while models alternately serve as question generators, contestants, and judges across diverse domains. An iterative weighting mechanism amplifies the influence of consistently reliable evaluators, aggregating peer judgments into consensus-based rankings that reflect collective model agreement. Our experiments demonstrate strong correlations with established benchmarks including MMLU-Pro and GPQA (respectively 78\% and 63\%), validating this peer-driven evaluation paradigm. The multi-judge design significantly outperforms single-judge baselines, confirming that distributed evaluation produces more robust and human-consistent assessments. AutoBench offers a scalable, contamination-resistant alternative to static benchmarks for the continuous evaluation of evolving language models. --- ### ATOM: AdapTive and OptiMized dynamic temporal knowledge graph construction using LLMs Authors: Yassir Lairgi, Ludovic Moncla, Khalid Benabdeslem, Rémy Cazabet, Pierre Cléau Published: 2025-10-26 Categories: cs.AI, cs.CL, cs.IR arXiv: https://arxiv.org/abs/2510.22590v2 In today's rapidly expanding data landscape, knowledge extraction from unstructured text is vital for real-time analytics, temporal inference, and dynamic memory frameworks. However, traditional static knowledge graph (KG) construction often overlooks the dynamic and time-sensitive nature of real-world data, limiting adaptability to continuous changes. Moreover, recent zero- or few-shot approaches that avoid domain-specific fine-tuning or reliance on prebuilt ontologies often suffer from instability across multiple runs, as well as incomplete coverage of key facts. To address these challenges, we introduce ATOM (AdapTive and OptiMized), a few-shot and scalable approach that builds and continuously updates Temporal Knowledge Graphs (TKGs) from unstructured texts. ATOM splits input documents into minimal, self-contained "atomic" facts, improving extraction exhaustivity and stability. Then, it constructs atomic TKGs from these facts, employing a dual-time modeling that distinguishes between when information is observed and when it is valid. The resulting atomic TKGs are subsequently merged in parallel. Empirical evaluations demonstrate that ATOM achieves ~18% higher exhaustivity, ~33% better stability, and over ~90% latency reduction compared to baseline methods, demonstrating a strong scalability potential for dynamic TKG construction. --- ### PSScreen V2: Partially Supervised Multiple Retinal Disease Screening Authors: Boyi Zheng, Yalin Zheng, Hrvoje Bogunović, Qing Liu Published: 2025-10-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.22589v2 In this work, we propose PSScreen V2, a partially supervised self-training framework for multiple retinal disease screening. Unlike previous methods that rely on fully labelled or single-domain datasets, PSScreen V2 is designed to learn from multiple partially labelled datasets with different distributions, addressing both label absence and domain shift challenges. To this end, PSScreen V2 adopts a three-branch architecture with one teacher and two student networks. The teacher branch generates pseudo labels from weakly augmented images to address missing labels, while the two student branches introduce novel feature augmentation strategies: Low-Frequency Dropout (LF-Dropout), which enhances domain robustness by randomly discarding domain-related low-frequency components, and Low-Frequency Uncertainty (LF-Uncert), which estimates uncertain domain variability via adversarially learned Gaussian perturbations of low-frequency statistics. Extensive experiments on multiple in-domain and out-of-domain fundus datasets demonstrate that PSScreen V2 achieves state-of-the-art performance and superior domain generalization ability. Furthermore, compatibility tests with diverse backbones, including the vision foundation model DINOv2, as well as evaluations on chest X-ray datasets, highlight the universality and adaptability of the proposed framework. The codes are available at https://github.com/boyiZheng99/PSScreen_V2. --- ### UltraVoice: Scaling Fine-Grained Style-Controlled Speech Conversations for Spoken Dialogue Models Authors: Wenming Tu, Guanrou Yang, Ruiqi Yan, Wenxi Chen, Ziyang Ma, Yipeng Kang, Kai Yu, Xie Chen, Zilong Zheng Published: 2025-10-26 Categories: eess.AS, cs.CL arXiv: https://arxiv.org/abs/2510.22588v2 Spoken dialogue models currently lack the ability for fine-grained speech style control, a critical capability for human-like interaction that is often overlooked in favor of purely functional capabilities like reasoning and question answering. To address this limitation, we introduce UltraVoice, the first large-scale speech dialogue dataset engineered for multiple fine-grained speech style control. Encompassing over 830 hours of speech dialogues, UltraVoice provides instructions across six key speech stylistic dimensions: emotion, speed, volume, accent, language, and composite styles. Fine-tuning leading models such as SLAM-Omni and VocalNet on UltraVoice significantly enhances their fine-grained speech stylistic controllability without degrading core conversational abilities. Specifically, our fine-tuned models achieve improvements of 29.12-42.33% in Mean Opinion Score (MOS) and 14.61-40.09 percentage points in Instruction Following Rate (IFR) on multi-dimensional control tasks designed in the UltraVoice. Moreover, on the URO-Bench benchmark, our fine-tuned models demonstrate substantial gains in core understanding, reasoning, and conversational abilities, with average improvements of +10.84% on the Basic setting and +7.87% on the Pro setting. Furthermore, the dataset's utility extends to training controllable Text-to-Speech (TTS) models, underscoring its high quality and broad applicability for expressive speech synthesis. The complete dataset and model checkpoints are available at: https://github.com/bigai-nlco/UltraVoice. --- ### Prediction-Powered Semi-Supervised Learning with Online Power Tuning Authors: Noa Shoham, Ron Dorfman, Shalev Shaer, Kfir Y. Levy, Yaniv Romano Published: 2025-10-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.22586v1 Prediction-Powered Inference (PPI) is a recently proposed statistical inference technique for parameter estimation that leverages pseudo-labels on both labeled and unlabeled data to construct an unbiased, low-variance estimator. In this work, we extend its core idea to semi-supervised learning (SSL) for model training, introducing a novel unbiased gradient estimator. This extension addresses a key challenge in SSL: while unlabeled data can improve model performance, its benefit heavily depends on the quality of pseudo-labels. Inaccurate pseudo-labels can introduce bias, leading to suboptimal models.To balance the contributions of labeled and pseudo-labeled data, we utilize an interpolation parameter and tune it on the fly, alongside the model parameters, using a one-dimensional online learning algorithm. We verify the practical advantage of our approach through experiments on both synthetic and real datasets, demonstrating improved performance over classic SSL baselines and PPI methods that tune the interpolation parameter offline. --- ### MobileGeo: Exploring Hierarchical Knowledge Distillation for Resource-Efficient Cross-view Drone Geo-Localization Authors: Jian Sun, Kangdao Liu, Chi Zhang, Chuangquan Chen, Junge Shen, C. L. Philip Chen, Chi-Man Vong Published: 2025-10-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.22582v3 Cross-view geo-localization (CVGL) plays a vital role in drone-based multimedia applications, enabling precise localization by matching drone-captured aerial images against geo-tagged satellite databases in GNSS-denied environments. However, existing methods rely on resource-intensive feature alignment and multi-branch architectures, incurring high inference costs that limit their deployment on edge devices. We propose MobileGeo, a mobile-friendly framework designed for efficient on-device CVGL: 1) During training, a Hierarchical Distillation (HD-CVGL) paradigm, coupled with Uncertainty-Aware Prediction Alignment (UAPA), distills essential information into a compact model without incurring inference overhead. 2) During inference, an efficient Multi-view Selection Refinement Module (MSRM) leverages mutual information to filter redundant views and reduce computational load. Extensive experiments demonstrate that MobileGeo outperforms previous state-of-the-art methods, achieving a 4.19% improvement in AP on University1652 dataset while being over 5 times efficient in FLOPs and 3 times faster. Crucially, MobileGeo runs at 251.5 FPS on an NVIDIA AGX Orin edge device, demonstrating its practical viability for real-time on-device drone geo-localization. The code is available at https://github.com/SkyEyeLoc/MobileGeo. --- ### Pedagogy-driven Evaluation of Generative AI-powered Intelligent Tutoring Systems Authors: Kaushal Kumar Maurya, Ekaterina Kochmar Published: 2025-10-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2510.22581v1 The interdisciplinary research domain of Artificial Intelligence in Education (AIED) has a long history of developing Intelligent Tutoring Systems (ITSs) by integrating insights from technological advancements, educational theories, and cognitive psychology. The remarkable success of generative AI (GenAI) models has accelerated the development of large language model (LLM)-powered ITSs, which have potential to imitate human-like, pedagogically rich, and cognitively demanding tutoring. However, the progress and impact of these systems remain largely untraceable due to the absence of reliable, universally accepted, and pedagogy-driven evaluation frameworks and benchmarks. Most existing educational dialogue-based ITS evaluations rely on subjective protocols and non-standardized benchmarks, leading to inconsistencies and limited generalizability. In this work, we take a step back from mainstream ITS development and provide comprehensive state-of-the-art evaluation practices, highlighting associated challenges through real-world case studies from careful and caring AIED research. Finally, building on insights from previous interdisciplinary AIED research, we propose three practical, feasible, and theoretically grounded research directions, rooted in learning science principles and aimed at establishing fair, unified, and scalable evaluation methodologies for ITSs. --- ### Optimal Anytime Algorithms for Online Convex Optimization with Adversarial Constraints Authors: Dhruv Sarkar, Abhishek Sinha Published: 2025-10-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.22579v1 We propose an anytime online algorithm for the problem of learning a sequence of adversarial convex cost functions while approximately satisfying another sequence of adversarial online convex constraints. A sequential algorithm is called \emph{anytime} if it provides a non-trivial performance guarantee for any intermediate timestep $t$ without requiring prior knowledge of the length of the entire time horizon $T$. Our proposed algorithm achieves optimal performance bounds without resorting to the standard doubling trick, which has poor practical performance due to multiple restarts. Our core technical contribution is the use of time-varying Lyapunov functions to keep track of constraint violations. This must be contrasted with prior works that used a fixed Lyapunov function tuned to the known horizon length $T$. The use of time-varying Lyapunov function poses unique analytical challenges as properties, such as \emph{monotonicity}, on which the prior proofs rest, no longer hold. By introducing a new analytical technique, we show that our algorithm achieves $O(\sqrt{t})$ regret and $\tilde{O}(\sqrt{t})$ cumulative constraint violation bounds for any $t\geq 1$. We extend our results to the dynamic regret setting, achieving bounds that adapt to the path length of the comparator sequence without prior knowledge of its total length. We also present an adaptive algorithm in the optimistic setting, whose performance gracefully scales with the cumulative prediction error. We demonstrate the practical utility of our algorithm through numerical experiments involving the online shortest path problem. --- ### From Pixels to Views: Learning Angular-Aware and Physics-Consistent Representations for Light Field Microscopy Authors: Feng He, Guodong Tan, Qiankun Li, Jun Yu, Quan Wen Published: 2025-10-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.22577v1 Light field microscopy (LFM) has become an emerging tool in neuroscience for large-scale neural imaging in vivo, notable for its single-exposure volumetric imaging, broad field of view, and high temporal resolution. However, learning-based 3D reconstruction in XLFM remains underdeveloped due to two core challenges: the absence of standardized datasets and the lack of methods that can efficiently model its angular-spatial structure while remaining physically grounded. We address these challenges by introducing three key contributions. First, we construct the XLFM-Zebrafish benchmark, a large-scale dataset and evaluation suite for XLFM reconstruction. Second, we propose Masked View Modeling for Light Fields (MVN-LF), a self-supervised task that learns angular priors by predicting occluded views, improving data efficiency. Third, we formulate the Optical Rendering Consistency Loss (ORC Loss), a differentiable rendering constraint that enforces alignment between predicted volumes and their PSF-based forward projections. On the XLFM-Zebrafish benchmark, our method improves PSNR by 7.7% over state-of-the-art baselines. ---