# Research Papers — 2025-10 · bundle page 79 of 217 > Full-text markdown of papers #3901–3950 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/79 [← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/78.md) [Next bundle →](https://aipod.dev/papers/archive/2025-10/page/80.md) ### CBINNS: Cancer Biology-Informed Neural Network for Unknown Parameter Estimation and Missing Physics Identification Authors: Bishal Chhetri, B. V. Rathish Kumar Published: 2025-10-20 Categories: q-bio.QM, cs.AI arXiv: https://arxiv.org/abs/2510.17920v1 The dynamics of tumor-immune interactions within a complex tumor microenvironment are typically modeled using a system of ordinary differential equations or partial differential equations. These models introduce some unknown parameters that need to be estimated accurately and efficiently from the limited and noisy experimental data. Moreover, due to the intricate biological complexity and limitations in experimental measurements, tumor-immune dynamics are not fully understood, and therefore, only partial knowledge of the underlying physics may be available, resulting in unknown or missing terms within the system of equations. In this study, we develop a cancer biology-informed neural network model(CBINN) to infer the unknown parameters in the system of equations as well as to discover the missing physics from sparse and noisy measurements. We test the performance of the CBINN model on three distinct nonlinear compartmental tumor-immune models and evaluate its robustness across multiple synthetic noise levels. By harnessing these highly nonlinear dynamics, our CBINN framework effectively estimates the unknown model parameters and uncovers the underlying physical laws or mathematical structures that govern these biological systems, even from scattered and noisy measurements. The models chosen here represent the dynamic patterns commonly observed in compartmental models of tumor-immune interactions, thereby validating the generalizability and efficacy of our methodology. --- ### ParaVul: A Parallel Large Language Model and Retrieval-Augmented Framework for Smart Contract Vulnerability Detection Authors: Tenghui Huang, Jinbo Wen, Jiawen Kang, Siyong Chen, Zhengtao Li, Tao Zhang, Dongning Liu, Jiacheng Wang, Chengjun Cai, Yinqiu Liu, Dusit Niyato Published: 2025-10-20 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2510.17919v1 Smart contracts play a significant role in automating blockchain services. Nevertheless, vulnerabilities in smart contracts pose serious threats to blockchain security. Currently, traditional detection methods primarily rely on static analysis and formal verification, which can result in high false-positive rates and poor scalability. Large Language Models (LLMs) have recently made significant progress in smart contract vulnerability detection. However, they still face challenges such as high inference costs and substantial computational overhead. In this paper, we propose ParaVul, a parallel LLM and retrieval-augmented framework to improve the reliability and accuracy of smart contract vulnerability detection. Specifically, we first develop Sparse Low-Rank Adaptation (SLoRA) for LLM fine-tuning. SLoRA introduces sparsification by incorporating a sparse matrix into quantized LoRA-based LLMs, thereby reducing computational overhead and resource requirements while enhancing their ability to understand vulnerability-related issues. We then construct a vulnerability contract dataset and develop a hybrid Retrieval-Augmented Generation (RAG) system that integrates dense retrieval with Best Matching 25 (BM25), assisting in verifying the results generated by the LLM. Furthermore, we propose a meta-learning model to fuse the outputs of the RAG system and the LLM, thereby generating the final detection results. After completing vulnerability detection, we design chain-of-thought prompts to guide LLMs to generate comprehensive vulnerability detection reports. Simulation results demonstrate the superiority of ParaVul, especially in terms of F1 scores, achieving 0.9398 for single-label detection and 0.9330 for multi-label detection. --- ### JT-Safe: Intrinsically Enhancing the Safety and Trustworthiness of LLMs Authors: Junlan Feng, Fanyu Meng, Chong Long, Pengyu Cong, Duqing Wang, Yan Zheng, Yuyao Zhang, Xuanchang Gao, Ye Yuan, Yunfei Ma, Zhijie Ren, Fan Yang et al. Published: 2025-10-20 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2510.17918v1 The hallucination and credibility concerns of large language models (LLMs) are global challenges that the industry is collectively addressing. Recently, a significant amount of advances have been made on post-training and inference techniques to mitigate these challenges. However, it is widely agreed that unsafe and hallucinations of LLMs intrinsically originate from pre-training, involving pre-training data and the next-token prediction learning mechanism. In this paper, we focus on enhancing pre-training data to improve the trustworthiness and safety of LLMs. Since the data is vast, it's almost impossible to entirely purge the data of factual errors, logical inconsistencies, or distributional biases. Moreover, the pre-training data lack grounding in real-world knowledge. Each piece of data is treated as a sequence of tokens rather than as a representation of a part of the world. To overcome these issues, we propose approaches to enhancing our pre-training data with its context in the world and increasing a substantial amount of data reflecting industrial scenarios. We argue that most source data are created by the authors for specific purposes in a certain spatial-temporal context. They have played a role in the real world. By incorporating related world context information, we aim to better anchor pre-training data within real-world scenarios, thereby reducing uncertainty in model training and enhancing the model's safety and trustworthiness. We refer to our Data with World Context as DWC. We continue pre-training an earlier checkpoint of JT-35B-Base with 1.5 trillion of DWC tokens. We introduce our post-training procedures to activate the potentials of DWC. Compared with the Qwen model of a similar scale, JT-Safe-35B achieves an average performance improvement of 1.79% on the Safety and Trustworthy evaluation benchmarks, while being pretrained with only 6.2 trillion tokens. --- ### Not Every Time and Frequency Need to Be Forgotten in Diffusion Unlearning Authors: Jinseong Park, Mijung Park Published: 2025-10-20 Categories: cs.LG, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2510.17917v2 Data unlearning aims to remove the influence of specific training samples from a trained model. In fine-tuning methods, data unlearning relies primarily on loss maximization over forget samples, which often leads to quality degradation or incomplete forgetting. Existing methods perform unlearning uniformly across diffusion stages, ignoring diffusion dynamics from noise to data. Our systematic study of diffusion phases shows that forgetting in diffusion models is uneven across time and frequency, with theoretical justification of distributive distortion and forgetting-utility trade-off. By selectively forgetting time and frequency in diffusion models, we achieve both higher unlearning success rates and improved generation quality across diverse settings, including both conditional and unconditional scenarios. We also introduce an improved SSCD metric that measures dissimilarity using a normalized perturbation distance. Together, we provide practical insights for understanding and improving data unlearning in diffusion models. --- ### Self-Evidencing Through Hierarchical Gradient Decomposition: A Dissipative System That Maintains Non-Equilibrium Steady-State by Minimizing Variational Free Energy Authors: Michael James McCulloch Published: 2025-10-20 Categories: cs.NE, cs.AI, cs.LG, q-bio.NC arXiv: https://arxiv.org/abs/2510.17916v1 The Free Energy Principle (FEP) states that self-organizing systems must minimize variational free energy to persist, but the path from principle to implementable algorithm has remained unclear. We present a constructive proof that the FEP can be realized through exact local credit assignment. The system decomposes gradient computation hierarchically: spatial credit via feedback alignment, temporal credit via eligibility traces, and structural credit via a Trophic Field Map (TFM) that estimates expected gradient magnitude for each connection block. We prove these mechanisms are exact at their respective levels and validate the central claim empirically: the TFM achieves 0.9693 Pearson correlation with oracle gradients. This exactness produces emergent capabilities including 98.6% retention after task interference, autonomous recovery from 75% structural damage, self-organized criticality (spectral radius p ~= 1.0$), and sample-efficient reinforcement learning on continuous control tasks without replay buffers. The architecture unifies Prigogine's dissipative structures, Friston's free energy minimization, and Hopfield's attractor dynamics, demonstrating that exact hierarchical inference over network topology can be implemented with local, biologically plausible rules. --- ### ConsistEdit: Highly Consistent and Precise Training-free Visual Editing Authors: Zixin Yin, Ling-Hao Chen, Lionel Ni, Xili Dai Published: 2025-10-20 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.17803v1 Recent advances in training-free attention control methods have enabled flexible and efficient text-guided editing capabilities for existing generation models. However, current approaches struggle to simultaneously deliver strong editing strength while preserving consistency with the source. This limitation becomes particularly critical in multi-round and video editing, where visual errors can accumulate over time. Moreover, most existing methods enforce global consistency, which limits their ability to modify individual attributes such as texture while preserving others, thereby hindering fine-grained editing. Recently, the architectural shift from U-Net to MM-DiT has brought significant improvements in generative performance and introduced a novel mechanism for integrating text and vision modalities. These advancements pave the way for overcoming challenges that previous methods failed to resolve. Through an in-depth analysis of MM-DiT, we identify three key insights into its attention mechanisms. Building on these, we propose ConsistEdit, a novel attention control method specifically tailored for MM-DiT. ConsistEdit incorporates vision-only attention control, mask-guided pre-attention fusion, and differentiated manipulation of the query, key, and value tokens to produce consistent, prompt-aligned edits. Extensive experiments demonstrate that ConsistEdit achieves state-of-the-art performance across a wide range of image and video editing tasks, including both structure-consistent and structure-inconsistent scenarios. Unlike prior methods, it is the first approach to perform editing across all inference steps and attention layers without handcraft, significantly enhancing reliability and consistency, which enables robust multi-round and multi-region editing. Furthermore, it supports progressive adjustment of structural consistency, enabling finer control. --- ### Unbiased Gradient Low-Rank Projection Authors: Rui Pan, Yang Luo, Yuxing Liu, Yang You, Tong Zhang Published: 2025-10-20 Categories: cs.LG, cs.AI, math.OC arXiv: https://arxiv.org/abs/2510.17802v1 Memory-efficient optimization is critical for training increasingly large language models (LLMs). A popular strategy involves gradient low-rank projection, storing only the projected optimizer states, with GaLore being a representative example. However, a significant drawback of many such methods is their lack of convergence guarantees, as various low-rank projection approaches introduce inherent biases relative to the original optimization algorithms, which contribute to performance gaps compared to full-parameter training. Aiming to tackle this problem, this paper investigates the layerwise sampling technique for debiasing low-rank projection mechanisms. In particular, an instantiation of the paradigm gives rise to a novel and unbiased low-rank optimization method built upon GaLore's mechanism and the Muon algorithm, named GaLore Unbiased with Muon (GUM). We theoretically prove our method matches the convergence guarantees of the base Muon algorithm while preserving the memory efficiency of low-rank techniques. Empirical experiments on LLM fine-tuning and pretraining also demonstrate non-trivial improvements over GaLore and even better performance than full-parameter training. Further investigation shows that the improvement of this technique comes from a more uniform distribution of knowledge inside layers, leading to more efficient utilization of the model parameter space and better memorization. --- ### Robobench: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models as Embodied Brain Authors: Yulin Luo, Chun-Kai Fan, Menghang Dong, Jiayu Shi, Xiangju Mi, Mengdi Zhao, Bo-Wen Zhang, Cheng Chi, Jiaming Liu, Gaole Dai, Rongyu Zhang, Ruichuan An et al. Published: 2025-10-20 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2510.17801v2 Building robots that can perceive, reason, and act in dynamic, unstructured environments remains a central challenge. Recent embodied systems often follow a dual-system paradigm, where System 2 performs high-level reasoning and System 1 handles low-level control. We refer to System 2 as the embodied brain, the cognitive core for decision-making in manipulation. Although evaluating this embodied brain is crucial, existing benchmarks mainly measure execution success or cover only limited aspects of high-level cognition and task realism. We introduce RoboBench, a benchmark for evaluating multimodal large language models (MLLMs) as embodied brains. RoboBench covers five dimensions: Instruction Comprehension, Perception Reasoning, Generalized Planning, Affordance Prediction, and Failure Analysis. It spans 14 capabilities, 25 tasks, and 6,092 QA pairs. To improve realism, it draws from large-scale real robotic data and in-house collection across diverse embodiments, attribute-rich objects, multi-view scenes, and memory-driven navigation. For planning, RoboBench introduces an MLLM-as-world-simulator framework that assesses whether predicted plans can achieve critical object-state changes under physical and visual constraints, enabling more faithful evaluation of long-horizon reasoning than symbolic matching. Experiments on 18 state-of-the-art MLLMs reveal persistent limitations in implicit instruction understanding, spatiotemporal reasoning, cross-scenario planning, fine-grained affordance understanding, and failure diagnosis. We further analyze how embodied cognitive abilities relate to downstream robotic control. RoboBench offers a comprehensive scaffold for quantifying high-level cognition and guiding next-generation MLLMs toward more robust robotic intelligence. --- ### Glyph: Scaling Context Windows via Visual-Text Compression Authors: Jiale Cheng, Yusen Liu, Xinyu Zhang, Yulin Fei, Wenyi Hong, Ruiliang Lyu, Weihan Wang, Zhe Su, Xiaotao Gu, Xiao Liu, Yushi Bai, Jie Tang et al. Published: 2025-10-20 Categories: cs.CV, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2510.17800v2 Large language models (LLMs) increasingly rely on long-context modeling for tasks such as document understanding, code analysis, and multi-step reasoning. However, scaling context windows to the million-token level brings prohibitive computational and memory costs, limiting the practicality of long-context LLMs. In this work, we take a different perspective-visual context scaling-to tackle this challenge. Instead of extending token-based sequences, we propose Glyph, a framework that renders long texts into images and processes them with vision-language models (VLMs). This approach substantially compresses textual input while preserving semantic information, and we further design an LLM-driven genetic search to identify optimal visual rendering configurations for balancing accuracy and compression. Through extensive experiments, we demonstrate that our method achieves 3-4x token compression while maintaining accuracy comparable to leading LLMs such as Qwen3-8B on various long-context benchmarks. This compression also leads to around 4x faster prefilling and decoding, and approximately 2x faster SFT training. Furthermore, under extreme compression, a 128K-context VLM could scale to handle 1M-token-level text tasks. In addition, the rendered text data benefits real-world multimodal tasks, such as document understanding. Our code and model are released at https://github.com/thu-coai/Glyph. --- ### Enterprise Deep Research: Steerable Multi-Agent Deep Research for Enterprise Analytics Authors: Akshara Prabhakar, Roshan Ram, Zixiang Chen, Silvio Savarese, Frank Wang, Caiming Xiong, Huan Wang, Weiran Yao Published: 2025-10-20 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2510.17797v2 As information grows exponentially, enterprises face increasing pressure to transform unstructured data into coherent, actionable insights. While autonomous agents show promise, they often struggle with domain-specific nuances, intent alignment, and enterprise integration. We present Enterprise Deep Research (EDR), a multi-agent system that integrates (1) a Master Planning Agent for adaptive query decomposition, (2) four specialized search agents (General, Academic, GitHub, LinkedIn), (3) an extensible MCP-based tool ecosystem supporting NL2SQL, file analysis, and enterprise workflows, (4) a Visualization Agent for data-driven insights, and (5) a reflection mechanism that detects knowledge gaps and updates research direction with optional human-in-the-loop steering guidance. These components enable automated report generation, real-time streaming, and seamless enterprise deployment, as validated on internal datasets. On open-ended benchmarks including DeepResearch Bench and DeepConsult, EDR outperforms state-of-the-art agentic systems without any human steering. We release the EDR framework and benchmark trajectories to advance research on multi-agent reasoning applications. Code at https://github.com/SalesforceAIResearch/enterprise-deep-research and Dataset at https://huggingface.co/datasets/Salesforce/EDR-200 --- ### What Makes AI Research Replicable? Executable Knowledge Graphs as Scientific Knowledge Representations Authors: Yujie Luo, Zhuoyun Yu, Xuehai Wang, Yuqi Zhu, Ningyu Zhang, Lanning Wei, Lun Du, Da Zheng, Huajun Chen Published: 2025-10-20 Categories: cs.CL, cs.AI, cs.LG, cs.MA, cs.SE arXiv: https://arxiv.org/abs/2510.17795v3 Replicating AI research is a crucial yet challenging task for large language model (LLM) agents. Existing approaches often struggle to generate executable code, primarily due to insufficient background knowledge and the limitations of retrieval-augmented generation (RAG) methods, which fail to capture latent technical details hidden in referenced papers. Furthermore, previous approaches tend to overlook valuable implementation-level code signals and lack structured knowledge representations that support multi-granular retrieval and reuse. To overcome these challenges, we propose Executable Knowledge Graphs (xKG), a pluggable, paper-centric knowledge base that automatically integrates code snippets and technical insights extracted from scientific literature. When integrated into three agent frameworks with two different LLMs, xKG shows substantial performance gains (10.9% with o3-mini) on PaperBench, demonstrating its effectiveness as a general and extensible solution for automated AI research replication. Code is available at https://github.com/zjunlp/xKG. --- ### Functional Distribution Networks (FDN) Authors: Omer Haq Published: 2025-10-20 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2510.17794v3 Modern probabilistic regressors often remain overconfident under distribution shift. Functional Distribution Networks (FDN) place input-conditioned distributions over network weights, producing predictive mixtures whose dispersion adapts to the input; we train them with a Monte Carlo beta-ELBO objective. We pair FDN with an evaluation protocol that separates interpolation from extrapolation and emphasizes simple OOD sanity checks. On controlled 1D tasks and small/medium UCI-style regression benchmarks, FDN remains competitive in accuracy with strong Bayesian, ensemble, dropout, and hypernetwork baselines, while providing strongly input-dependent, shift-aware uncertainty and competitive calibration under matched parameter and update budgets. --- ### Foundational Automatic Evaluators: Scaling Multi-Task Generative Evaluator Training for Reasoning-Centric Domains Authors: Austin Xu, Xuan-Phi Nguyen, Yilun Zhou, Chien-Sheng Wu, Caiming Xiong, Shafiq Joty Published: 2025-10-20 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2510.17793v2 Finetuning specialized generative evaluators has emerged as a popular paradigm to meet the increasing demand for scalable evaluation during both training and test-time. However, recent work has largely focused on applying new methodology, such as reinforcement learning (RL), to training evaluators, shying away from large-scale, data-driven development. In this work, we focus on data scaling, curating a set of 2.5M samples spanning five unique evaluation tasks (pairwise, step-level, reference-free and reference-based verification, and single rating) and multiple domains focused on reasoning evaluation. With our data, we train Foundational Automatic Reasoning Evaluators (FARE), a family of 8B and 20B (with 3.6B active) parameter evaluators, with a simple iterative rejection-sampling supervised finetuning (SFT) approach. FARE-8B challenges larger specialized RL-trained evaluators and FARE-20B sets the new standard for open-source evaluators, surpassing specialized 70B+ evaluators. Beyond static benchmarks, we evaluate FARE in real-world tasks: As inference-time rerankers, FARE-20B achieves near-oracle performance on MATH. As verifiers in RL training, FARE improves the downstream RL-trained model performance by up to 14.1% vs. string-matching verifiers. When initialized from FARE, a continually-finetuned FARE-Code outperforms gpt-oss-20B by 65% on evaluating test-case quality. --- ### SoftMimic: Learning Compliant Whole-body Control from Examples Authors: Gabriel B. Margolis, Michelle Wang, Nolan Fey, Pulkit Agrawal Published: 2025-10-20 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2510.17792v1 We introduce SoftMimic, a framework for learning compliant whole-body control policies for humanoid robots from example motions. Imitating human motions with reinforcement learning allows humanoids to quickly learn new skills, but existing methods incentivize stiff control that aggressively corrects deviations from a reference motion, leading to brittle and unsafe behavior when the robot encounters unexpected contacts. In contrast, SoftMimic enables robots to respond compliantly to external forces while maintaining balance and posture. Our approach leverages an inverse kinematics solver to generate an augmented dataset of feasible compliant motions, which we use to train a reinforcement learning policy. By rewarding the policy for matching compliant responses rather than rigidly tracking the reference motion, SoftMimic learns to absorb disturbances and generalize to varied tasks from a single motion clip. We validate our method through simulations and real-world experiments, demonstrating safe and effective interaction with the environment. --- ### UltraCUA: A Foundation Model for Computer Use Agents with Hybrid Action Authors: Yuhao Yang, Zhen Yang, Zi-Yi Dou, Anh Nguyen, Keen You, Omar Attia, Andrew Szot, Michael Feng, Ram Ramrakhya, Alexander Toshev, Chao Huang, Yinfei Yang et al. Published: 2025-10-20 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2510.17790v3 Computer-use agents face a fundamental limitation. They rely exclusively on primitive GUI actions (click, type, scroll), creating brittle execution chains prone to cascading failures. While API-driven agents harness rich capabilities through structured interfaces and tools, computer-use agents remain constrained to low-level visual interactions. We present UltraCUA, a foundation model that transcends this limitation through hybrid action-seamlessly unifying primitive GUI operations with high-level tool execution. Our innovation rests on four critical advances. First, an automated pipeline extracts and scales tool capabilities from software documentation and code repositories. Second, a synthetic data engine produces 17,000+ verifiable tasks capturing real-world computer-use complexity. Third, comprehensive hybrid action trajectory collection incorporates both GUI primitives and strategic tool calls. Fourth, a two-stage training methodology combines supervised fine-tuning with online reinforcement learning, enabling intelligent action selection between GUI and API. Evaluation with our 7B and 32B UltraCUA models reveals transformative performance gains. On OSWorld, UltraCUA achieves 22% relative improvement while executing 11% faster than existing approaches, averagely. Cross-domain validation on WindowsAgentArena demonstrates robust generalization with 21.7% success rate, surpassing Windows-trained baselines. The hybrid action paradigm proves essential, reducing error propagation while improving execution efficiency. This work establishes a scalable paradigm bridging primitive GUI interactions and high-level tool intelligence, enabling more resilient and adaptable computer use agents for diverse environments and complex real-world tasks. --- ### Inference-Time Compute Scaling For Flow Matching Authors: Adam Stecklov, Noah El Rimawi-Fine, Mathieu Blanchette Published: 2025-10-20 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.17786v1 Allocating extra computation at inference time has recently improved sample quality in large language models and diffusion-based image generation. In parallel, Flow Matching (FM) has gained traction in language, vision, and scientific domains, but inference-time scaling methods for it remain under-explored. Concurrently, Kim et al., 2025 approach this problem but replace the linear interpolant with a non-linear variance-preserving (VP) interpolant at inference, sacrificing FM's efficient and straight sampling. Additionally, inference-time compute scaling for flow matching has only been applied to visual tasks, like image generation. We introduce novel inference-time scaling procedures for FM that preserve the linear interpolant during sampling. Evaluations of our method on image generation, and for the first time (to the best of our knowledge), unconditional protein generation, show that I) sample quality consistently improves as inference compute increases, and II) flow matching inference-time scaling can be applied to scientific domains. --- ### Botany-Bot: Digital Twin Monitoring of Occluded and Underleaf Plant Structures with Gaussian Splats Authors: Simeon Adebola, Chung Min Kim, Justin Kerr, Shuangyu Xie, Prithvi Akella, Jose Luis Susa Rincon, Eugen Solowjow, Ken Goldberg Published: 2025-10-20 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2510.17783v1 Commercial plant phenotyping systems using fixed cameras cannot perceive many plant details due to leaf occlusion. In this paper, we present Botany-Bot, a system for building detailed "annotated digital twins" of living plants using two stereo cameras, a digital turntable inside a lightbox, an industrial robot arm, and 3D segmentated Gaussian Splat models. We also present robot algorithms for manipulating leaves to take high-resolution indexable images of occluded details such as stem buds and the underside/topside of leaves. Results from experiments suggest that Botany-Bot can segment leaves with 90.8% accuracy, detect leaves with 86.2% accuracy, lift/push leaves with 77.9% accuracy, and take detailed overside/underside images with 77.3% accuracy. Code, videos, and datasets are available at https://berkeleyautomation.github.io/Botany-Bot/. --- ### SparseVILA: Decoupling Visual Sparsity for Efficient VLM Inference Authors: Samir Khaki, Junxian Guo, Jiaming Tang, Shang Yang, Yukang Chen, Konstantinos N. Plataniotis, Yao Lu, Song Han, Zhijian Liu Published: 2025-10-20 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.17777v1 Vision Language Models (VLMs) have rapidly advanced in integrating visual and textual reasoning, powering applications across high-resolution image understanding, long-video analysis, and multi-turn conversation. However, their scalability remains limited by the growing number of visual tokens that dominate inference latency. We present SparseVILA, a new paradigm for efficient VLM inference that decouples visual sparsity across the prefilling and decoding stages. SparseVILA distributes sparsity across stages by pruning redundant visual tokens during prefill and retrieving only query-relevant tokens during decoding. This decoupled design matches leading prefill pruning methods while preserving multi-turn fidelity by retaining most of the visual cache so that query-aware tokens can be retrieved at each conversation round. Built on an AWQ-optimized inference pipeline, SparseVILA achieves up to 4.0 times faster prefilling, 2.5 times faster decoding, and an overall 2.6 times end-to-end speedup on long-context video tasks -- while improving accuracy on document-understanding and reasoning tasks. By decoupling query-agnostic pruning and query-aware retrieval, SparseVILA establishes a new direction for efficient multimodal inference, offering a training-free, architecture-agnostic framework for accelerating large VLMs without sacrificing capability. --- ### Mapping Post-Training Forgetting in Language Models at Scale Authors: Jackson Harmon, Andreas Hochlehnert, Matthias Bethge, Ameya Prabhu Published: 2025-10-20 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2510.17776v1 Scaled post-training now drives many of the largest capability gains in language models (LMs), yet its effect on pretrained knowledge remains poorly understood. Not all forgetting is equal: Forgetting one fact (e.g., a U.S. president or an API call) does not "average out" by recalling another. Hence, we propose a sample-wise paradigm to measure what is forgotten and when backward transfer occurs. Our metric counts 1->0 transitions (correct before post-training, incorrect after) to quantify forgetting and 0->1 transitions to quantify backward transfer. Traditional task averages conflate these effects and obscure large changes. For multiple-choice benchmarks, we add chance-adjusted variants that subtract the expected contribution of random guessing from pre- and post-training accuracies. We apply this framework across post-training stages, model sizes, and data scales. Our large-scale analysis shows that: (1) Domain-continual pretraining induces moderate forgetting with low-to-moderate backward transfer; (2) RL/SFT post-training applied to base models and Instruction tuning yields moderate-to-large backward transfer on math and logic with overall low-to-moderate forgetting; (3) Applying RL/SFT to instruction-tuned models is sensitive on data scale: at small scales, both forgetting and backward transfer are small; at larger scales, effects are mixed and warrant further study with better controls; (4) Model merging does not reliably mitigate forgetting. Overall, our framework offers a practical yardstick for mapping how post-training alters pretrained knowledge at scale -- enabling progress towards generally capable AI systems. --- ### Towards Explainable Skin Cancer Classification: A Dual-Network Attention Model with Lesion Segmentation and Clinical Metadata Fusion Authors: Md. Enamul Atiq, Shaikh Anowarul Fattah Published: 2025-10-20 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.17773v1 Skin cancer is a life-threatening disease where early detection significantly improves patient outcomes. Automated diagnosis from dermoscopic images is challenging due to high intra-class variability and subtle inter-class differences. Many deep learning models operate as "black boxes," limiting clinical trust. In this work, we propose a dual-encoder attention-based framework that leverages both segmented lesions and clinical metadata to enhance skin lesion classification in terms of both accuracy and interpretability. A novel Deep-UNet architecture with Dual Attention Gates (DAG) and Atrous Spatial Pyramid Pooling (ASPP) is first employed to segment lesions. The classification stage uses two DenseNet201 encoders-one on the original image and another on the segmented lesion whose features are fused via multi-head cross-attention. This dual-input design guides the model to focus on salient pathological regions. In addition, a transformer-based module incorporates patient metadata (age, sex, lesion site) into the prediction. We evaluate our approach on the HAM10000 dataset and the ISIC 2018 and 2019 challenges. The proposed method achieves state-of-the-art segmentation performance and significantly improves classification accuracy and average AUC compared to baseline models. To validate our model's reliability, we use Gradient-weighted Class Activation Mapping (Grad-CAM) to generate heatmaps. These visualizations confirm that our model's predictions are based on the lesion area, unlike models that rely on spurious background features. These results demonstrate that integrating precise lesion segmentation and clinical data with attention-based fusion leads to a more accurate and interpretable skin cancer classification model. --- ### Atlas-based Manifold Representations for Interpretable Riemannian Machine Learning Authors: Ryan A. Robinett, Sophia A. Madejski, Kyle Ruark, Samantha J. Riesenfeld, Lorenzo Orecchia Published: 2025-10-20 Categories: cs.LG, stat.AP arXiv: https://arxiv.org/abs/2510.17772v1 Despite the popularity of the manifold hypothesis, current manifold-learning methods do not support machine learning directly on the latent $d$-dimensional data manifold, as they primarily aim to perform dimensionality reduction into $\mathbb{R}^D$, losing key manifold features when the embedding dimension $D$ approaches $d$. On the other hand, methods that directly learn the latent manifold as a differentiable atlas have been relatively underexplored. In this paper, we aim to give a proof of concept of the effectiveness and potential of atlas-based methods. To this end, we implement a generic data structure to maintain a differentiable atlas that enables Riemannian optimization over the manifold. We complement this with an unsupervised heuristic that learns a differentiable atlas from point cloud data. We experimentally demonstrate that this approach has advantages in terms of efficiency and accuracy in selected settings. Moreover, in a supervised classification task over the Klein bottle and in RNA velocity analysis of hematopoietic data, we showcase the improved interpretability and robustness of our approach. --- ### Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMs Authors: Zhining Liu, Ziyi Chen, Hui Liu, Chen Luo, Xianfeng Tang, Suhang Wang, Joy Zeng, Zhenwei Dai, Zhan Shi, Tianxin Wei, Benoit Dumoulin, Hanghang Tong Published: 2025-10-20 Categories: cs.AI, cs.CV arXiv: https://arxiv.org/abs/2510.17771v1 Vision-Language Models (VLMs) achieve strong results on multimodal tasks such as visual question answering, yet they can still fail even when the correct visual evidence is present. In this work, we systematically investigate whether these failures arise from not perceiving the evidence or from not leveraging it effectively. By examining layer-wise attention dynamics, we find that shallow layers focus primarily on text, while deeper layers sparsely but reliably attend to localized evidence regions. Surprisingly, VLMs often perceive the visual evidence when outputting incorrect answers, a phenomenon we term ``seeing but not believing'' that widely exists in major VLM families. Building on this, we introduce an inference-time intervention that highlights deep-layer evidence regions through selective attention-based masking. It requires no training and consistently improves accuracy across multiple families, including LLaVA, Qwen, Gemma, and InternVL. These results show that VLMs encode reliable evidence internally but under-utilize it, making such signals explicit can bridge the gap between perception and reasoning, advancing the diagnostic understanding and reliability of VLMs. --- ### Evaluating Medical LLMs by Levels of Autonomy: A Survey Moving from Benchmarks to Applications Authors: Xiao Ye, Jacob Dineen, Zhaonan Li, Zhikun Xu, Weiyu Chen, Shijie Lu, Yuxi Huang, Ming Shen, Phu Tran, Ji-Eun Irene Yum, Muhammad Ali Khan, Muhammad Umar Afzal et al. Published: 2025-10-20 Categories: cs.CL arXiv: https://arxiv.org/abs/2510.17764v1 Medical Large language models achieve strong scores on standard benchmarks; however, the transfer of those results to safe and reliable performance in clinical workflows remains a challenge. This survey reframes evaluation through a levels-of-autonomy lens (L0-L3), spanning informational tools, information transformation and aggregation, decision support, and supervised agents. We align existing benchmarks and metrics with the actions permitted at each level and their associated risks, making the evaluation targets explicit. This motivates a level-conditioned blueprint for selecting metrics, assembling evidence, and reporting claims, alongside directions that link evaluation to oversight. By centering autonomy, the survey moves the field beyond score-based claims toward credible, risk-aware evidence for real clinical use. --- ### VERA-V: Variational Inference Framework for Jailbreaking Vision-Language Models Authors: Qilin Liao, Anamika Lochab, Ruqi Zhang Published: 2025-10-20 Categories: cs.CR, cs.CL, cs.CV, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2510.17759v2 Vision-Language Models (VLMs) extend large language models with visual reasoning, but their multimodal design also introduces new, underexplored vulnerabilities. Existing multimodal red-teaming methods largely rely on brittle templates, focus on single-attack settings, and expose only a narrow subset of vulnerabilities. To address these limitations, we introduce VERA-V, a variational inference framework that recasts multimodal jailbreak discovery as learning a joint posterior distribution over paired text-image prompts. This probabilistic view enables the generation of stealthy, coupled adversarial inputs that bypass model guardrails. We train a lightweight attacker to approximate the posterior, allowing efficient sampling of diverse jailbreaks and providing distributional insights into vulnerabilities. VERA-V further integrates three complementary strategies: (i) typography-based text prompts that embed harmful cues, (ii) diffusion-based image synthesis that introduces adversarial signals, and (iii) structured distractors to fragment VLM attention. Experiments on HarmBench and HADES benchmarks show that VERA-V consistently outperforms state-of-the-art baselines on both open-source and frontier VLMs, achieving up to 53.75% higher attack success rate (ASR) over the best baseline on GPT-4o. We include the code on the project page available here: https://github.com/kxwhiowo/VERA-V --- ### Prediction of Sea Ice Velocity and Concentration in the Arctic Ocean using Physics-informed Neural Network Authors: Younghyun Koo, Maryam Rahnemoonfar Published: 2025-10-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2510.17756v1 As an increasing amount of remote sensing data becomes available in the Arctic Ocean, data-driven machine learning (ML) techniques are becoming widely used to predict sea ice velocity (SIV) and sea ice concentration (SIC). However, fully data-driven ML models have limitations in generalizability and physical consistency due to their excessive reliance on the quantity and quality of training data. In particular, as Arctic sea ice entered a new phase with thinner ice and accelerated melting, there is a possibility that an ML model trained with historical sea ice data cannot fully represent the dynamically changing sea ice conditions in the future. In this study, we develop physics-informed neural network (PINN) strategies to integrate physical knowledge of sea ice into the ML model. Based on the Hierarchical Information-sharing U-net (HIS-Unet) architecture, we incorporate the physics loss function and the activation function to produce physically plausible SIV and SIC outputs. Our PINN model outperforms the fully data-driven model in the daily predictions of SIV and SIC, even when trained with a small number of samples. The PINN approach particularly improves SIC predictions in melting and early freezing seasons and near fast-moving ice regions. --- ### Human-AI Interactions: Cognitive, Behavioral, and Emotional Impacts Authors: Celeste Riley, Omar Al-Refai, Yadira Colunga Reyes, Eman Hammad Published: 2025-10-20 Categories: cs.HC, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2510.17753v2 As stories of human-AI interactions continue to be highlighted in the news and research platforms, the challenges are becoming more pronounced, including potential risks of overreliance, cognitive offloading, social and emotional manipulation, and the nuanced degradation of human agency and judgment. This paper surveys recent research on these issues through the lens of the psychological triad: cognition, behavior, and emotion. Observations seem to suggest that while AI can substantially enhance memory, creativity, and engagement, it also introduces risks such as diminished critical thinking, skill erosion, and increased anxiety. Emotional outcomes are similarly mixed, with AI systems showing promise for support and stress reduction, but raising concerns about dependency, inappropriate attachments, and ethical oversight. This paper aims to underscore the need for responsible and context-aware AI design, highlighting gaps for longitudinal research and grounded evaluation frameworks to balance benefits with emerging human-centric risks. --- ### A Multi-Threading Kernel for Enabling Neuromorphic Edge Applications Authors: Lars Niedermeier, Vyom Shah, Jeffrey L. Krichmar Published: 2025-10-20 Categories: cs.NE, cs.AI arXiv: https://arxiv.org/abs/2510.17745v1 Spiking Neural Networks (SNNs) have sparse, event driven processing that can leverage neuromorphic applications. In this work, we introduce a multi-threading kernel that enables neuromorphic applications running at the edge, meaning they process sensory input directly and without any up-link to or dependency on a cloud service. The kernel shows speed-up gains over single thread processing by a factor of four on moderately sized SNNs and 1.7X on a Synfire network. Furthermore, it load-balances all cores available on multi-core processors, such as ARM, which run today's mobile devices and is up to 70% more energy efficient compared to statical core assignment. The present work can enable the development of edge applications that have low Size, Weight, and Power (SWaP), and can prototype the integration of neuromorphic chips. --- ### Joint Multi-Condition Representation Modelling via Matrix Factorisation for Visual Place Recognition Authors: Timur Ismagilov, Shakaiba Majeed, Michael Milford, Tan Viet Tuyen Nguyen, Sarvapali D. Ramchurn, Shoaib Ehsan Published: 2025-10-20 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.17739v1 We address multi-reference visual place recognition (VPR), where reference sets captured under varying conditions are used to improve localisation performance. While deep learning with large-scale training improves robustness, increasing data diversity and model complexity incur extensive computational cost during training and deployment. Descriptor-level fusion via voting or aggregation avoids training, but often targets multi-sensor setups or relies on heuristics with limited gains under appearance and viewpoint change. We propose a training-free, descriptor-agnostic approach that jointly models places using multiple reference descriptors via matrix decomposition into basis representations, enabling projection-based residual matching. We also introduce SotonMV, a structured benchmark for multi-viewpoint VPR. On multi-appearance data, our method improves Recall@1 by up to ~18% over single-reference and outperforms multi-reference baselines across appearance and viewpoint changes, with gains of ~5% on unstructured data, demonstrating strong generalisation while remaining lightweight. --- ### Efficient Tensor Completion Algorithms for Highly Oscillatory Operators Authors: Navjot Singh, Edgar Solomonik, Xiaoye Sherry Li, Yang Liu Published: 2025-10-20 Categories: math.NA, cs.LG arXiv: https://arxiv.org/abs/2510.17734v3 This paper presents low-complexity tensor completion algorithms and their efficient implementation to reconstruct highly oscillatory operators discretized as $n\times n$ matrices. The underlying tensor decomposition is based on the reshaping of the input matrix and its butterfly decomposition into an order $O (\log n)$ tensor. The reshaping of the input matrix into a tensor allows for representation of the butterfly decomposition as a tensor decomposition with dense tensors. This leads to efficient utilization of the existing software infrastructure for dense and sparse tensor computations. We propose two tensor completion algorithms in the butterfly format, using alternating least squares and gradient-based optimization, as well as a novel strategy that uses low-rank matrix completion to efficiently generate an initial guess for the proposed algorithms. To demonstrate the efficiency and applicability of our proposed algorithms, we perform three numerical experiments using simulated oscillatory operators in seismic applications. In these experiments, we use $O (n \log n)$ observed entries in the input matrix and demonstrate an $O(n\log^3 n)$ computational cost of the proposed algorithms, leading to a speedup of orders of magnitudes per iteration for large matrices compared to the low-rank matrix and quantized tensor-train completion. Moreover, the proposed butterfly completion algorithms, equipped with the novel initial guess generation strategy, achieve reconstruction errors that are smaller by an order of magnitude, enabling accurate recovery of the underlying structure compared to the state-of-the-art completion algorithms. --- ### Train for Truth, Keep the Skills: Binary Retrieval-Augmented Reward Mitigates Hallucinations Authors: Tong Chen, Akari Asai, Luke Zettlemoyer, Hannaneh Hajishirzi, Faeze Brahman Published: 2025-10-20 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2510.17733v1 Language models often generate factually incorrect information unsupported by their training data, a phenomenon known as extrinsic hallucination. Existing mitigation approaches often degrade performance on open-ended generation and downstream tasks, limiting their practical utility. We propose an online reinforcement learning method using a novel binary retrieval-augmented reward (RAR) to address this tradeoff. Unlike continuous reward schemes, our approach assigns a reward of one only when the model's output is entirely factually correct, and zero otherwise. We evaluate our method on Qwen3 reasoning models across diverse tasks. For open-ended generation, binary RAR achieves a 39.3% reduction in hallucination rates, substantially outperforming both supervised training and continuous-reward RL baselines. In short-form question answering, the model learns calibrated abstention, strategically outputting "I don't know" when faced with insufficient parametric knowledge. This yields 44.4% and 21.7% fewer incorrect answers on PopQA and GPQA, respectively. Crucially, these factuality gains come without performance degradation on instruction following, math, or code, whereas continuous-reward RL, despite improving factuality, induces quality regressions. --- ### Can Image-To-Video Models Simulate Pedestrian Dynamics? Authors: Aaron Appelle, Jerome P. Lynch Published: 2025-10-20 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.17731v1 Recent high-performing image-to-video (I2V) models based on variants of the diffusion transformer (DiT) have displayed remarkable inherent world-modeling capabilities by virtue of training on large scale video datasets. We investigate whether these models can generate realistic pedestrian movement patterns in crowded public scenes. Our framework conditions I2V models on keyframes extracted from pedestrian trajectory benchmarks, then evaluates their trajectory prediction performance using quantitative measures of pedestrian dynamics. --- ### Enabling Fine-Grained Operating Points for Black-Box LLMs Authors: Ege Beyazit, KL Navaneet, Prashant Mathur, Roi Blanco, Vidit Bansal, Karim Bouyarmane Published: 2025-10-20 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.17727v2 Black-box Large Language Models (LLMs) provide practical and accessible alternatives to other machine learning methods, as they require minimal labeled data and machine learning expertise to develop solutions for various decision making problems. However, for applications that need operating with constraints on specific metrics (e.g., precision $\geq$ 95%), decision making with black-box LLMs remains unfavorable, due to their low numerical output cardinalities. This results in limited control over their operating points, preventing fine-grained adjustment of their decision making behavior. In this paper, we study using black-box LLMs as classifiers, focusing on efficiently improving their operational granularity without performance loss. Specifically, we first investigate the reasons behind their low-cardinality numerical outputs and show that they are biased towards generating rounded but informative verbalized probabilities. Then, we experiment with standard prompt engineering, uncertainty estimation and confidence elicitation techniques, and observe that they do not effectively improve operational granularity without sacrificing performance or increasing inference cost. Finally, we propose efficient approaches to significantly increase the number and diversity of available operating points. Our proposed approaches provide finer-grained operating points and achieve comparable to or better performance than the benchmark methods across 11 datasets and 3 LLMs. --- ### AcademicEval: Live Long-Context LLM Benchmark Authors: Haozhen Zhang, Tao Feng, Pengrui Han, Jiaxuan You Published: 2025-10-20 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2510.17725v1 Large Language Models (LLMs) have recently achieved remarkable performance in long-context understanding. However, current long-context LLM benchmarks are limited by rigid context length, labor-intensive annotation, and the pressing challenge of label leakage issues during LLM training. Therefore, we propose \textsc{AcademicEval}, a live benchmark for evaluating LLMs over long-context generation tasks. \textsc{AcademicEval} adopts papers on arXiv to introduce several academic writing tasks with long-context inputs, \textit{i.e.}, \textsc{Title}, \textsc{Abstract}, \textsc{Introduction}, and \textsc{Related Work}, which cover a wide range of abstraction levels and require no manual labeling. Moreover, \textsc{AcademicEval} integrates high-quality and expert-curated few-shot demonstrations from a collected co-author graph to enable flexible context length. Especially, \textsc{AcademicEval} features an efficient live evaluation, ensuring no label leakage. We conduct a holistic evaluation on \textsc{AcademicEval}, and the results illustrate that LLMs perform poorly on tasks with hierarchical abstraction levels and tend to struggle with long few-shot demonstrations, highlighting the challenge of our benchmark. Through experimental analysis, we also reveal some insights for enhancing LLMs' long-context modeling capabilities. Code is available at https://github.com/ulab-uiuc/AcademicEval --- ### Signature Forgery Detection: Improving Cross-Dataset Generalization Authors: Matheus Ramos Parracho Published: 2025-10-20 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.17724v1 Automated signature verification is a critical biometric technique used in banking, identity authentication, and legal documentation. Despite the notable progress achieved by deep learning methods, most approaches in offline signature verification still struggle to generalize across datasets, as variations in handwriting styles and acquisition protocols often degrade performance. This study investigates feature learning strategies for signature forgery detection, focusing on improving cross-dataset generalization -- that is, model robustness when trained on one dataset and tested on another. Using three public benchmarks -- CEDAR, ICDAR, and GPDS Synthetic -- two experimental pipelines were developed: one based on raw signature images and another employing a preprocessing method referred to as shell preprocessing. Several behavioral patterns were identified and analyzed; however, no definitive superiority between the two approaches was established. The results show that the raw-image model achieved higher performance across benchmarks, while the shell-based model demonstrated promising potential for future refinement toward robust, cross-domain signature verification. --- ### MT-Video-Bench: A Holistic Video Understanding Benchmark for Evaluating Multimodal LLMs in Multi-Turn Dialogues Authors: Yaning Pan, Qianqian Xie, Guohui Zhang, Zekun Wang, Yongqian Wen, Yuanxing Zhang, Haoxuan Hu, Zhiyu Pan, Yibing Huang, Zhidong Gan, Yonghong Lin, An Ping et al. Published: 2025-10-20 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.17722v2 The recent development of Multimodal Large Language Models (MLLMs) has significantly advanced AI's ability to understand visual modalities. However, existing evaluation benchmarks remain limited to single-turn question answering, overlooking the complexity of multi-turn dialogues in real-world scenarios. To bridge this gap, we introduce MT-Video-Bench, a holistic video understanding benchmark for evaluating MLLMs in multi-turn dialogues. Specifically, our MT-Video-Bench mainly assesses 6 core competencies that focus on perceptivity and interactivity, encompassing 1,000 meticulously curated multi-turn dialogues from diverse domains. These capabilities are rigorously aligned with real-world applications, such as interactive sports analysis and multi-turn video-based intelligent tutoring. With MT-Video-Bench, we extensively evaluate various state-of-the-art open-source and closed-source MLLMs, revealing their significant performance discrepancies and limitations in handling multi-turn video dialogues. The benchmark will be publicly available to foster future research. --- ### PANER: A Paraphrase-Augmented Framework for Low-Resource Named Entity Recognition Authors: Nanda Kumar Rengarajan, Jun Yan, Chun Wang Published: 2025-10-20 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2510.17720v1 Named Entity Recognition (NER) is a critical task that requires substantial annotated data, making it challenging in low-resource scenarios where label acquisition is expensive. While zero-shot and instruction-tuned approaches have made progress, they often fail to generalize to domain-specific entities and do not effectively utilize limited available data. We present a lightweight few-shot NER framework that addresses these challenges through two key innovations: (1) a new instruction tuning template with a simplified output format that combines principles from prior IT approaches to leverage the large context window of recent state-of-the-art LLMs; (2) introducing a strategic data augmentation technique that preserves entity information while paraphrasing the surrounding context, thereby expanding our training data without compromising semantic relationships. Experiments on benchmark datasets show that our method achieves performance comparable to state-of-the-art models on few-shot and zero-shot tasks, with our few-shot approach attaining an average F1 score of 80.1 on the CrossNER datasets. Models trained with our paraphrasing approach show consistent improvements in F1 scores of up to 17 points over baseline versions, offering a promising solution for groups with limited NER training data and compute power. --- ### OmniRaindrop: A Comprehensive Benchmark for Dual-Focused Raindrop Removal under Day and Night Scenes Authors: Zhiqiang Teng, Tingting Chen, Beibei Lin, Zifeng Yuan, Xuanyi Li, Xuanyu Zhang, Shunli Zhang Published: 2025-10-20 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.17719v3 Adherent raindrops on camera lenses are a pervasive source of real-world image degradation, undermining outdoor visual perception for downstream applications such as robotics and 3D scene reconstruction. However, most existing raindrop datasets and methods target only background-focused 2D images, overlooking the equally common raindrop-focused case and the impact of raindrops on 3D scene reconstruction. Moreover, they are almost exclusively daytime, lacking the complex artificial lighting and reflections of nighttime raindrop scenes. To address these issues, we introduce OmniRaindrop, a real-world dataset and benchmark that jointly evaluates 2D raindrop removal and 3D scene reconstruction across raindrop-focused and background-focused captures, under both daytime and nighttime conditions. OmniRaindrop provides spatially aligned raindrop-focused, background-focused, and rain-free ground-truth image sets: a 2D subset of 15,186 aligned pairs/triplets and a 3D subset of 40 real-world scenes evenly split between day and night, totaling 3,822 multi-view images. On top of this data, the benchmark traces how raindrop degradation propagates from 2D restoration to 3D reconstruction by testing representative restoration methods and their effects on pose estimation, point cloud initialization, and 3DGS optimization. Comprehensive experiments reveal that raindrop-focused scenes are substantially harder than background-focused ones, and that errors in 2D restoration and pose/point cloud initialization cascade into degraded 3D reconstruction, pointing to clear directions for more robust 2D restoration and 3DGS design. Building on the dataset, we have organized two consecutive NTIRE challenges (2025 and 2026) on Day and Night Raindrop Removal for Dual-Focused Images. --- ### Automatic Classification of Circulating Blood Cell Clusters based on Multi-channel Flow Cytometry Imaging Authors: Suqiang Ma, Subhadeep Sengupta, Yao Lee, Beikang Gu, Xianyan Chen, Xianqiao Wang, Yang Liu, Mengjia Xu, Galit H. Frydman, He Li Published: 2025-10-20 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.17716v1 Circulating blood cell clusters (CCCs) containing red blood cells (RBCs), white blood cells(WBCs), and platelets are significant biomarkers linked to conditions like thrombosis, infection, and inflammation. Flow cytometry, paired with fluorescence staining, is commonly used to analyze these cell clusters, revealing cell morphology and protein profiles. While computational approaches based on machine learning have advanced the automatic analysis of single-cell flow cytometry images, there is a lack of effort to build tools to automatically analyze images containing CCCs. Unlike single cells, cell clusters often exhibit irregular shapes and sizes. In addition, these cell clusters often consist of heterogeneous cell types, which require multi-channel staining to identify the specific cell types within the clusters. This study introduces a new computational framework for analyzing CCC images and identifying cell types within clusters. Our framework uses a two-step analysis strategy. First, it categorizes images into cell cluster and non-cluster groups by fine-tuning the You Only Look Once(YOLOv11) model, which outperforms traditional convolutional neural networks (CNNs), Vision Transformers (ViT). Then, it identifies cell types by overlaying cluster contours with regions from multi-channel fluorescence stains, enhancing accuracy despite cell debris and staining artifacts. This approach achieved over 95% accuracy in both cluster classification and phenotype identification. In summary, our automated framework effectively analyzes CCC images from flow cytometry, leveraging both bright-field and fluorescence data. Initially tested on blood cells, it holds potential for broader applications, such as analyzing immune and tumor cell clusters, supporting cellular research across various diseases. --- ### QueST: Incentivizing LLMs to Generate Difficult Problems Authors: Hanxu Hu, Xingxing Zhang, Jannis Vamvas, Rico Sennrich, Furu Wei Published: 2025-10-20 Categories: cs.CL arXiv: https://arxiv.org/abs/2510.17715v1 Large Language Models have achieved strong performance on reasoning tasks, solving competition-level coding and math problems. However, their scalability is limited by human-labeled datasets and the lack of large-scale, challenging coding problem training data. Existing competitive coding datasets contain only thousands to tens of thousands of problems. Previous synthetic data generation methods rely on either augmenting existing instruction datasets or selecting challenging problems from human-labeled data. In this paper, we propose QueST, a novel framework which combines difficulty-aware graph sampling and difficulty-aware rejection fine-tuning that directly optimizes specialized generators to create challenging coding problems. Our trained generators demonstrate superior capability compared to even GPT-4o at creating challenging problems that benefit downstream performance. We leverage QueST to generate large-scale synthetic coding problems, which we then use to distill from strong teacher models with long chain-of-thought or to conduct reinforcement learning for smaller models, proving effective in both scenarios. Our distillation experiments demonstrate significant performance gains. Specifically, after fine-tuning Qwen3-8B-base on 100K difficult problems generated by QueST, we surpass the performance of the original Qwen3-8B on LiveCodeBench. With an additional 112K examples (i.e., 28K human-written problems paired with multiple synthetic solutions), our 8B model matches the performance of the much larger DeepSeek-R1-671B. These findings indicate that generating complex problems via QueST offers an effective and scalable approach to advancing the frontiers of competitive coding and reasoning for large language models. --- ### The Marked Edge Walk: A Novel MCMC Algorithm for Sampling of Graph Partitions Authors: Atticus McWhorter, Daryl DeFord Published: 2025-10-20 Categories: cs.DS, cs.LG, physics.soc-ph arXiv: https://arxiv.org/abs/2510.17714v3 Novel Markov Chain Monte Carlo (MCMC) methods have enabled the generation of large ensembles of redistricting plans modeled as a graph partitioning problem. However, existing algorithms such as Reversible Recombination (RevReCom) and Metropolized Forest Recombination (MFR) have strong preferences for distributions related to the spanning tree measure. In this paper we introduce the Marked Edge Walk (MEW), a novel Markov chain proposal for sampling from the space of graph partitions. The walk operates on the space of spanning trees with marked edges, allowing for calculable transition probabilities for use in the Metropolis-Hastings algorithm. Empirical results on real-world dual graphs show convergence under a broad class of target distributions less constrained by spanning tree counts, including policy-based distributions, such as competitiveness on New Hampshire that are independent of spanning trees, and compactness and partisan symmetry distributions on New Hampshire and Texas that, while related to spanning trees, can now be properly targeted with a smaller degree of spanning tree bias, which represents an advancement in flexible ensemble generation. --- ### Bi-Level Reinforcement Learning Pathway for Sim-to-Real Optimality Authors: Akhil S Anand, Shambhuraj Sawant, Paavo Parmas, Jasper Hoffmann, Dirk Reinhardt, Sebastien Gros Published: 2025-10-20 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2510.17709v2 Training Reinforcement Learning (RL) policies using simulation models before deployment in real-world environments is a common strategy when real-world interaction is expensive. This approach is used in sim-to-real RL and in dyna-style model-based RL. A key limitation of this approach is that the policies trained in simulation often perform poorly in the real world due to discrepancies between the simulation model and the real-world environment, referred to as the sim-to-real gap. This gap reflects the objective mismatch: simulation models are typically constructed for predictive accuracy, whereas policies are trained to maximize task performance. Since the policy learned in simulation is implicitly defined by the simulation parameters, understanding the sensitivity of the learned policy to these parameters enables gradient-based adaptation of the simulation model to improve real-world policy performance. Motivated by this, we derive the sensitivity of locally converged policies trained with Stochastic Policy Gradient (SPG) methods in an actor-critic setting, which is the most widely used approach in RL. Based on this sensitivity analysis, we formulate a bi-level RL approach that can address the objective mismatch problem by learning simulation parameters using gradients of real-world policy performance, thereby directly coupling simulation model adaptation with policy performance. We provide a thorough convergence analysis of the proposed bi-level RL approach and illustrate the concept through a proof-of-concept bi-level PPO algorithm. --- ### Contextual Attention Modulation: Towards Efficient Multi-Task Adaptation in Large Language Models Authors: Dayan Pan, Zhaoyang Fu, Jingyuan Wang, Xiao Han, Yue Zhu, Xiangyu Zhao Published: 2025-10-20 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2510.17705v1 Large Language Models (LLMs) possess remarkable generalization capabilities but struggle with multi-task adaptation, particularly in balancing knowledge retention with task-specific specialization. Conventional fine-tuning methods suffer from catastrophic forgetting and substantial resource consumption, while existing parameter-efficient methods perform suboptimally in complex multi-task scenarios. To address this, we propose Contextual Attention Modulation (CAM), a novel mechanism that dynamically modulates the representations of self-attention modules in LLMs. CAM enhances task-specific features while preserving general knowledge, thereby facilitating more effective and efficient adaptation. For effective multi-task adaptation, CAM is integrated into our Hybrid Contextual Attention Modulation (HyCAM) framework, which combines a shared, full-parameter CAM module with multiple specialized, lightweight CAM modules, enhanced by a dynamic routing strategy for adaptive knowledge fusion. Extensive experiments on heterogeneous tasks, including question answering, code generation, and logical reasoning, demonstrate that our approach significantly outperforms existing approaches, achieving an average performance improvement of 3.65%. The implemented code and data are available to ease reproducibility at https://github.com/Applied-Machine-Learning-Lab/HyCAM. --- ### Enhancing Cross-Patient Generalization in AI-Based Parkinson s Disease Detection Authors: Mhd Adnan Albani, Riad Sonbol Published: 2025-10-20 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2510.17703v2 Parkinson's disease (PD) is a neurodegenerative disease affecting about 1% of people over the age of 60, causing motor impairments that impede hand coordination activities such as writing and drawing. Many approaches have tried to support early detection of Parkinson's disease based on hand-drawn images; however, we identified two major limitations in the related works: (1) the lack of sufficient datasets, (2) the robustness when dealing with unseen patient data. In this paper, we propose a new approach to detect Parkinson's disease that consists of two stages: The first stage classifies based on their drawing type(circle, meander, spiral), and the second stage extracts the required features from the images and detects Parkinson's disease. We overcame the previous two limitations by applying a chunking strategy where we divide each image into 2x2 chunks. Each chunk is processed separately when extracting features and recognizing Parkinson's disease indicators. To make the final classification, an ensemble method is used to merge the decisions made from each chunk. Our evaluation shows that our proposed approach outperforms the top performing state-of-the-art approaches, in particular on unseen patients. On the NewHandPD dataset our approach, it achieved 97.08% accuracy for seen patients and 94.91% for unseen patients, our proposed approach maintained a gap of only 2.17 percentage points, compared to the 4.76-point drop observed in prior work. --- ### Elastic ViTs from Pretrained Models without Retraining Authors: Walter Simoncini, Michael Dorkenwald, Tijmen Blankevoort, Cees G. M. Snoek, Yuki M. Asano Published: 2025-10-20 Categories: cs.CV arXiv: https://arxiv.org/abs/2510.17700v2 Vision foundation models achieve remarkable performance but are only available in a limited set of pre-determined sizes, forcing sub-optimal deployment choices under real-world constraints. We introduce SnapViT: Single-shot network approximation for pruned Vision Transformers, a new post-pretraining structured pruning method that enables elastic inference across a continuum of compute budgets. Our approach efficiently combines gradient information with cross-network structure correlations, approximated via an evolutionary algorithm, does not require labeled data, generalizes to models without a classification head, and is retraining-free. Experiments on DINO, SigLIPv2, DeIT, and AugReg models demonstrate superior performance over state-of-the-art methods across various sparsities, requiring less than five minutes on a single A100 GPU to generate elastic models that can be adjusted to any computational budget. Our key contributions include an efficient pruning strategy for pretrained Vision Transformers, a novel evolutionary approximation of Hessian off-diagonal structures, and a self-supervised importance scoring mechanism that maintains strong performance without requiring retraining or labels. Code and pruned models are available at: https://elastic.ashita.nl/ --- ### GAS: Improving Discretization of Diffusion ODEs via Generalized Adversarial Solver Authors: Aleksandr Oganov, Ilya Bykov, Eva Neudachina, Mishan Aliev, Alexander Tolmachev, Alexander Sidorov, Aleksandr Zuev, Andrey Okhotin, Denis Rakitin, Aibek Alanov Published: 2025-10-20 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2510.17699v2 While diffusion models achieve state-of-the-art generation quality, they still suffer from computationally expensive sampling. Recent works address this issue with gradient-based optimization methods that distill a few-step ODE diffusion solver from the full sampling process, reducing the number of function evaluations from dozens to just a few. However, these approaches often rely on intricate training techniques and do not explicitly focus on preserving fine-grained details. In this paper, we introduce the Generalized Solver: a simple parameterization of the ODE sampler that does not require additional training tricks and improves quality over existing approaches. We further combine the original distillation loss with adversarial training, which mitigates artifacts and enhances detail fidelity. We call the resulting method the Generalized Adversarial Solver and demonstrate its superior performance compared to existing solver training methods under similar resource constraints. Code is available at https://github.com/3145tttt/GAS. --- ### Towards Mining Effective Pedagogical Strategies from Learner-LLM Educational Dialogues Authors: Liqun He, Manolis Mavrikis, Mutlu Cukurova Published: 2025-10-20 Categories: cs.CL arXiv: https://arxiv.org/abs/2510.17698v1 Dialogue plays a crucial role in educational settings, yet existing evaluation methods for educational applications of large language models (LLMs) primarily focus on technical performance or learning outcomes, often neglecting attention to learner-LLM interactions. To narrow this gap, this AIED Doctoral Consortium paper presents an ongoing study employing a dialogue analysis approach to identify effective pedagogical strategies from learner-LLM dialogues. The proposed approach involves dialogue data collection, dialogue act (DA) annotation, DA pattern mining, and predictive model building. Early insights are outlined as an initial step toward future research. The work underscores the need to evaluate LLM-based educational applications by focusing on dialogue dynamics and pedagogical strategies. --- ### A Principle of Targeted Intervention for Multi-Agent Reinforcement Learning Authors: Anjie Liu, Jianhong Wang, Samuel Kaski, Jun Wang, Mengyue Yang Published: 2025-10-20 Categories: cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2510.17697v4 Steering cooperative multi-agent reinforcement learning (MARL) towards desired outcomes is challenging, particularly when the global guidance from a human on the whole multi-agent system is impractical in a large-scale MARL. On the other hand, designing external mechanisms (e.g., intrinsic rewards and human feedback) to coordinate agents mostly relies on empirical studies, lacking a easy-to-use research tool. In this work, we employ multi-agent influence diagrams (MAIDs) as a graphical framework to address the above issues. First, we introduce the concept of MARL interaction paradigms (orthogonal to MARL learning paradigms), using MAIDs to analyze and visualize both unguided self-organization and global guidance mechanisms in MARL. Then, we design a new MARL interaction paradigm, referred to as the targeted intervention paradigm that is applied to only a single targeted agent, so the problem of global guidance can be mitigated. In implementation, we introduce a causal inference technique, referred to as Pre-Strategy Intervention (PSI), to realize the targeted intervention paradigm. Since MAIDs can be regarded as a special class of causal diagrams, a composite desired outcome that integrates the primary task goal and an additional desired outcome can be achieved by maximizing the corresponding causal effect through the PSI. Moreover, the bundled relevance graph analysis of MAIDs provides a tool to identify whether an MARL learning paradigm is workable under the design of an MARL interaction paradigm. In experiments, we demonstrate the effectiveness of our proposed targeted intervention, and verify the result of relevance graph analysis. --- ### Efficient Algorithms for Mitigating Uncertainty and Risk in Reinforcement Learning Authors: Xihong Su Published: 2025-10-20 Categories: cs.LG arXiv: https://arxiv.org/abs/2510.17690v1 This dissertation makes three main contributions. First, We identify a new connection between policy gradient and dynamic programming in MMDPs and propose the Coordinate Ascent Dynamic Programming (CADP) algorithm to compute a Markov policy that maximizes the discounted return averaged over the uncertain models. CADP adjusts model weights iteratively to guarantee monotone policy improvements to a local maximum. Second, We establish sufficient and necessary conditions for the exponential ERM Bellman operator to be a contraction and prove the existence of stationary deterministic optimal policies for ERM-TRC and EVaR-TRC. We also propose exponential value iteration, policy iteration, and linear programming algorithms for computing optimal stationary policies for ERM-TRC and EVaR-TRC. Third, We propose model-free Q-learning algorithms for computing policies with risk-averse objectives: ERM-TRC and EVaR-TRC. The challenge is that Q-learning ERM Bellman may not be a contraction. Instead, we use the monotonicity of Q-learning ERM Bellman operators to derive a rigorous proof that the ERM-TRC and the EVaR-TRC Q-learning algorithms converge to the optimal risk-averse value functions. The proposed Q-learning algorithms compute the optimal stationary policy for ERM-TRC and EVaR-TRC. --- ### Quantum Synthetic Data Generation for Industrial Bioprocess Monitoring Authors: Shawn M. Gibford, Mohammad Reza Boskabadi, Christopher J. Savoie, Seyed Soheil Mansouri Published: 2025-10-20 Categories: cs.ET, cs.LG arXiv: https://arxiv.org/abs/2510.17688v1 Data scarcity and sparsity in bio-manufacturing poses challenges for accurate model development, process monitoring, and optimization. We aim to replicate and capture the complex dynamics of industrial bioprocesses by proposing the use of a Quantum Wasserstein Generative Adversarial Network with Gradient Penalty (QWGAN-GP) to generate synthetic time series data for industrially relevant processes. The generator within our GAN is comprised of a Parameterized Quantum Circuit (PQC). This methodology offers potential advantages in process monitoring, modeling, forecasting, and optimization, enabling more efficient bioprocess management by reducing the dependence on scarce experimental data. Our results demonstrate acceptable performance in capturing the temporal dynamics of real bioprocess data. We focus on Optical Density, a key measurement for Dry Biomass estimation. The data generated showed high fidelity to the actual historical experimental data. This intersection of quantum computing and machine learning has opened new frontiers in data analysis and generation, particularly in computationally intensive fields, for use cases such as increasing prediction accuracy for soft sensor design or for use in predictive control. --- ### CrossGuard: Safeguarding MLLMs against Joint-Modal Implicit Malicious Attacks Authors: Xu Zhang, Hao Li, Zhichao Lu Published: 2025-10-20 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2510.17687v2 Multimodal Large Language Models (MLLMs) achieve strong reasoning and perception capabilities but are increasingly vulnerable to jailbreak attacks. While existing work focuses on explicit attacks, where malicious content resides in a single modality, recent studies reveal implicit attacks, in which benign text and image inputs jointly express unsafe intent. Such joint-modal threats are difficult to detect and remain underexplored, largely due to the scarcity of high-quality implicit data. We propose ImpForge, an automated red-teaming pipeline that leverages reinforcement learning with tailored reward modules to generate diverse implicit samples across 14 domains. Building on this dataset, we further develop CrossGuard, an intent-aware safeguard providing robust and comprehensive defense against both explicit and implicit threats. Extensive experiments across safe and unsafe benchmarks, implicit and explicit attacks, and multiple out-of-domain settings demonstrate that CrossGuard significantly outperforms existing defenses, including advanced MLLMs and guardrails, achieving stronger security while maintaining high utility. This offers a balanced and practical solution for enhancing MLLM robustness against real-world multimodal threats. Our code is released: https://github.com/ZhangXu0963/CrossGuard. ---