机器学习学术速递[10.28]

点击阅读原文访问arxivdaily.com，涵盖CS|物理|数学|经济|统计|金融|生物|电气领域，更有搜索、收藏等功能！

cs.LG 方向，今日共计356篇

大模型相关(36篇)

【1】AutoStreamPipe: LLM Assisted Automatic Generation of Data Stream Processing Pipelines
标题：AutoStreamPipe：LLM辅助的数据流处理流水线自动生成
链接：https://arxiv.org/abs/2510.23408

作者：Abolfazl Younesi, Zahra Najafabadi Samani, Thomas Fahringer
备注：Under review
摘要：Data pipelines are essential in stream processing as they enable the efficient collection, processing, and delivery of real-time data, supporting rapid data analysis. In this paper, we present AutoStreamPipe, a novel framework that employs Large Language Models (LLMs) to automate the design, generation, and deployment of stream processing pipelines. AutoStreamPipe bridges the semantic gap between high-level user intent and platform-specific implementations across distributed stream processing systems for structured multi-agent reasoning by integrating a Hypergraph of Thoughts (HGoT) as an extended version of GoT. AutoStreamPipe combines resilient execution strategies, advanced query analysis, and HGoT to deliver pipelines with good accuracy. Experimental evaluations on diverse pipelines demonstrate that AutoStreamPipe significantly reduces development time (x6.3) and error rates (x5.19), as measured by a novel Error-Free Score (EFS), compared to LLM code-generation methods.

【2】Increasing LLM Coding Capabilities through Diverse Synthetic Coding Tasks
标题：通过多样化的合成编码任务提高LLM编码能力
链接：https://arxiv.org/abs/2510.23208

作者：Amal Abed, Ivan Lukic, Jörg K.H. Franke, Frank Hutter
备注：Presented at the 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop: The 4th Deep Learning for Code Workshop (DL4C)
摘要：Large language models (LLMs) have shown impressive promise in code generation, yet their progress remains limited by the shortage of large-scale datasets that are both diverse and well-aligned with human reasoning. Most existing resources pair problems with solutions, but omit the intermediate thought process that guides coding. To close this gap, we present a scalable synthetic data generation pipeline that produces nearly 800k instruction-reasoning-code-test quadruplets. Each sample combines a task, a step-by-step reasoning trace, a working solution, and executable tests, enabling models to learn not just the what but also the how of problem solving. Our pipeline combines four key components: curated contest problems, web-mined content filtered by relevance classifiers, data expansion guided by reasoning patterns, and multi-stage execution-based validation. A genetic mutation algorithm further increases task diversity while maintaining consistency between reasoning traces and code implementations. Our key finding is that fine-tuning LLMs on this dataset yields consistent improvements on coding benchmarks. Beyond raw accuracy, reasoning-aware data can substitute for model scaling, generalize across architectures, and outperform leading open-source alternatives under identical sample budgets. Our work establishes reasoning-centered synthetic data generation as an efficient approach for advancing coding capabilities in LLMs. We publish our dataset and generation pipeline to facilitate further research.

【3】LLM Meets Diffusion: A Hybrid Framework for Crystal Material Generation
标题：LLM满足扩散：晶体材料生成的混合框架
链接：https://arxiv.org/abs/2510.23040

作者：Subhojyoti Khastagir, Kishalay Das, Pawan Goyal, Seung-Cheol Lee, Satadeep Bhattacharjee, Niloy Ganguly
备注：NeurIPS 2025
摘要：Recent advances in generative modeling have shown significant promise in designing novel periodic crystal structures. Existing approaches typically rely on either large language models (LLMs) or equivariant denoising models, each with complementary strengths: LLMs excel at handling discrete atomic types but often struggle with continuous features such as atomic positions and lattice parameters, while denoising models are effective at modeling continuous variables but encounter difficulties in generating accurate atomic compositions. To bridge this gap, we propose CrysLLMGen, a hybrid framework that integrates an LLM with a diffusion model to leverage their complementary strengths for crystal material generation. During sampling, CrysLLMGen first employs a fine-tuned LLM to produce an intermediate representation of atom types, atomic coordinates, and lattice structure. While retaining the predicted atom types, it passes the atomic coordinates and lattice structure to a pre-trained equivariant diffusion model for refinement. Our framework outperforms state-of-the-art generative models across several benchmark tasks and datasets. Specifically, CrysLLMGen not only achieves a balanced performance in terms of structural and compositional validity but also generates more stable and novel materials compared to LLM-based and denoisingbased models Furthermore, CrysLLMGen exhibits strong conditional generation capabilities, effectively producing materials that satisfy user-defined constraints. Code is available at https://github.com/kdmsit/crysllmgen

【4】Incentivizing Agentic Reasoning in LLM Judges via Tool-Integrated Reinforcement Learning
标题：通过工具集成强化学习激励LLM评委的抽象推理
链接：https://arxiv.org/abs/2510.23038

作者：Ran Xu, Jingjing Chen, Jiayu Ye, Yu Wu, Jun Yan, Carl Yang, Hongkun Yu
备注：Work in Progress
摘要：Large Language Models (LLMs) are widely used as judges to evaluate response quality, providing a scalable alternative to human evaluation. However, most LLM judges operate solely on intrinsic text-based reasoning, limiting their ability to verify complex constraints or perform accurate computation. Motivated by the success of tool-integrated reasoning (TIR) in numerous tasks, we propose TIR-Judge, an end-to-end RL framework for training LLM judges that integrates a code executor for precise evaluation. TIR-Judge is built on three principles: (i) diverse training across verifiable and non-verifiable domains, (ii) flexible judgment formats (pointwise, pairwise, listwise), and (iii) iterative RL that bootstraps directly from the initial model without distillation. On seven public benchmarks, TIR-Judge surpasses strong reasoning-based judges by up to 6.4% (pointwise) and 7.7% (pairwise), and achieves listwise performance comparable to Claude-Opus-4 despite having only 8B parameters. Remarkably, TIR-Judge-Zero - trained entirely without distilled judge trajectories, matches the performance of distilled variants, demonstrating that tool-augmented judges can self-evolve through iterative reinforcement learning.

【5】Can Language Models Compose Skills In-Context?
标题：语言模型可以在上下文中编写技能吗？
链接：https://arxiv.org/abs/2510.22993

作者：Zidong Liu, Zhuoyan Xu, Zhenmei Shi, Yingyu Liang
摘要：Composing basic skills from simple tasks to accomplish composite tasks is crucial for modern intelligent systems. We investigate the in-context composition ability of language models to perform composite tasks that combine basic skills demonstrated in in-context examples. This is more challenging than the standard setting, where skills and their composition can be learned in training. We conduct systematic experiments on various representative open-source language models, utilizing linguistic and logical tasks designed to probe composition abilities. The results reveal that simple task examples can have a surprising negative impact on the performance, because the models generally struggle to recognize and assemble the skills correctly, even with Chain-of-Thought examples. Theoretical analysis further shows that it is crucial to align examples with the corresponding steps in the composition. This inspires a method for the probing tasks, whose improved performance provides positive support for our insights.

【6】The Reasoning Trap: How Enhancing LLM Reasoning Amplifies Tool Hallucination
标题：推理陷阱：增强LLM推理如何放大工具幻觉
链接：https://arxiv.org/abs/2510.22977

作者：Chenlong Yin, Zeyang Sha, Shiwen Cui, Changhua Meng
备注：18 pages, 5 figures
摘要：Enhancing the reasoning capabilities of Large Language Models (LLMs) is a key strategy for building Agents that "think then act." However, recent observations, like OpenAI's o3, suggest a paradox: stronger reasoning often coincides with increased hallucination, yet no prior work has systematically examined whether reasoning enhancement itself causes tool hallucination. To address this gap, we pose the central question: Does strengthening reasoning increase tool hallucination? To answer this, we introduce SimpleToolHalluBench, a diagnostic benchmark measuring tool hallucination in two failure modes: (i) no tool available, and (ii) only distractor tools available. Through controlled experiments, we establish three key findings. First, we demonstrate a causal relationship: progressively enhancing reasoning through RL increases tool hallucination proportionally with task performance gains. Second, this effect transcends overfitting - training on non-tool tasks (e.g., mathematics) still amplifies subsequent tool hallucination. Third, the effect is method-agnostic, appearing when reasoning is instilled via supervised fine-tuning and when it is merely elicited at inference by switching from direct answers to step-by-step thinking. We also evaluate mitigation strategies including Prompt Engineering and Direct Preference Optimization (DPO), revealing a fundamental reliability-capability trade-off: reducing hallucination consistently degrades utility. Mechanistically, Reasoning RL disproportionately collapses tool-reliability-related representations, and hallucinations surface as amplified divergences concentrated in late-layer residual streams. These findings reveal that current reasoning enhancement methods inherently amplify tool hallucination, highlighting the need for new training objectives that jointly optimize for capability and reliability.

【7】Robust Uncertainty Quantification for Self-Evolving Large Language Models via Continual Domain Pretraining
标题：通过连续领域预训练进行自进化大型语言模型的鲁棒不确定性量化
链接：https://arxiv.org/abs/2510.22931

作者：Xiaofan Zhou, Lu Cheng
摘要：Continual Learning (CL) is essential for enabling self-evolving large language models (LLMs) to adapt and remain effective amid rapid knowledge growth. Yet, despite its importance, little attention has been given to establishing statistical reliability guarantees for LLMs under CL, particularly in the setting of continual domain pretraining (CDP). Conformal Prediction (CP) has shown promise in offering correctness guarantees for LLMs, but it faces major challenges in CDP: testing data often stems from unknown or shifting domain distributions, under which CP may no longer provide valid guarantees. Moreover, when high coverage is required, CP can yield excessively large prediction sets for unanswerable queries, reducing informativeness. To address these challenges, we introduce an adaptive rejection and non-exchangeable CP framework. Our method first estimates the distribution of questions across domains in the test set using transformer-based clustering, then reweights or resamples the calibration data accordingly. Building on this, adaptive rejection CP allows the LLM to selectively abstain from answering when its confidence or competence shifts significantly. Extensive experiments demonstrate that our framework enhances both the effectiveness and reliability of CP under CDP scenarios. Our code is available at: https://anonymous.4open.science/r/CPCL-8C12/

【8】Simple Denoising Diffusion Language Models
标题：简单去噪扩散语言模型
链接：https://arxiv.org/abs/2510.22926

作者：Huaisheng Zhu, Zhengyu Chen, Shijie Zhou, Zhihui Xie, Yige Yuan, Zhimeng Guo, Siyuan Xu, Hangfan Zhang, Vasant Honavar, Teng Xiao
摘要：Diffusion models have recently been extended to language generation through Masked Diffusion Language Models (MDLMs), which achieve performance competitive with strong autoregressive models. However, MDLMs tend to degrade in the few-step regime and cannot directly adopt existing few-step distillation methods designed for continuous diffusion models, as they lack the intrinsic property of mapping from noise to data. Recent Uniform-state Diffusion Models (USDMs), initialized from a uniform prior, alleviate some limitations but still suffer from complex loss formulations that hinder scalability. In this work, we propose a simplified denoising-based loss for USDMs that optimizes only noise-replaced tokens, stabilizing training and matching ELBO-level performance. Furthermore, by framing denoising as self-supervised learning, we introduce a simple modification to our denoising loss with contrastive-inspired negative gradients, which is practical and yield additional improvements in generation quality.

【9】Interpreting and Mitigating Unwanted Uncertainty in LLMs
标题：解释和缓解法学硕士中不必要的不确定性
链接：https://arxiv.org/abs/2510.22866

作者：Tiasa Singha Roy, Ayush Rajesh Jhaveri, Ilias Triantafyllopoulos
摘要：Despite their impressive capabilities, Large Language Models (LLMs) exhibit unwanted uncertainty, a phenomenon where a model changes a previously correct answer into an incorrect one when re-prompted. This behavior undermines trust and poses serious risks in high-stakes domains. In this work, we investigate the mechanisms that drive this phenomenon. We adapt the Needle-in-a-Haystack retrieval framework and integrate a Flip-style re-evaluation prompt to simulate realistic answer-flipping scenarios. We find that retrieval heads are not primarily responsible for avoiding uncertainty. Instead, we identify a small set of non-retrieval attention heads that disproportionately attend to misleading tokens in uncertain contexts. Masking these heads yields significant improvements, reducing flip behavior by up to 15% without introducing incoherence or overcorrection. However, when tested for downstream tasks, we observe trade-offs with flip behavior. Our findings contribute to the growing field of mechanistic interpretability and present a simple yet effective technique for mitigating uncertainty-driven failure modes in LLMs.

【10】Encoder-Decoder Diffusion Language Models for Efficient Training and Inference
标题：用于高效训练和推理的编码器-解码器扩散语言模型
链接：https://arxiv.org/abs/2510.22852

作者：Marianne Arriola, Yair Schiff, Hao Phung, Aaron Gokaslan, Volodymyr Kuleshov
备注：NeurIPS 2025. We provide the code at this https URL
摘要：Discrete diffusion models enable parallel token sampling for faster inference than autoregressive approaches. However, prior diffusion models use a decoder-only architecture, which requires sampling algorithms that invoke the full network at every denoising step and incur high computational cost. Our key insight is that discrete diffusion models perform two types of computation: 1) representing clean tokens and 2) denoising corrupted tokens, which enables us to use separate modules for each task. We propose an encoder-decoder architecture to accelerate discrete diffusion inference, which relies on an encoder to represent clean tokens and a lightweight decoder to iteratively refine a noised sequence. We also show that this architecture enables faster training of block diffusion models, which partition sequences into blocks for better quality and are commonly used in diffusion language model inference. We introduce a framework for Efficient Encoder-Decoder Diffusion (E2D2), consisting of an architecture with specialized training and sampling algorithms, and we show that E2D2 achieves superior trade-offs between generation quality and inference throughput on summarization, translation, and mathematical reasoning tasks. We provide the code, model weights, and blog post on the project page: https://m-arriola.com/e2d2

【11】Exploration of Summarization by Generative Language Models for Automated Scoring of Long Essays
标题：长论文自动评分的生成语言模型总结探索
链接：https://arxiv.org/abs/2510.22830

作者：Haowei Hua (1), Hong Jiao (2), Xinyi Wang (3) ((1) Princeton University, (2) University of Maryland, College Park, (3) University of Maryland, College Park & Beijing Normal University)
备注：19 pages, 5 Tables 7 Figures, Presentation at Artificial Intelligence in Measurement and Education Conference (AIME-Con)
摘要：BERT and its variants are extensively explored for automated scoring. However, a limit of 512 tokens for these encoder-based models showed the deficiency in automated scoring of long essays. Thus, this research explores generative language models for automated scoring of long essays via summarization and prompting. The results revealed great improvement of scoring accuracy with QWK increased from 0.822 to 0.8878 for the Learning Agency Lab Automated Essay Scoring 2.0 dataset.

【12】VEHME: A Vision-Language Model For Evaluating Handwritten Mathematics Expressions
标题：VEHME：一个手写数学表达式的视觉语言模型
链接：https://arxiv.org/abs/2510.22798

作者：Thu Phuong Nguyen, Duc M. Nguyen, Hyotaek Jeon, Hyunwook Lee, Hyunmin Song, Sungahn Ko, Taehwan Kim
备注：EMNLP 2025. Project Website: this https URL
摘要：Automatically assessing handwritten mathematical solutions is an important problem in educational technology with practical applications, but it remains a significant challenge due to the diverse formats, unstructured layouts, and symbolic complexity of student work. To address this challenge, we introduce VEHME-a Vision-Language Model for Evaluating Handwritten Mathematics Expressions-designed to assess open-form handwritten math responses with high accuracy and interpretable reasoning traces. VEHME integrates a two-phase training pipeline: (i) supervised fine-tuning using structured reasoning data, and (ii) reinforcement learning that aligns model outputs with multi-dimensional grading objectives, including correctness, reasoning depth, and error localization. To enhance spatial understanding, we propose an Expression-Aware Visual Prompting Module, trained on our synthesized multi-line math expressions dataset to robustly guide attention in visually heterogeneous inputs. Evaluated on AIHub and FERMAT datasets, VEHME achieves state-of-the-art performance among open-source models and approaches the accuracy of proprietary systems, demonstrating its potential as a scalable and accessible tool for automated math assessment. Our training and experiment code is publicly available at our GitHub repository.

【13】TELL-TALE: Task Efficient LLMs with Task Aware Layer Elimination
标题：TELL-TARE：具有任务感知层消除的任务高效LLM
链接：https://arxiv.org/abs/2510.22767

作者：Omar Naim, Krish Sharma, Nicholas Asher
摘要：In this paper we introduce Tale, Task-Aware Layer Elimination, an inference-time algorithm that prunes entire transformer layers in an LLM by directly optimizing task-specific validation performance. We evaluate TALE on 9 tasks and 5 models, including LLaMA 3.1 8B, Qwen 2.5 7B, Qwen 2.5 0.5B, Mistral 7B, and Lucie 7B, under both zero-shot and few-shot settings. Unlike prior approaches, TALE requires no retraining and consistently improves accuracy while reducing computational cost across all benchmarks. Furthermore, applying TALE during finetuning leads to additional performance gains. Finally, TALE provides flexible user control over trade-offs between accuracy and efficiency. Mutual information analysis shows that certain layers act as bottlenecks, degrading task-relevant representations. Tale's selective layer removal remedies this problem, producing smaller, faster, and more accurate models that are also faster to fine-tune while offering new insights into transformer interpretability.

【14】SALSA: Single-pass Autoregressive LLM Structured Classification
标题：SALSA：单通道自回归LLM结构化分类
链接：https://arxiv.org/abs/2510.22691

作者：Ruslan Berdichevsky, Shai Nahum-Gefen, Elad Ben Zaken
摘要：Despite their impressive generalization capabilities, instruction-tuned Large Language Models often underperform on text classification benchmarks. We introduce SALSA, a coherent pipeline that combines structured prompting, class-to-token mapping, and parameter-efficient fine-tuning, thereby avoiding cold-start training. Each class label is mapped to a distinct output token, and prompts are constructed to elicit a single-token response. During inference, the model's output is projected only onto the logits of the relevant class tokens, enabling efficient and accurate classification in a single forward pass. SALSA achieves state-of-the-art results across diverse benchmarks, demonstrating its robustness and scalability for LLM-based classification applications.

【15】FastVLM: Self-Speculative Decoding for Fast Vision-Language Model Inference
标题：FastVLM：快速视觉语言模型推理的自我推测解码
链接：https://arxiv.org/abs/2510.22641

作者：Divya Jyoti Bajpai, Manjesh Kumar Hanawal
备注：Accepted for presentation at the main Conference IJCNLP-AACL 2025
摘要：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.

【16】Breaking Agent Backbones: Evaluating the Security of Backbone LLMs in AI Agents
标题：打破代理主干：评估人工智能代理中主干LLM的安全性
链接：https://arxiv.org/abs/2510.22620

作者：Julia Bazinska, Max Mathys, Francesco Casucci, Mateo Rojas-Carulla, Xander Davies, Alexandra Souly, Niklas Pfister
备注：Julia Bazinska and Max Mathys contributed equally
摘要：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 $\operatorname{b}^3$ benchmark, a security benchmark based on 194331 unique crowdsourced adversarial attacks. We then evaluate 31 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.

【17】Text to Trust: Evaluating Fine-Tuning and LoRA Trade-offs in Language Models for Unfair Terms of Service Detection
标题：文本到信任：评估不公平服务条款检测的语言模型中的微调和LoRA权衡
链接：https://arxiv.org/abs/2510.22531

作者：Noshitha Padma Pratyusha Juttu, Sahithi Singireddy, Sravani Gona, Sujal Timilsina
备注：6 pages, including figures and tables. All experiments are reproducible. Code and fine-tuned models are publicly available on: GitHub: (this https URL) and Hugging Face: (this https URL)
摘要：Large Language Models (LLMs) have transformed text understanding, yet their adaptation to specialized legal domains remains constrained by the cost of full fine-tuning. This study provides a systematic evaluation of fine tuning, parameter efficient adaptation (LoRA, QLoRA), and zero-shot prompting strategies for unfair clause detection in Terms of Service (ToS) documents, a key application in legal NLP. We finetune BERT and DistilBERT, apply 4-bit Low-Rank Adaptation (LoRA) to models such as TinyLlama, LLaMA 3B/7B, and SaulLM, and evaluate GPT-4o and O-versions in zero-shot settings. Experiments on the CLAUDETTE-ToS benchmark and the Multilingual Scraper Corpus show that full fine-tuning achieves the strongest precision recall balance, while LoRA-based models provide competitive recall with up to 3x lower memory cost. These findings highlight practical design trade-offs for efficient and domain-adapted LLMs, contributing open baselines for fine-tuning research in legal text processing.

【18】Accelerating Materials Design via LLM-Guided Evolutionary Search
标题：通过LLM引导的进化搜索加速材料设计
链接：https://arxiv.org/abs/2510.22503

作者：Nikhil Abhyankar, Sanchit Kabra, Saaketh Desai, Chandan K. Reddy
摘要：Materials discovery requires navigating vast chemical and structural spaces while satisfying multiple, often conflicting, objectives. We present LLM-guided Evolution for MAterials design (LLEMA), a unified framework that couples the scientific knowledge embedded in large language models with chemistry-informed evolutionary rules and memory-based refinement. At each iteration, an LLM proposes crystallographically specified candidates under explicit property constraints; a surrogate-augmented oracle estimates physicochemical properties; and a multi-objective scorer updates success/failure memories to guide subsequent generations. Evaluated on 14 realistic tasks spanning electronics, energy, coatings, optics, and aerospace, LLEMA discovers candidates that are chemically plausible, thermodynamically stable, and property-aligned, achieving higher hit-rates and stronger Pareto fronts than generative and LLM-only baselines. Ablation studies confirm the importance of rule-guided generation, memory-based refinement, and surrogate prediction. By enforcing synthesizability and multi-objective trade-offs, LLEMA delivers a principled pathway to accelerate practical materials discovery. Code: https://github.com/scientific-discovery/LLEMA

【19】Frustratingly Easy Task-aware Pruning for Large Language Models
标题：大型语言模型的简单任务感知修剪
链接：https://arxiv.org/abs/2510.22489

作者：Yuanhe Tian, Junjie Liu, Xican Yang, Haishan Ye, Yan Song
备注：8 pages, 3 figures
摘要：Pruning provides a practical solution to reduce the resources required to run large language models (LLMs) to benefit from their effective capabilities as well as control their cost for training and inference. Research on LLM pruning often ranks the importance of LLM parameters using their magnitudes and calibration-data activations and removes (or masks) the less important ones, accordingly reducing LLMs' size. However, these approaches primarily focus on preserving the LLM's ability to generate fluent sentences, while neglecting performance on specific domains and tasks. In this paper, we propose a simple yet effective pruning approach for LLMs that preserves task-specific capabilities while shrinking their parameter space. We first analyze how conventional pruning minimizes loss perturbation under general-domain calibration and extend this formulation by incorporating task-specific feature distributions into the importance computation of existing pruning algorithms. Thus, our framework computes separate importance scores using both general and task-specific calibration data, partitions parameters into shared and exclusive groups based on activation-norm differences, and then fuses their scores to guide the pruning process. This design enables our method to integrate seamlessly with various foundation pruning techniques and preserve the LLM's specialized abilities under compression. Experiments on widely used benchmarks demonstrate that our approach is effective and consistently outperforms the baselines with identical pruning ratios and different settings.

【20】Backward-Friendly Optimization: Training Large Language Models with Approximate Gradients under Memory Constraints
标题：向后友好的优化：在内存约束下训练具有近似子集的大型语言模型
链接：https://arxiv.org/abs/2510.22467

作者：Jing Yang, Kaitong Cai, Yijia Fan, Yufeng Yang, Keze Wang

【21】Label Smoothing Improves Gradient Ascent in LLM Unlearning
标题：标签平滑改善LLM取消学习中的梯度上升
链接：https://arxiv.org/abs/2510.22376

作者：Zirui Pang, Hao Zheng, Zhijie Deng, Ling Li, Zixin Zhong, Jiaheng Wei

【22】Mapping Faithful Reasoning in Language Models
标题：语言模型中的忠实推理映射
链接：https://arxiv.org/abs/2510.22362

作者：Jiazheng Li, Andreas Damianou, J Rosser, José Luis Redondo García, Konstantina Palla
备注：9 pages, Accepted to the Mechanistic Interpretability Workshop at NeurIPS 2025

【23】LIFT: Interpretable truck driving risk prediction with literature-informed fine-tuned LLMs
标题：LTIP：通过文献知情的微调LLM进行可解释的卡车驾驶风险预测
链接：https://arxiv.org/abs/2510.22333

作者：Xiao Hu, Yuansheng Lian, Ke Zhang, Yunxuan Li, Yuelong Su, Meng Li

【24】LacMaterial: Large Language Models as Analogical Chemists for Materials Discovery
标题：LacMaterial：大型语言模型作为材料发现的类比化学家
链接：https://arxiv.org/abs/2510.22312

作者：Hongyu Guo

【25】When Fewer Layers Break More Chains: Layer Pruning Harms Test-Time Scaling in LLMs
标题：当更少的层打破更多的链时：层修剪损害LLM中的测试时间缩放
链接：https://arxiv.org/abs/2510.22228

作者：Keyu Wang, Tian Lyu, Guinan Su, Jonas Geiping, Lu Yin, Marco Canini, Shiwei Liu

【26】Edit Less, Achieve More: Dynamic Sparse Neuron Masking for Lifelong Knowledge Editing in LLMs
标题：编辑更少，实现更多：LLM终身知识编辑的动态稀疏神经元掩蔽
链接：https://arxiv.org/abs/2510.22139

作者：Jinzhe Liu, Junshu Sun, Shufan Shen, Chenxue Yang, Shuhui Wang
备注：19 pages, 11 figures, Accepted by NeurIPS 2025

【27】Mint: A Simple Test-Time Adaptation of Vision-Language Models against Common Corruptions
标题：薄荷：视觉语言模型针对常见腐败的简单测试时调整
链接：https://arxiv.org/abs/2510.22127

作者：Wenxuan Bao, Ruxi Deng, Jingrui He
备注：Accepted by NeurIPS 2025

【28】Scaling Up Efficient Small Language Models Serving and Deployment for Semantic Job Search
标题：扩展高效的小语言模型，为语义职位搜索服务和部署
链接：https://arxiv.org/abs/2510.22101

作者：Kayhan Behdin, Qingquan Song, Sriram Vasudevan, Jian Sheng, Xiaojing Ma, Z Zhou, Chuanrui Zhu, Guoyao Li, Chanh Nguyen, Sayan Ghosh, Hejian Sang, Ata Fatahi Baarzi, Sundara Raman Ramachandran, Xiaoqing Wang, Qing Lan, Vinay Y S, Qi Guo, Caleb Johnson, Zhipeng Wang, Fedor Borisyuk

【29】QuArch: A Benchmark for Evaluating LLM Reasoning in Computer Architecture
标题：QuArch：评估计算机架构中LLM推理的基准
链接：https://arxiv.org/abs/2510.22087

作者：Shvetank Prakash, Andrew Cheng, Arya Tschand, Mark Mazumder, Varun Gohil, Jeffrey Ma, Jason Yik, Zishen Wan, Jessica Quaye, Elisavet Lydia Alvanaki, Avinash Kumar, Chandrashis Mazumdar, Tuhin Khare, Alexander Ingare, Ikechukwu Uchendu, Radhika Ghosal, Abhishek Tyagi, Chenyu Wang, Andrea Mattia Garavagno, Sarah Gu, Alice Guo, Grace Hur, Luca Carloni, Tushar Krishna, Ankita Nayak, Amir Yazdanbakhsh, Vijay Janapa Reddi

【30】Jailbreak Mimicry: Automated Discovery of Narrative-Based Jailbreaks for Large Language Models
标题：越狱模仿：大型语言模型基于叙事的越狱的自动发现
链接：https://arxiv.org/abs/2510.22085

作者：Pavlos Ntais
备注：18 pages, 5 figures

【31】LLM-AR: LLM-powered Automated Reasoning Framework
标题：LLM-AR：LLM-powered Automated Reasoning Framework
链接：https://arxiv.org/abs/2510.22034

作者：Rick Chen, Joseph Ternasky, Aaron Ontoyin Yin, Xianling Mu, Fuat Alican, Yigit Ihlamur

【32】Embedding Trust: Semantic Isotropy Predicts Nonfactuality in Long-Form Text Generation
标题：嵌入信任：语义各向同性预测长篇文本生成中的非真实性
链接：https://arxiv.org/abs/2510.21891

作者：Dhrupad Bhardwaj, Julia Kempe, Tim G. J. Rudner

【33】TowerVision: Understanding and Improving Multilinguality in Vision-Language Models
标题：TowerVision：理解和改进视觉语言模型中的多语言性
链接：https://arxiv.org/abs/2510.21849

作者：André G. Viveiros, Patrick Fernandes, Saul Santos, Sonal Sannigrahi, Emmanouil Zaranis, Nuno M. Guerreiro, Amin Farajian, Pierre Colombo, Graham Neubig, André F. T. Martins
备注：15 pages, 7 figures, submitted to arXiv October 2025. All models, datasets, and training code will be released at this https URL

【34】KARIPAP: Quantum-Inspired Tensor Network Compression of Large Language Models Using Infinite Projected Entangled Pair States and Tensor Renormalization Group
标题：卡里普：使用无限投影纠缠对状态和张量重正化群对大型语言模型的量子启发张量网络压缩
链接：https://arxiv.org/abs/2510.21844

作者：Azree Nazri
备注：28 pages

【35】Restoring Pruned Large Language Models via Lost Component Compensation
标题：通过丢失组件补偿恢复修剪后的大型语言模型
链接：https://arxiv.org/abs/2510.21834

作者：Zijian Feng, Hanzhang Zhou, Zixiao Zhu, Tianjiao Li, Jia Jim Deryl Chua, Lee Onn Mak, Gee Wah Ng, Kezhi Mao
备注：NeurIPS 2025 Spotlight

【36】asLLR: LLM based Leads Ranking in Auto Sales
标题：asTLR：基于LLM的汽车销售领先者排名
链接：https://arxiv.org/abs/2510.21713

作者：Yin Sun, Yiwen Liu, Junjie Song, Chenyu Zhang, Xinyuan Zhang, Lingjie Liu, Siqi Chen, Yuji Cao

Graph相关(图学习|图神经网络|图优化等)(24篇)

【1】TAMI: Taming Heterogeneity in Temporal Interactions for Temporal Graph Link Prediction
标题：TAMI：驯服时间交互中的异源以进行时间图链接预测
链接：https://arxiv.org/abs/2510.23577

作者：Zhongyi Yu, Jianqiu Wu, Zhenghao Wu, Shuhan Zhong, Weifeng Su, Chul-Ho Lee, Weipeng Zhuo
备注：Accepted to NeurIPS 2025

【2】A Deep Latent Factor Graph Clustering with Fairness-Utility Trade-off Perspective
标题：基于公平-效用权衡视角的深潜在因素图聚集
链接：https://arxiv.org/abs/2510.23507

作者：Siamak Ghodsi, Amjad Seyedi, Tai Le Quy, Fariba Karimi, Eirini Ntoutsi
备注：Accepted to IEEE Big-Data 2025 main research track. The paper is 10 main pages and 4 pages of Appendix

【3】Adaptive Dual Prompting: Hierarchical Debiasing for Fairness-aware Graph Neural Networks
标题：自适应双重预算分配：公平性感知图神经网络的分层去偏置
链接：https://arxiv.org/abs/2510.23469

作者：Yuhan Yang, Xingbo Fu, Jundong Li

【4】Improving Predictions of Molecular Properties with Graph Featurisation and Heterogeneous Ensemble Models
标题：用图特征化和非均相包络模型改进分子性质的预测
链接：https://arxiv.org/abs/2510.23428

作者：Michael L. Parker, Samar Mahmoud, Bailey Montefiore, Mario Öeren, Himani Tandon, Charlotte Wharrick, Matthew D. Segall
备注：None

【5】Learning from Frustration: Torsor CNNs on Graphs
标题：从挫败中学习：图形上的Torsor CNN
链接：https://arxiv.org/abs/2510.23288

作者：Daiyuan Li, Shreya Arya, Robert Ghrist
备注：19 pages (main text + appendices), 1 figure

【6】Grassmanian Interpolation of Low-Pass Graph Filters: Theory and Applications
标题：低通图过滤器的格拉斯曼内插：理论与应用
链接：https://arxiv.org/abs/2510.23235

作者：Anton Savostianov, Michael T. Schaub, Benjamin Stamm
备注：13 pages

【7】AirFed: Federated Graph-Enhanced Multi-Agent Reinforcement Learning for Multi-UAV Cooperative Mobile Edge Computing
标题：AirFed：用于多无人机协作移动边缘计算的联邦图增强型多智能体强化学习
链接：https://arxiv.org/abs/2510.23053

作者：Zhiyu Wang, Suman Raj, Rajkumar Buyya

【8】QoSGMAA: A Robust Multi-Order Graph Attention and Adversarial Framework for Sparse QoS Prediction
标题：MQSGMAA：用于稀疏服务质量预测的稳健多阶图注意力和对抗框架
链接：https://arxiv.org/abs/2510.22982

作者：Guanchen Du, Jianlong Xu, Mingtong Li, Ruiqi Wang, Qianqing Guo, Caiyi Chen, Qingcao Dai, Yuxiang Zeng

【9】Charting the Design Space of Neural Graph Representations for Subgraph Matching
标题：绘制神经图表示的设计空间以进行子图匹配
链接：https://arxiv.org/abs/2510.22897

作者：Vaibhav Raj, Indradyumna Roy, Ashwin Ramachandran, Soumen Chakrabarti, Abir De
备注：ICLR 2025

【10】Inductive Transfer Learning for Graph-Based Recommenders
标题：基于图的推荐器的归纳迁移学习
链接：https://arxiv.org/abs/2510.22799

作者：Florian Grötschla, Elia Trachsel, Luca A. Lanzendörfer, Roger Wattenhofer
备注：Accepted at the New Perspectives in Graph Machine Learning Workshop at NeurIPS 2025

【11】Enhancing Graph Classification Robustness with Singular Pooling
标题：利用奇异池增强图分类的鲁棒性
链接：https://arxiv.org/abs/2510.22643

作者：Sofiane Ennadir, Oleg Smirnov, Yassine Abbahaddou, Lele Cao, Johannes F. Lutzeyer
备注：Accepted at Neurips 2025

【12】Cross-Paradigm Graph Backdoor Attacks with Promptable Subgraph Triggers
标题：具有可攻击子图触发器的跨范式图后门攻击
链接：https://arxiv.org/abs/2510.22555

作者：Dongyi Liu, Jiangtong Li, Dawei Cheng, Changjun Jiang

【13】Iteratively Refined Early Interaction Alignment for Subgraph Matching based Graph Retrieval
标题：基于子图匹配的图检索的迭代改进早期交互对齐
链接：https://arxiv.org/abs/2510.22538

作者：Ashwin Ramachandran, Vaibhav Raj, Indrayumna Roy, Soumen Chakrabarti, Abir De
备注：None

【14】Random Search Neural Networks for Efficient and Expressive Graph Learning
标题：随机搜索神经网络实现高效且富有表现力的图学习
链接：https://arxiv.org/abs/2510.22520

作者：Michael Ito, Danai Koutra, Jenna Wiens
备注：NEURIPS 2025; version with full appendix

【15】Toward Robust Signed Graph Learning through Joint Input-Target Denoising
标题：通过联合输入目标去噪实现稳健的符号图学习
链接：https://arxiv.org/abs/2510.22513

作者：Junran Wu, Beng Chin Ooi, Ke Xu
备注：ACM MM 2025

【16】Contextual Tokenization for Graph Inverted Indices
标题：图倒置索引的上下文标记化
链接：https://arxiv.org/abs/2510.22479

作者：Pritish Chakraborty, Indradyumna Roy, Soumen Chakrabarti, Abir De
备注：None

【17】GraphTOP: Graph Topology-Oriented Prompting for Graph Neural Networks
标题：GraphTop：图神经网络的面向图布局
链接：https://arxiv.org/abs/2510.22451

作者：Xingbo Fu, Zhenyu Lei, Zihan Chen, Binchi Zhang, Chuxu Zhang, Jundong Li
备注：Accepted by the 39 Annual Conference on Neural Information Processing Systems (NeurIPS 2025)

【18】Dynamic Graph Neural Network for Data-Driven Physiologically Based Pharmacokinetic Modeling
标题：动态图神经网络用于数据驱动的基于生理的药代动力学建模
链接：https://arxiv.org/abs/2510.22096

作者：Su Liu, Xin Hu, Shurong Wen, Jiaqi Liu, Jiexi Xu, Lanruo Wang

【19】Hierarchical Graph Networks for Accurate Weather Forecasting via Lightweight Training
标题：通过轻量级训练实现准确天气预报的分层图网络
链接：https://arxiv.org/abs/2510.22094

作者：Thomas Bailie, S. Karthik Mukkavilli, Varvara Vetrova, Yun Sing Koh

【20】Pruning and Quantization Impact on Graph Neural Networks
标题：修剪和量化对图神经网络的影响
链接：https://arxiv.org/abs/2510.22058

作者：Khatoon Khedri, Reza Rawassizadeh, Qifu Wen, Mehdi Hosseinzadeh

【21】Deep Learning on Real-World Graphs
标题：现实世界图形上的深度学习
链接：https://arxiv.org/abs/2510.21994

作者：Emanuele Rossi
备注：The thesis was submitted for the degree of Doctor of Philosophy in Computing at Imperial College London (February 2024), under the supervision of Prof. Michael M. Bronstein. It includes work published at ICML, ICLR, NeurIPS, and the Learning on Graphs Conference

【22】Boltzmann Graph Ensemble Embeddings for Aptamer Libraries
标题：Boltzmann图嵌入适体库
链接：https://arxiv.org/abs/2510.21980

作者：Starlika Bauskar, Jade Jiao, Narayanan Kannan, Alexander Kimm, Justin M. Baker, Matthew J. Tyler, Andrea L. Bertozzi, Anne M. Andrews

【23】Prefetching Cache Optimization Using Graph Neural Networks: A Modular Framework and Conceptual Analysis
标题：使用图神经网络预取缓存优化：模块化框架和概念分析
链接：https://arxiv.org/abs/2510.21865

作者：F. I. Qowy

【24】Morphology-Aware KOA Classification: Integrating Graph Priors with Vision Models
标题：形态学感知KOA分类：将图先验与视觉模型集成
链接：https://arxiv.org/abs/2510.21801

作者：Marouane Tliba, Mohamed Amine Kerkouri, Yassine Nasser, Nour Aburaed, Aladine Chetouani, Ulas Bagci, Rachid Jennane
备注：Submitted to ICASSP 2026

Transformer(6篇)

【1】A U-Net and Transformer Pipeline for Multilingual Image Translation
标题：用于多语言图像翻译的U-Net和Transformer管道
链接：https://arxiv.org/abs/2510.23554

作者：Siddharth Sahay, Radhika Agarwal
备注：6 pages, 3 figures, 5 tables, and 2 algorithms. Prepared in IEEE double-column format

【2】Transforming volcanic monitoring: A dataset and benchmark for onboard volcano activity detection
标题：改变火山监测：机载火山活动检测的数据集和基准
链接：https://arxiv.org/abs/2510.22889

作者：Darshana Priyasad, Tharindu Fernando, Maryam Haghighat, Harshala Gammulle, Clinton Fookes
备注：Preprint to appear in IEEE IGARSS 2025

【3】Transformer Key-Value Memories Are Nearly as Interpretable as Sparse Autoencoders
标题：Transformer key Value内存几乎与稀疏自动编码器一样可解释
链接：https://arxiv.org/abs/2510.22332

作者：Mengyu Ye, Jun Suzuki, Tatsuro Inaba, Tatsuki Kuribayashi
备注：NeurIPS 2025

【4】Transformer Based Linear Attention with Optimized GPU Kernel Implementation
标题：基于Transformer的线性注意力，并优化了图形处理器内核实现
链接：https://arxiv.org/abs/2510.21956

作者：Armin Gerami, Ramani Duraiswami

【5】Enabling Robust In-Context Memory and Rapid Task Adaptation in Transformers with Hebbian and Gradient-Based Plasticity
标题：在具有Hebbian和基于顺应性的可塑性的Transformers中实现鲁棒的上下文记忆和快速任务适应
链接：https://arxiv.org/abs/2510.21908

作者：Siddharth Chaudhary

【6】Numerical Fragility in Transformers: A Layer-wise Theory for Explaining, Forecasting, and Mitigating Instability
标题：Transformer中的数字脆弱性：解释、预测和缓解不稳定性的分层理论
链接：https://arxiv.org/abs/2510.21770

作者：Jinwoo Baek
备注：15 pages

GAN|对抗|攻击|生成相关(13篇)

【1】Track, Inpaint, Resplat: Subject-driven 3D and 4D Generation with Progressive Texture Infilling
标题：Track、Inpaint、Resplat：采用渐进纹理填充的主题驱动的3D和4D生成
链接：https://arxiv.org/abs/2510.23605

作者：Shuhong Zheng, Ashkan Mirzaei, Igor Gilitschenski
备注：NeurIPS 2025, 38 pages, 22 figures

【2】Symbolic Neural Generation with Applications to Lead Discovery in Drug Design
标题：符号神经生成应用于引领药物设计发现
链接：https://arxiv.org/abs/2510.23379

作者：Ashwin Srinivasan, A Baskar, Tirtharaj Dash, Michael Bain, Sanjay Kumar Dey, Mainak Banerjee
备注：37 pages, 15 figures; partial overlap of experimental results with this https URL

【3】Accelerating Eigenvalue Dataset Generation via Chebyshev Subspace Filter
标题：通过Chebyshev子空间过滤器加速特征值数据集生成
链接：https://arxiv.org/abs/2510.23215

作者：Hong Wang, Jie Wang, Jian Luo, huanshuo dong, Yeqiu Chen, Runmin Jiang, Zhen huang

【4】RL-AUX: Reinforcement Learning for Auxiliary Task Generation
标题：RL-UX：用于辅助任务生成的强化学习
链接：https://arxiv.org/abs/2510.22940

作者：Judah Goldfeder, Matthew So, Hod Lipson

【5】Logical GANs: Adversarial Learning through Ehrenfeucht Fraisse Games
标题：逻辑GAN：通过Escrum feucht Frisse游戏进行对抗学习
链接：https://arxiv.org/abs/2510.22824

作者：Mirco A. Mannucci
备注：12

【6】Optimal Anytime Algorithms for Online Convex Optimization with Adversarial Constraints
标题：具有对抗约束的在线凸优化的最优任意时间算法
链接：https://arxiv.org/abs/2510.22579

作者：Dhruv Sarkar, Abhishek Sinha

【7】Open Multimodal Retrieval-Augmented Factual Image Generation
标题：开放式多模式检索增强事实图像生成
链接：https://arxiv.org/abs/2510.22521

作者：Yang Tian, Fan Liu, Jingyuan Zhang, Wei Bi, Yupeng Hu, Liqiang Nie
备注：Preprint

【8】LAMP: Data-Efficient Linear Affine Weight-Space Models for Parameter-Controlled 3D Shape Generation and Extrapolation
标题：LMA：数据高效的线性仿射权重空间模型，用于参数控制的3D形状生成和外推
链接：https://arxiv.org/abs/2510.22491

作者：Ghadi Nehme, Yanxia Zhang, Dule Shu, Matt Klenk, Faez Ahmed

【9】SecureLearn - An Attack-agnostic Defense for Multiclass Machine Learning Against Data Poisoning Attacks
标题：SecureLearn -针对数据中毒攻击的多类机器学习的攻击不可知防御
链接：https://arxiv.org/abs/2510.22274

作者：Anum Paracha, Junaid Arshad, Mohamed Ben Farah, Khalid Ismail

【10】MAGIC-Flow: Multiscale Adaptive Conditional Flows for Generation and Interpretable Classification
标题：MAGIC-Flow：用于生成和可解释分类的多尺度自适应条件流
链接：https://arxiv.org/abs/2510.22070

作者：Luca Caldera, Giacomo Bottacini, Lara Cavinato

【11】PF$Δ$: A Benchmark Dataset for Power Flow under Load, Generation, and Topology Variations
标题：PF$Δ$：负载、发电和布局变化下潮流的基准数据集
链接：https://arxiv.org/abs/2510.22048

作者：Ana K. Rivera, Anvita Bhagavathula, Alvaro Carbonero, Priya Donti
备注：31 pages, 14 figures. Accepted at NeurIPS 2025

【12】Adversarial Déjà Vu: Jailbreak Dictionary Learning for Stronger Generalization to Unseen Attacks
标题：对抗性Déjà Vu：越狱词典学习以更强有力地概括隐形攻击
链接：https://arxiv.org/abs/2510.21910

作者：Mahavir Dabas, Tran Huynh, Nikhil Reddy Billa, Jiachen T. Wang, Peng Gao, Charith Peris, Yao Ma, Rahul Gupta, Ming Jin, Prateek Mittal, Ruoxi Jia

【13】A Multimodal, Multitask System for Generating E Commerce Text Listings from Images
标题：用于从图像生成电子商务文本列表的多模式、多任务系统
链接：https://arxiv.org/abs/2510.21835

作者：Nayan Kumar Singh
备注：24 pages, 10 figures, 11 tables. Code can be found at: this https URL

半/弱/无/有监督|不确定性|主动学习(17篇)

【1】T-REGS: Minimum Spanning Tree Regularization for Self-Supervised Learning
标题：T-REGS：自我监督学习的最小生成树正规化
链接：https://arxiv.org/abs/2510.23484

作者：Julie Mordacq, David Loiseaux, Vicky Kalogeiton, Steve Oudot
备注：NeurIPS 2025

【2】Schrodinger Neural Network and Uncertainty Quantification: Quantum Machine
标题：Schrodinger神经网络和不确定性量化：量子机
链接：https://arxiv.org/abs/2510.23449

作者：M. M. Hammad
备注：29 pages, 16 figures

【3】Multitask Multimodal Self-Supervised Learning for Medical Images
标题：医学图像的多任务多模式自我监督学习
链接：https://arxiv.org/abs/2510.23325

作者：Cristian Simionescu

【4】Self-induced stochastic resonance: A physics-informed machine learning approach
标题：自诱导随机共振：一种基于物理学的机器学习方法
链接：https://arxiv.org/abs/2510.22848

作者：Divyesh Savaliya, Marius E. Yamakou
备注：22 pages, 10 figures, 58 references

【5】A Theory of the Mechanics of Information: Generalization Through Measurement of Uncertainty (Learning is Measuring)
标题：信息力学理论：通过不确定性测量进行概括（学习就是测量）
链接：https://arxiv.org/abs/2510.22809

作者：Christopher J. Hazard, Michael Resnick, Jacob Beel, Jack Xia, Cade Mack, Dominic Glennie, Matthew Fulp, David Maze, Andrew Bassett, Martin Koistinen
备注：117 pages

【6】Learning Without Augmenting: Unsupervised Time Series Representation Learning via Frame Projections
标题：无需增强的学习：通过帧投影的无监督时间序列表示学习
链接：https://arxiv.org/abs/2510.22655

作者：Berken Utku Demirel, Christian Holz
备注：Published at the Conference on Neural Information Processing Systems (NeurIPS) 2025

【7】Prediction-Powered Semi-Supervised Learning with Online Power Tuning
标题：具有在线功率调整的预测驱动半监督学习
链接：https://arxiv.org/abs/2510.22586

作者：Noa Shoham, Ron Dorfman, Shalev Shaer, Kfir Y. Levy, Yaniv Romano
备注：NeurIPS 2025

【8】Scalable Oversight via Partitioned Human Supervision
标题：通过分区人力监督进行可扩展的监督
链接：https://arxiv.org/abs/2510.22500

作者：Ren Yin, Takashi Ishida, Masashi Sugiyama

【9】Uncertainty quantification in model discovery by distilling interpretable material constitutive models from Gaussian process posteriors
标题：通过从高斯过程后验中提取可解释的材料本构模型来量化模型发现中的不确定性
链接：https://arxiv.org/abs/2510.22345

作者：David Anton, Henning Wessels, Ulrich Römer, Alexander Henkes, Jorge-Humberto Urrea-Quintero

【10】Joint Score-Threshold Optimization for Interpretable Risk Assessment Under Partial Supervision
标题：部分监督下可解释风险评估的联合得分阈值优化
链接：https://arxiv.org/abs/2510.21934

作者：Fardin Gankhanloo, Emmett Springer, Erik H. Hoyer, Daniel L. Young, Kimia Ghobadi

【11】A supervised discriminant data representation: application to pattern classification
标题：有监督的区分数据表示：在模式分类中的应用
链接：https://arxiv.org/abs/2510.21898

作者：Fadi Dornaika, Ahmad Khoder, Abdelmalik Moujahid, Wassim Khoder
备注：None

【12】Multi-Agent Pose Uncertainty: A Differentiable Rendering Cramér-Rao Bound
标题：多智能体姿势不确定性：可区分渲染Cramér-Rao界
链接：https://arxiv.org/abs/2510.21785

作者：Arun Muthukkumar
备注：5 pages, 3 figures, 1 table. Presented at IEEE/CVF International Conference on Computer Vision (ICCV 2025) and IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025)

【13】Macroeconomic Forecasting for the G7 countries under Uncertainty Shocks
标题：不确定性冲击下七国集团国家宏观经济预测
链接：https://arxiv.org/abs/2510.23347

作者：Shovon Sengupta, Sunny Kumar Singh, Tanujit Chakraborty

【14】Semi-Supervised Learning under General Causal Models
标题：一般因果模型下的半监督学习
链接：https://arxiv.org/abs/2510.22567

作者：Archer Moore, Heejung Shim, Jingge Zhu, Mingming Gong
备注：None

【15】Semi-supervised Vertex Hunting, with Applications in Network and Text Analysis
标题：半监督的端点搜寻，及其在网络和文本分析中的应用
链接：https://arxiv.org/abs/2510.22526

作者：Yicong Jiang, Zheng Tracy Ke
备注：None

【16】RGC: a radio AGN classifier based on deep learning. I. A semi-supervised model for the VLA images of bent radio AGNs
标题：RGC：基于深度学习的射电AGN分类器。I.弯曲射电活动星系核VLA图像的半监督模型
链接：https://arxiv.org/abs/2510.22190

作者：M.S. Hossain (1), M.S.H. Shahal (2 and 3), A. Khan (1 and 2), K.M.B. Asad (2 and 4), P. Saikia (5), F. Akter (6), A. Ali (1 and 3), M.A. Amin (1 and 3), A. Momen (1 and 2 and 4), M. Hasan (3), A.K.M.M. Rahman (1 and 3) ((1) Center for Computational and Data Sciences, Independent University, Bangladesh, (2) Center for Astronomy, Space Science and Astrophysics, Independent University, Bangladesh, (3) Department of Computer Science and Engineering, Independent University, Bangladesh, (4) Department of Physical Sciences, Independent University, Bangladesh, (5) Department of Astronomy and Physics, Yale University, USA, (6) Department of Agricultural and Biosystems Engineering, North Dakota State University, USA)
备注：12 pages, 7 pages appendix, 6 figures, submitted to A&A

【17】Frequentist Validity of Epistemic Uncertainty Estimators
标题：认识不确定性估计器的频繁有效性
链接：https://arxiv.org/abs/2510.22063

作者：Anchit Jain, Stephen Bates

迁移|Zero/Few/One-Shot|自适应(17篇)

【1】PTPP-Aware Adaptation Scaling Laws: Predicting Domain-Adaptation Performance at Unseen Pre-Training Budgets
标题：PTPP感知的适应缩放律：在不可见的预训练预算下预测领域适应性能
链接：https://arxiv.org/abs/2510.23198

作者：Etienne Goffinet, Shane Bergsma, Avraham Sheinin, Natalia Vassilieva, Shaheer Muhammad, Preslav Nakov, Gurpreet Gosal

【2】DREaM: Drug-Drug Relation Extraction via Transfer Learning Method
标题：DREaM：通过迁移学习方法提取药物关系
链接：https://arxiv.org/abs/2510.23189

作者：Ali Fata, Hossein Rahmani, Parinaz Soltanzadeh, Amirhossein Derakhshan, Behrouz Minaei Bidgoli

【3】TARC: Time-Adaptive Robotic Control
标题：TARC：时间自适应机器人控制
链接：https://arxiv.org/abs/2510.23176

作者：Arnav Sukhija, Lenart Treven, Jin Cheng, Florian Dörfler, Stelian Coros, Andreas Krause

【4】AG-Fusion: adaptive gated multimodal fusion for 3d object detection in complex scenes
标题：AG-Fusion：自适应门控多模式融合，用于复杂场景中的3D对象检测
链接：https://arxiv.org/abs/2510.23151

作者：Sixian Liu, Chen Xu, Qiang Wang, Donghai Shi, Yiwen Li

【5】DeepSalt: Bridging Laboratory and Satellite Spectra through Domain Adaptation and Knowledge Distillation for Large-Scale Soil Salinity Estimation
标题：DeepSalt：通过领域适应和知识蒸馏将实验室和卫星光谱连接起来，以进行大规模土壤盐分估计
链接：https://arxiv.org/abs/2510.23124

作者：Rupasree Dey, Abdul Matin, Everett Lewark, Tanjim Bin Faruk, Andrei Bachinin, Sam Leuthold, M. Francesca Cotrufo, Shrideep Pallickara, Sangmi Lee Pallickara

【6】Beyond Higher Rank: Token-wise Input-Output Projections for Efficient Low-Rank Adaptation
标题：超越更高等级：基于代币的投入-产出预测，实现高效的低等级适应
链接：https://arxiv.org/abs/2510.23123

作者：Shiwei Li, Xiandi Luo, Haozhao Wang, Xing Tang, Ziqiang Cui, Dugang Liu, Yuhua Li, Xiuqiang He, Ruixuan Li
备注：Accepted by NeurIPS 2025

【7】MoEMeta: Mixture-of-Experts Meta Learning for Few-Shot Relational Learning
标题：MoEMeta：用于Few-Shot关系学习的专家混合Meta学习
链接：https://arxiv.org/abs/2510.23013

作者：Han Wu, Jie Yin
备注：Accepted by NeurIPS 2025

【8】Adaptive Forests For Classification
标题：适应性森林分类
链接：https://arxiv.org/abs/2510.22991

作者：Dimitris Bertsimas, Yubing Cui
备注：Under review at JMLR

【9】SmartMixed: A Two-Phase Training Strategy for Adaptive Activation Function Learning in Neural Networks
标题：SmartMixed：神经网络中自适应激活函数学习的两阶段训练策略
链接：https://arxiv.org/abs/2510.22450

作者：Amin Omidvar

【10】Simplifying Knowledge Transfer in Pretrained Models
标题：简化预训练模型中的知识转移
链接：https://arxiv.org/abs/2510.22208

作者：Siddharth Jain, Shyamgopal Karthik, Vineet Gandhi
备注：12 pages, 3 figures, 6 tables, Accepted at TMLR 2025

【11】ATLAS: Adaptive Transfer Scaling Laws for Multilingual Pretraining, Finetuning, and Decoding the Curse of Multilinguality
标题：ATLAS：多语言预训练、微调和解码多语言诅咒的自适应转移缩放定律
链接：https://arxiv.org/abs/2510.22037

作者：Shayne Longpre, Sneha Kudugunta, Niklas Muennighoff, I-Hung Hsu, Isaac Caswell, Alex Pentland, Sercan Arik, Chen-Yu Lee, Sayna Ebrahimi

【12】GAPO: Group Adaptive Policy Optimization for Real-World Code Edit
标题：GAPO：针对现实世界代码的群体自适应政策优化编辑
链接：https://arxiv.org/abs/2510.21830

作者：Jianqing Zhang, Zhezheng Hao, Wei Xia, Hande Dong, Hong Wang, Chenxing Wei, Yuyan Zhou, Yubin Qi, Qiang Lin, Jian Cao

【13】Quantifying Multimodal Imbalance: A GMM-Guided Adaptive Loss for Audio-Visual Learning
标题：量化多模式失衡：GMM引导的视听学习适应性损失
链接：https://arxiv.org/abs/2510.21797

作者：Zhaocheng Liu, Zhiwen Yu, Xiaoqing Liu

【14】Words to Waves: Emotion-Adaptive Music Recommendation System
标题：Words to Waves：描述自适应音乐推荐系统
链接：https://arxiv.org/abs/2510.21724

作者：Apoorva Chavali, Reeve Menezes

【15】Synthetic-to-Real Transfer Learning for Chromatin-Sensitive PWS Microscopy
标题：染色敏感PWS显微镜的综合到真实转移学习
链接：https://arxiv.org/abs/2510.22239

作者：Jahidul Arafat, Sanjaya Poudel
备注：24 pages, 5 figures and 4 tables

【16】Input Adaptive Bayesian Model Averaging
标题：输入自适应Bayesian模型平均
链接：https://arxiv.org/abs/2510.22054

作者：Yuli Slavutsky, Sebastian Salazar, David M. Blei

【17】Adaptive Split-MMD Training for Small-Sample Cross-Dataset P300 EEG Classification
标题：小样本交叉数据集P300脑电分类的自适应分裂MMD训练
链接：https://arxiv.org/abs/2510.21969

作者：Weiyu Chen, Arnaud Delorme
备注：8 pages, 5 figures. Submitted to IEEE BIBM 2025 Workshop on Machine Learning for EEG Signal Processing (MLESP)

强化学习(16篇)

【1】Learning to Reason Efficiently with Discounted Reinforcement Learning
标题：通过折扣强化学习学习有效推理
链接：https://arxiv.org/abs/2510.23486

作者：Alex Ayoub, Kavosh Asadi, Dale Schuurmans, Csaba Szepesvári, Karim Bouyarmane

【2】The Best of N Worlds: Aligning Reinforcement Learning with Best-of-N Sampling via max@k Optimisation
标题：N个世界中的最佳：通过max@k优化将强化学习与N个中的最佳采样保持一致
链接：https://arxiv.org/abs/2510.23393

作者：Farid Bagirov, Mikhail Arkhipov, Ksenia Sycheva, Evgeniy Glukhov, Egor Bogomolov

【3】Human-Like Goalkeeping in a Realistic Football Simulation: a Sample-Efficient Reinforcement Learning Approach
标题：真实足球模拟中的类人守门员：一种样本高效的强化学习方法
链接：https://arxiv.org/abs/2510.23216

作者：Alessandro Sestini, Joakim Bergdahl, Jean-Philippe Barrette-LaPierre, Florian Fuchs, Brady Chen, Micheal Jones, Linus Gisslén

【4】Adapting Interleaved Encoders with PPO for Language-Guided Reinforcement Learning in BabyAI
标题：在BabyAI中调整交错编码器与PPO，以实现百分比引导的强化学习
链接：https://arxiv.org/abs/2510.23148

作者：Aryan Mathur, Asaduddin Ahmed
备注：Undergraduate research project, IIT Palakkad, 2025

【5】Towards Stable and Effective Reinforcement Learning for Mixture-of-Experts
标题：面向混合专家的稳定有效强化学习
链接：https://arxiv.org/abs/2510.23027

作者：Di Zhang, Xun Wu, Shaohan Huang, Yaru Hao, Li Dong, Zewen Chi, Zhifang Sui, Furu Wei

【6】Guardian: Decoupling Exploration from Safety in Reinforcement Learning
标题：卫报：强化学习中探索与安全脱钩
链接：https://arxiv.org/abs/2510.22859

作者：Kaitong Cai, Jusheng Zhang, Jing Yang, Keze Wang

【7】HRM-Agent: Training a recurrent reasoning model in dynamic environments using reinforcement learning
标题：HRM-Agent：使用强化学习在动态环境中训练循环推理模型
链接：https://arxiv.org/abs/2510.22832

作者：Long H Dang, David Rawlinson
备注：14 pages, 9 figures, 1 table

【8】FlowCritic: Bridging Value Estimation with Flow Matching in Reinforcement Learning
标题：FlowCritic：桥接强化学习中的流匹配值估计
链接：https://arxiv.org/abs/2510.22686

作者：Shan Zhong, Shutong Ding, He Diao, Xiangyu Wang, Kah Chan Teh, Bei Peng

【9】BLIP-FusePPO: A Vision-Language Deep Reinforcement Learning Framework for Lane Keeping in Autonomous Vehicles
标题：BLIP-FusePPO：用于自动驾驶车辆车道保持的视觉语言深度强化学习框架
链接：https://arxiv.org/abs/2510.22370

作者：Seyed Ahmad Hosseini Miangoleh, Amin Jalal Aghdasian, Farzaneh Abdollahi
备注：his https URL

【10】Solving Continuous Mean Field Games: Deep Reinforcement Learning for Non-Stationary Dynamics
标题：解决连续平均场博弈：非平稳动力学的深度强化学习
链接：https://arxiv.org/abs/2510.22158

作者：Lorenzo Magnino, Kai Shao, Zida Wu, Jiacheng Shen, Mathieu Laurière
备注：Neurips 2025

【11】Agentic Reinforcement Learning for Real-World Code Repair
标题：用于现实世界代码修复的抽象强化学习
链接：https://arxiv.org/abs/2510.22075

作者：Siyu Zhu, Anastasiya Karpovich, Albert Chen, Jessica Koscheka, Shailesh Jannu, Di Wen, Yuqing Zhu, Rohit Jain, Alborz Geramifard

【12】Online Optimization for Offline Safe Reinforcement Learning
标题：线下安全强化学习的在线优化
链接：https://arxiv.org/abs/2510.22027

作者：Yassine Chemingui, Aryan Deshwal, Alan Fern, Thanh Nguyen-Tang, Janardhan Rao Doppa
备注：To appear in NeurIPS 2025 Conference

【13】Do You Trust the Process?: Modeling Institutional Trust for Community Adoption of Reinforcement Learning Policies
标题：您相信这个过程吗？：社区采用强化学习政策的机构信任建模
链接：https://arxiv.org/abs/2510.22017

作者：Naina Balepur, Xingrui Pei, Hari Sundaram

【14】Computational Hardness of Reinforcement Learning with Partial $q^π$-Realizability
标题：具有部分$q & pi $-可实现性的强化学习的计算难度
链接：https://arxiv.org/abs/2510.21888

作者：Shayan Karimi, Xiaoqi Tan
备注：to be published in NeurIPS 2025

【15】Taxonomy and Trends in Reinforcement Learning for Robotics and Control Systems: A Structured Review
标题：机器人和控制系统强化学习的分类和趋势：结构化评论
链接：https://arxiv.org/abs/2510.21758

作者：Kumater Ter, RexCharles Donatus, Ore-Ofe Ajayi, Daniel Udekwe

【16】Reinforcement learning-guided optimization of critical current in high-temperature superconductors
标题：强化学习引导的高温超导体临界电流优化
链接：https://arxiv.org/abs/2510.22424

作者：Mouyang Cheng, Qiwei Wan, Bowen Yu, Eunbi Rha, Michael J Landry, Mingda Li
备注：7 pages, 4 figures

元学习(4篇)

【1】An Information-Theoretic Analysis of Out-of-Distribution Generalization in Meta-Learning with Applications to Meta-RL
标题：元学习中分布外概括的信息论分析及其在元RL的应用
链接：https://arxiv.org/abs/2510.23448

作者：Xingtu Liu

【2】SwiftTS: A Swift Selection Framework for Time Series Pre-trained Models via Multi-task Meta-Learning
标题：SwiftTS：通过多任务元学习进行时间序列预训练模型的Swift选择框架
链接：https://arxiv.org/abs/2510.23051

作者：Tengxue Zhang, Biao Ouyang, Yang Shu, Xinyang Chen, Chenjuan Guo, Bin Yang
备注：10 pages,6 figures

【3】Training data membership inference via Gaussian process meta-modeling: a post-hoc analysis approach
标题：通过高斯过程元建模训练数据隶属度推断：事后分析方法
链接：https://arxiv.org/abs/2510.21846

作者：Yongchao Huang, Pengfei Zhang, Shahzad Mumtaz
备注：10 pages

【4】MetaCaDI: A Meta-Learning Framework for Scalable Causal Discovery with Unknown Interventions
标题：MetaCaDI：一个用于使用未知干预的可扩展原因发现的元学习框架
链接：https://arxiv.org/abs/2510.22298

作者：Hans Jarett Ong, Yoichi Chikahara, Tomoharu Iwata
备注：8 pages, 2 figures

符号|符号学习(1篇)

【1】Predicting symbolic ODEs from multiple trajectories
标题：从多个轨迹预测象征性ODE
链接：https://arxiv.org/abs/2510.23295

作者：Yakup Emre Şahin, Niki Kilbertus, Sören Becker
备注：Published at: 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop: Machine Learning and the Physical Sciences

医学相关(7篇)

【1】Progressive Growing of Patch Size: Curriculum Learning for Accelerated and Improved Medical Image Segmentation
标题：补丁大小的逐步增长：加速和改进医学图像分割的课程学习
链接：https://arxiv.org/abs/2510.23241

作者：Stefan M. Fischer, Johannes Kiechle, Laura Daza, Lina Felsner, Richard Osuala, Daniel M. Lang, Karim Lekadir, Jan C. Peeken, Julia A. Schnabel
备注：Journal Extension of "Progressive Growing of Patch Size: Resource-Efficient Curriculum Learning for Dense Prediction Tasks" (MICCAI2024) submitted to MedIA

【2】Privacy-Aware Federated nnU-Net for ECG Page Digitization
标题：隐私意识联邦nnU-Net用于心电图页面数字化
链接：https://arxiv.org/abs/2510.22387

作者：Nader Nemati

【3】AnyECG-Lab: An Exploration Study of Fine-tuning an ECG Foundation Model to Estimate Laboratory Values from Single-Lead ECG Signals
标题：AnyECG-Lab：微调心电图基础模型以从单导心电图信号估计实验室值的探索性研究
链接：https://arxiv.org/abs/2510.22301

作者：Yujie Xiao, Gongzhen Tang, Wenhui Liu, Jun Li, Guangkun Nie, Zhuoran Kan, Deyun Zhang, Qinghao Zhao, Shenda Hong

【4】Predicting Metabolic Dysfunction-Associated Steatotic Liver Disease using Machine Learning Methods
标题：使用机器学习方法预测代谢功能障碍相关脂肪性肝脏疾病
链接：https://arxiv.org/abs/2510.22293

作者：Mary E. An, Paul Griffin, Jonathan G. Stine, Ramakrishna Balakrishnan, Ram Sriram, Soundar Kumara

【5】Efficient Utility-Preserving Machine Unlearning with Implicit Gradient Surgery
标题：通过隐式梯度手术实现高效的效用保留机器去学习
链接：https://arxiv.org/abs/2510.22124

作者：Shiji Zhou (Institute of Artificial Intelligence, Beihang University, Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing, Beihang University), Tianbai Yu (University of Illinois at Urbana-Champaign), Zhi Zhang (University of Amsterdam), Heng Chang (Tsinghua University), Xiao Zhou (Tsinghua University), Dong Wu (YanTron Technology Co. Ltd), Han Zhao (University of Illinois at Urbana-Champaign)
备注：Corresponding author: Shiji Zhou (zhoushiji25@buaa.this http URL). Shiji Zhou and Tianbai Yu contributed equally

【6】Unlocking Biomedical Insights: Hierarchical Attention Networks for High-Dimensional Data Interpretation
标题：解锁生物医学洞察：用于多维数据解释的分层注意力网络
链接：https://arxiv.org/abs/2510.21820

作者：Rekha R Nair, Tina Babu, Alavikunhu Panthakkan, Hussain Al-Ahmad, Balamurugan Balusamy

【7】Expert Validation of Synthetic Cervical Spine Radiographs Generated with a Denoising Diffusion Probabilistic Model
标题：用降噪扩散概率模型生成的合成宫颈X光片的专家验证
链接：https://arxiv.org/abs/2510.22166

作者：Austin A. Barr, Brij S. Karmur, Anthony J. Winder, Eddie Guo, John T. Lysack, James N. Scott, William F. Morrish, Muneer Eesa, Morgan Willson, David W. Cadotte, Michael M.H. Yang, Ian Y.M. Chan, Sanju Lama, Garnette R. Sutherland
备注：10 pages, 4 figures, 1 table

蒸馏|知识提取(1篇)

【1】Sentinel: Dynamic Knowledge Distillation for Personalized Federated Intrusion Detection in Heterogeneous IoT Networks
标题：Sentinel：用于异类物联网网络中个性化联邦入侵检测的动态知识提炼
链接：https://arxiv.org/abs/2510.23019

作者：Gurpreet Singh, Keshav Sood, P. Rajalakshmi, Yong Xiang
备注：This is a preprint version of a paper currently under review for possible publication in IEEE TDSC

推荐(2篇)

【1】Unifying Inductive, Cross-Domain, and Multimodal Learning for Robust and Generalizable Recommendation
标题：统一归纳、跨域和多模态学习以实现鲁棒和可推广的推荐
链接：https://arxiv.org/abs/2510.21812

作者：Chanyoung Chung, Kyeongryul Lee, Sunbin Park, Joyce Jiyoung Whang
备注：7 pages, 3 figures, and 4 tables. International Workshop on Multimodal Generative Search and Recommendation (MMGenSR) at The 34th ACM International Conference on Information and Knowledge Management (CIKM 2025)

【2】DiffGRM: Diffusion-based Generative Recommendation Model
标题：迪夫GRM：基于扩散的生成式推荐模型
链接：https://arxiv.org/abs/2510.21805

作者：Zhao Liu, Yichen Zhu, Yiqing Yang, Guoping Tang, Rui Huang, Qiang Luo, Xiao Lv, Ruiming Tang, Kun Gai, Guorui Zhou
备注：13 pages, 5 figures

聚类(5篇)

【1】Coresets for Clustering Under Stochastic Noise
标题：随机噪声下聚类的核心集
链接：https://arxiv.org/abs/2510.23438

作者：Lingxiao Huang, Zhize Li, Nisheeth K. Vishnoi, Runkai Yang, Haoyu Zhao
备注：This paper has been accepted by NeurIPS 2025

【2】GCAO: Group-driven Clustering via Gravitational Attraction and Optimization
标题：GCAO：通过引力吸引和优化的群体驱动集群
链接：https://arxiv.org/abs/2510.23259

作者：Qi Li, Jun Wang

【3】A method for outlier detection based on cluster analysis and visual expert criteria
标题：一种基于集群分析和视觉专家准则的离群点检测方法
链接：https://arxiv.org/abs/2510.23136

作者：Juan A. Lara, David Lizcano, Víctor Rampérez, Javier Soriano
备注：None

【4】Clustering by Denoising: Latent plug-and-play diffusion for single-cell data
标题：通过去噪进行聚集：单单元数据的潜在即插即用扩散
链接：https://arxiv.org/abs/2510.22835

作者：Dominik Meier, Shixing Yu, Sagnik Nandy, Promit Ghosal, Kyra Gan

【5】A Scalable Global Optimization Algorithm For Constrained Clustering
标题：一种可扩展的约束聚集全局优化算法
链接：https://arxiv.org/abs/2510.22519

作者：Pedro Chumpitaz-Flores, My Duong, Cristobal Heredia, Kaixun Hua
备注：21 pages, 4 figures, 9 tables

超分辨率|去噪|去模糊|去雾(2篇)

【1】DDTR: Diffusion Denoising Trace Recovery
标题：DDTR：扩散降噪微量回收
链接：https://arxiv.org/abs/2510.22553

作者：Maximilian Matyash, Avigdor Gal, Arik Senderovich

【2】A Free Probabilistic Framework for Denoising Diffusion Models: Entropy, Transport, and Reverse Processes
标题：去噪扩散模型的自由概率框架：熵、输运和反向过程
链接：https://arxiv.org/abs/2510.22778

作者：Swagatam Das

自动驾驶|车辆|车道检测等(2篇)

【1】GRAD: Real-Time Gated Recurrent Anomaly Detection in Autonomous Vehicle Sensors Using Reinforced EMA and Multi-Stage Sliding Window Techniques
标题：GRAD：使用增强型EMA和多阶段滑动窗口技术在自动驾驶车辆传感器中进行实时门控复发异常检测
链接：https://arxiv.org/abs/2510.23327

作者：Mohammad Hossein Jafari Naeimi, Ali Norouzi, Athena Abdi

【2】Learn2Drive: A neural network-based framework for socially compliant automated vehicle control
标题：Learn 2Drive：一个基于神经网络的框架，用于符合社会要求的自动化车辆控制
链接：https://arxiv.org/abs/2510.21736

作者：Yuhui Liu, Samannita Halder, Shian Wang, Tianyi Li

点云|SLAM|雷达|激光|深度RGBD相关(2篇)

【1】Linearized Optimal Transport for Analysis of High-Dimensional Point-Cloud and Single-Cell Data
标题：用于分析多维点云和单细胞数据的线性化最佳传输
链接：https://arxiv.org/abs/2510.22033

作者：Tianxiang Wang, Yingtong Ke, Dhananjay Bhaskar, Smita Krishnaswamy, Alexander Cloninger
备注：11 pages, 5 figures

【2】Generative AI in Depth: A Survey of Recent Advances, Model Variants, and Real-World Applications
标题：深入生成人工智能：最新进展、模型变体和现实应用的概览
链接：https://arxiv.org/abs/2510.21887

作者：Shamim Yazdani, Akansha Singh, Nripsuta Saxena, Zichong Wang, Avash Palikhe, Deng Pan, Umapada Pal, Jie Yang, Wenbin Zhang
备注：Accepted by the Journal of Big Data

联邦学习|隐私保护|加密(4篇)

【1】Differential Privacy as a Perk: Federated Learning over Multiple-Access Fading Channels with a Multi-Antenna Base Station
标题：差异隐私作为福利：具有多天线基站的多址衰落通道上的联邦学习
链接：https://arxiv.org/abs/2510.23463

作者：Hao Liang, Haifeng Wen, Kaishun Wu, Hong Xing
备注：15 pages, 5 figures, submitted for possible publication

【2】SGFusion: Stochastic Geographic Gradient Fusion in Federated Learning
标题：SGFusion：联邦学习中的随机地理梯度融合
链接：https://arxiv.org/abs/2510.23455

作者：Khoa Nguyen, Khang Tran, NhatHai Phan, Cristian Borcea, Rouming Jin, Issa Khalil

【3】Learning Reconfigurable Representations for Multimodal Federated Learning with Missing Data
标题：缺失数据的多模式联邦学习学习的可重配置表示
链接：https://arxiv.org/abs/2510.22880

作者：Duong M. Nguyen, Trong Nghia Hoang, Thanh Trung Huynh, Quoc Viet Hung Nguyen, Phi Le Nguyen
备注：Accepted at NeurIPS 2025

【4】Power to the Clients: Federated Learning in a Dictatorship Setting
标题：客户的权力：独裁环境下的联邦学习
链接：https://arxiv.org/abs/2510.22149

作者：Mohammadsajad Alipour, Mohammad Mohammadi Amiri

推理|分析|理解|解释(19篇)

【1】When No Paths Lead to Rome: Benchmarking Systematic Neural Relational Reasoning
标题：当无路可达罗马：系统神经关系推理的基准
链接：https://arxiv.org/abs/2510.23532

作者：Anirban Das, Irtaza Khalid, Rafael Peñaloza, Steven Schockaert
备注：accepted at NeurIPS 2025 D&B track

【2】Bayes-Split-Edge: Bayesian Optimization for Constrained Collaborative Inference in Wireless Edge Systems
标题：Bayes-Split-Edge：无线边缘系统中约束协作推理的Bayesian优化
链接：https://arxiv.org/abs/2510.23503

作者：Fatemeh Zahra Safaeipour, Jacob Chakareski, Morteza Hashemi

【3】PAHQ: Accelerating Automated Circuit Discovery through Mixed-Precision Inference Optimization
标题：PAHQ：通过混合精度推理优化加速自动电路发现
链接：https://arxiv.org/abs/2510.23264

作者：Xinhai Wang, Shu Yang, Liangyu Wang, Lin Zhang, Huanyi Xie, Lijie Hu, Di Wang

【4】Deep Active Inference with Diffusion Policy and Multiple Timescale World Model for Real-World Exploration and Navigation
标题：用于现实世界探索和导航的扩散政策和多时间尺度世界模型的深度主动推理
链接：https://arxiv.org/abs/2510.23258

作者：Riko Yokozawa, Kentaro Fujii, Yuta Nomura, Shingo Murata
备注：Preprint version

【5】Towards Personalized Treatment Plan: Geometrical Model-Agnostic Approach to Counterfactual Explanations
标题：走向个性化的治疗计划：几何模型不可知的反事实解释方法
链接：https://arxiv.org/abs/2510.22911

作者：Daniel Sin, Milad Toutounchian
备注：This paper is 15 pages long consisting of multiple sections including an abstract, introduction, related works, methodology, results, ablation studies, conclusion, future works, and an appendix section. There are 10 figures and 5 tables in total

【6】Once Upon an Input: Reasoning via Per-Instance Program Synthesis
标题：一旦输入：通过逐实例程序合成进行推理
链接：https://arxiv.org/abs/2510.22849

作者：Adam Stein, Neelay Velingker, Mayur Naik, Eric Wong
备注：Accepted at NeurIPS 2025. 34 pages, 7 figures

【7】A roadmap for curvature-based geometric data analysis and learning
标题：基于弯曲的几何数据分析和学习路线图
链接：https://arxiv.org/abs/2510.22599

作者：Yasharth Yadav, Kelin Xia

【8】Combining Deep Learning and Explainable AI for Toxicity Prediction of Chemical Compounds
标题：结合深度学习和可解释人工智能进行化合物毒性预测
链接：https://arxiv.org/abs/2510.22572

作者：Eduard Popescu, Adrian Groza, Andreea Cernat

【9】FAPO: Flawed-Aware Policy Optimization for Efficient and Reliable Reasoning
标题：FAPO：有缺陷的策略优化，以实现高效可靠的推理
链接：https://arxiv.org/abs/2510.22543

作者：Yuyang Ding, Chi Zhang, Juntao Li, Haibin Lin, Xin Liu, Min Zhang
备注：Project page: this https URL

【10】Benchmarking Egocentric Multimodal Goal Inference for Assistive Wearable Agents
标题：可穿戴辅助智能体自我中心多模态目标推理的基准测试
链接：https://arxiv.org/abs/2510.22443

作者：Vijay Veerabadran, Fanyi Xiao, Nitin Kamra, Pedro Matias, Joy Chen, Caley Drooff, Brett D Roads, Riley Williams, Ethan Henderson, Xuanyi Zhao, Kevin Carlberg, Joseph Tighe, Karl Ridgeway
备注：Accepted as a spotlight paper at the 39th Conference on Neural Information Processing Systems (NeurIPS 2025)

【11】DynaSolidGeo: A Dynamic Benchmark for Genuine Spatial Mathematical Reasoning of VLMs in Solid Geometry
标题：DynaSolidGeo：实体几何中VLM真正空间数学推理的动态基准
链接：https://arxiv.org/abs/2510.22340

作者：Changti Wu, Shijie Lian, Zihao Liu, Lei Zhang, Laurence Tianruo Yang, Kai Chen
备注：The code and dataset are available at \href{this https URL}{DynaSolidGeo}

【12】A Multi-level Analysis of Factors Associated with Student Performance: A Machine Learning Approach to the SAEB Microdata
标题：与学生表现相关因素的多层次分析：SAEB微数据的机器学习方法
链接：https://arxiv.org/abs/2510.22266

作者：Rodrigo Tertulino, Ricardo Almeida
备注：This article is being prepared for submission to the International Journal of Educational Technology in Higher Education

【13】LUNA: Efficient and Topology-Agnostic Foundation Model for EEG Signal Analysis
标题：LUNA：高效且不可知的脑电信号分析基础模型
链接：https://arxiv.org/abs/2510.22257

作者：Berkay Döner, Thorir Mar Ingolfsson, Luca Benini, Yawei Li
备注：NeurIPS camera-ready version, 27 pages, 10 figures, 13 tables

【14】Beyond Reasoning Gains: Mitigating General Capabilities Forgetting in Large Reasoning Models
标题：超越推理收益：减轻大型推理模型中的一般能力遗忘
链接：https://arxiv.org/abs/2510.21978

作者：Hoang Phan, Xianjun Yang, Kevin Yao, Jingyu Zhang, Shengjie Bi, Xiaocheng Tang, Madian Khabsa, Lijuan Liu, Deren Lei

【15】The Mirror Loop: Recursive Non-Convergence in Generative Reasoning Systems
标题：镜像循环：生成推理系统中的渐进非收敛
链接：https://arxiv.org/abs/2510.21861

作者：Bentley DeVilling (Course Correct Labs, Independent Research Group)
备注：18 pages, 2 figures. Category: cs.LG. Code and data: this https URL

【16】Towards Interpretable Deep Learning and Analysis of Dynamical Systems via the Discrete Empirical Interpolation Method
标题：通过离散经验插值方法实现可解释深度学习和动态系统分析
链接：https://arxiv.org/abs/2510.21852

作者：Hojin Kim, Romit Maulik
备注：9 pages, 12 figures

【17】Analysis of accuracy and efficiency of neural networks to simulate Navier-Stokes fluid flows with obstacles
标题：神经网络模拟有障碍物的纳维尔-斯托克斯流体流动的准确性和效率分析
链接：https://arxiv.org/abs/2510.22976

作者：Rui Hespanha, Elliot McGuire, João Hespanha

【18】Statistical Analysis of the Sinkhorn Iterations for Two-Sample Schrödinger Bridge Estimation
标题：双样本薛定汉桥估计的Sinkhorn迭代的统计分析
链接：https://arxiv.org/abs/2510.22560

作者：Ibuki Maeda, Rentian Yao, Atsushi Nitanda
备注：30 pages

【19】Bridging Prediction and Attribution: Identifying Forward and Backward Causal Influence Ranges Using Assimilative Causal Inference
标题：桥梁预测和归因：使用同化因果推理识别前向和后向因果影响范围
链接：https://arxiv.org/abs/2510.21889

作者：Marios Andreou, Nan Chen
备注：39 pages (Main Text pp. 1--25; Supplementary Materials/Appendix pp. 26--35), 9 figures (all in Main Text). Submitted for peer-review to SIAM/ASA Journal on Uncertainty Quantification. Code available upon request. For more info see this https URL

检测相关(5篇)

【1】Diffuse to Detect: A Generalizable Framework for Anomaly Detection with Diffusion Models Applications to UAVs and Beyond
标题：扩散检测：具有扩散模型应用于无人机及其他领域的异常检测的通用框架
链接：https://arxiv.org/abs/2510.22928

作者：Mingze Gong, Juan Du, Jianbang You

【2】CLEANet: Robust and Efficient Anomaly Detection in Contaminated Multivariate Time Series
标题：CLEANet：多变量污染时间序列的鲁棒高效异常检测
链接：https://arxiv.org/abs/2510.22619

作者：Songhan Zhang, Yuanhao Lai, Pengfei Zheng, Boxi Yu, Xiaoying Tang, Qiuai Fu, Pinjia He

【3】Optimal Detection for Language Watermarks with Pseudorandom Collision
标题：伪随机冲突语言水印的最佳检测
链接：https://arxiv.org/abs/2510.22007

作者：T. Tony Cai, Xiang Li, Qi Long, Weijie J. Su, Garrett G. Wen

【4】Quantum Autoencoders for Anomaly Detection in Cybersecurity
标题：用于网络安全异常检测的量子自动编码器
链接：https://arxiv.org/abs/2510.21837

作者：Rohan Senthil, Swee Liang Wong

【5】A Feature Engineering Approach for Business Impact-Oriented Failure Detection in Distributed Instant Payment Systems
标题：分布式即时支付系统中面向业务影响的故障检测的特征工程方法
链接：https://arxiv.org/abs/2510.21710

作者：Lorenzo Porcelli

分类|识别(11篇)

【1】Arabic Little STT: Arabic Children Speech Recognition Dataset
标题：阿拉伯语Little STT：阿拉伯语儿童语音识别数据集
链接：https://arxiv.org/abs/2510.23319

作者：Mouhand Alkadri, Dania Desouki, Khloud Al Jallad

【2】Enabling Vibration-Based Gesture Recognition on Everyday Furniture via Energy-Efficient FPGA Implementation of 1D Convolutional Networks
标题：通过节能的1D卷积网络实现日常家具基于振动的手势识别
链接：https://arxiv.org/abs/2510.23156

作者：Koki Shibata, Tianheng Ling, Chao Qian, Tomokazu Matsui, Hirohiko Suwa, Keiichi Yasumoto, Gregor Schiele
备注：9 pages, 5 figures, 5 tables, accepted by 2025 IEEE Annual Congress on Artificial Intelligence of Things (IEEE AIoT)

【3】AI based signage classification for linguistic landscape studies
标题：基于人工智能的标牌分类用于语言景观研究
链接：https://arxiv.org/abs/2510.22885

作者：Yuqin Jiang, Song Jiang, Jacob Algrim, Trevor Harms, Maxwell Koenen, Xinya Lan, Xingyu Li, Chun-Han Lin, Jia Liu, Jiayang Sun, Henry Zenger

【4】Identification of Causal Direction under an Arbitrary Number of Latent Confounders
标题：任意数量潜在混杂因素下因果方向的识别
链接：https://arxiv.org/abs/2510.22711

作者：Wei Chen, Linjun Peng, Zhiyi Huang, Haoyue Dai, Zhifeng Hao, Ruichu Cai, Kun Zhang

【5】Does Homophily Help in Robust Test-time Node Classification?
标题：同质性有助于稳健的测试时节点分类吗？
链接：https://arxiv.org/abs/2510.22289

作者：Yan Jiang, Ruihong Qiu, Zi Huang

【6】Cost-Sensitive Evaluation for Binary Classifiers
标题：二进制分类器的成本敏感评估
链接：https://arxiv.org/abs/2510.22016

作者：Pierangelo Lombardo, Antonio Casoli, Cristian Cingolani, Shola Oshodi, Michele Zanatta

【7】RatioWaveNet: A Learnable RDWT Front-End for Robust and Interpretable EEG Motor-Imagery Classification
标题：RatioWaveNet：一个可学习的RDWT前端，用于稳健且可解释的脑电运动图像分类
链接：https://arxiv.org/abs/2510.21841

作者：Marco Siino, Giuseppe Bonomo, Rosario Sorbello, Ilenia Tinnirello

【8】Minimizing Human Intervention in Online Classification
标题：最大限度地减少在线分类中的人为干预
链接：https://arxiv.org/abs/2510.23557

作者：William Réveillard, Vasileios Saketos, Alexandre Proutiere, Richard Combes
备注：49 pages, 8 figures

【9】Quantum Phase Classification of Rydberg Atom Systems Using Resource-Efficient Variational Quantum Circuits and Classical Shadows
标题：使用资源高效的变分量子电路和经典阴影对里德伯原子系统进行量子相分类
链接：https://arxiv.org/abs/2510.23489

作者：Hemish Ahuja, Samradh Bhardwaj, Kirti Dhir, Roman Bagdasarian, Ziwoong Jang
备注：7 pages, 2 tables, and 3 figures. for associated code files, see this https URL

【10】Rate-optimal Design for Anytime Best Arm Identification
标题：随时最佳手臂识别的速率优化设计
链接：https://arxiv.org/abs/2510.23199

作者：Junpei Komiyama, Kyoungseok Jang, Junya Honda

【11】Treble10: A high-quality dataset for far-field speech recognition, dereverberation, and enhancement
标题：Treble10：用于远场语音识别、去回响和增强的高质量数据集
链接：https://arxiv.org/abs/2510.23141

作者：Sarabeth S. Mullins, Georg Götz, Eric Bezzam, Steven Zheng, Daniel Gert Nielsen

表征(1篇)

【1】A Novel Framework for Multi-Modal Protein Representation Learning
标题：多模式蛋白质表示学习的新框架
链接：https://arxiv.org/abs/2510.23273

作者：Runjie Zheng, Zhen Wang, Anjie Qiao, Jiancong Xie, Jiahua Rao, Yuedong Yang
备注：35 pages, 5 figures, 4 tables

3D|3D重建等相关(4篇)

【1】LOC: A General Language-Guided Framework for Open-Set 3D Occupancy Prediction
标题：：开放集3D占用预测的通用地理引导框架
链接：https://arxiv.org/abs/2510.22141

作者：Yuhang Gao, Xiang Xiang, Sheng Zhong, Guoyou Wang

【2】Learning 3D Anisotropic Noise Distributions Improves Molecular Force Field Modeling
标题：学习3D各向异性噪音分布改进分子力场建模
链接：https://arxiv.org/abs/2510.22123

作者：Xixian Liu, Rui Jiao, Zhiyuan Liu, Yurou Liu, Yang Liu, Ziheng Lu, Wenbing Huang, Yang Zhang, Yixin Cao

【3】OpenEM: Large-scale multi-structural 3D datasets for electromagnetic methods
标题：OpenEM：用于电磁方法的大规模多结构3D数据集
链接：https://arxiv.org/abs/2510.21859

作者：Shuang Wang, Xuben Wang, Fei Deng, Peifan Jiang, Jian Chen, Gianluca Fiandaca

【4】Residual-guided AI-CFD hybrid method enables stable and scalable simulations: from 2D benchmarks to 3D applications
标题：残余引导的AI-CFA混合方法实现稳定且可扩展的模拟：从2D基准到3D应用
链接：https://arxiv.org/abs/2510.21804

作者：Shilaj Baral, Youngkyu Lee, Sangam Khanal, Joongoo Jeon

编码器(3篇)

【1】Bi-Encoder Contrastive Learning for Fingerprint and Iris Biometrics
标题：指纹和虹膜生物识别的双编码器对比学习
链接：https://arxiv.org/abs/2510.22937

作者：Matthew So, Judah Goldfeder, Mark Lis, Hod Lipson

【2】COLA: Continual Learning via Autoencoder Retrieval of Adapters
标题：COLA：通过适配器的自动编码器检索进行持续学习
链接：https://arxiv.org/abs/2510.21836

作者：Jaya Krishna Mandivarapu

【3】SITS-DECO: A Generative Decoder Is All You Need For Multitask Satellite Image Time Series Modelling
标题：SITS-DICO：多任务卫星图像时间序列建模所需的生成解码器
链接：https://arxiv.org/abs/2510.21813

作者：Samuel J. Barrett, Docko Sow
备注：27 pages, 7 figures

优化|敛散性(11篇)

【1】Lightweight Robust Direct Preference Optimization
标题：稳健的轻量级直接偏好优化
链接：https://arxiv.org/abs/2510.23590

作者：Cheol Woo Kim, Shresth Verma, Mauricio Tec, Milind Tambe

【2】BBOPlace-Bench: Benchmarking Black-Box Optimization for Chip Placement
标题：BBOPlace-Bench：芯片布局黑匣子优化基准
链接：https://arxiv.org/abs/2510.23472

作者：Ke Xue, Ruo-Tong Chen, Rong-Xi Tan, Xi Lin, Yunqi Shi, Siyuan Xu, Mingxuan Yuan, Chao Qian

【3】Offline Preference Optimization via Maximum Marginal Likelihood Estimation
标题：通过最大边际似然估计进行线下偏好优化
链接：https://arxiv.org/abs/2510.22881

作者：Saeed Najafi, Alona Fyshe

【4】Distributionally Robust Optimization via Diffusion Ambiguity Modeling
标题：基于扩散模糊模型的分布鲁棒优化
链接：https://arxiv.org/abs/2510.22757

作者：Jiaqi Wen, Jianyi Yang

【5】GRPO-Guard: Mitigating Implicit Over-Optimization in Flow Matching via Regulated Clipping
标题：GRPO-Guard：通过调节限幅缓解流量匹配中的隐性过度优化
链接：https://arxiv.org/abs/2510.22319

作者：Jing Wang, Jiajun Liang, Jie Liu, Henglin Liu, Gongye Liu, Jun Zheng, Wanyuan Pang, Ao Ma, Zhenyu Xie, Xintao Wang, Meng Wang, Pengfei Wan, Xiaodan Liang

【6】Visual Model Selection using Feature Importance Clusters in Fairness-Performance Similarity Optimized Space
标题：在公平性-性能相似性优化空间中使用特征重要性集群的视觉模型选择
链接：https://arxiv.org/abs/2510.22209

作者：Sofoklis Kitharidis, Cor J. Veenman, Thomas Bäck, Niki van Stein

【7】Probing Neural Combinatorial Optimization Models
标题：探索神经组合优化模型
链接：https://arxiv.org/abs/2510.22131

作者：Zhiqin Zhang, Yining Ma, Zhiguang Cao, Hoong Chuin Lau
备注：39 pages, 16 figures. Accepted as Spotlight at NeurIPS 2025

【8】FlowOpt: Fast Optimization Through Whole Flow Processes for Training-Free Editing
标题：FlowOpt：通过全流程快速优化，实现免训练编辑
链接：https://arxiv.org/abs/2510.22010

作者：Or Ronai, Vladimir Kulikov, Tomer Michaeli
备注：Project's webpage at this https URL

【9】SynCast: Synergizing Contradictions in Precipitation Nowcasting via Diffusion Sequential Preference Optimization
标题：SynCast：通过扩散序列偏好优化协同降水近播中的矛盾
链接：https://arxiv.org/abs/2510.21847

作者：Kaiyi Xu, Junchao Gong, Wenlong Zhang, Ben Fei, Lei Bai, Wanli Ouyang

【10】Variance-Reduction Guidance: Sampling Trajectory Optimization for Diffusion Models
标题：方差缩减指南：扩散模型的抽样轨迹优化
链接：https://arxiv.org/abs/2510.21792

作者：Shifeng Xu, Yanzhu Liu, Adams Wai-Kin Kong
备注：None

【11】Online Mixture of Experts: No-Regret Learning for Optimal Collective Decision-Making
标题：在线专家混合：无悔学习以实现最佳集体决策
链接：https://arxiv.org/abs/2510.21788

作者：Larkin Liu, Jalal Etesami
备注：39th Conference on Neural Information Processing Systems (NeurIPS 2025)

预测|估计(16篇)

【1】Towards Scaling Deep Neural Networks with Predictive Coding: Theory and Practice
标题：利用预测编码扩展深度神经网络：理论与实践
链接：https://arxiv.org/abs/2510.23323

作者：Francesco Innocenti

【2】Seeing Structural Failure Before it Happens: An Image-Based Physics-Informed Neural Network (PINN) for Spaghetti Bridge Load Prediction
标题：在结构故障发生之前了解情况：基于图像的物理信息神经网络（PINN）用于意大利面桥负载预测
链接：https://arxiv.org/abs/2510.23117

作者：Omer Jauhar Khan, Sudais Khan, Hafeez Anwar
备注：12 pages, 17 figures. Preprint

【3】Sublinear Sketches for Approximate Nearest Neighbor and Kernel Density Estimation
标题：近似最近邻和核密度估计的次线性草图
链接：https://arxiv.org/abs/2510.23039

作者：Ved Danait, Srijan Das, Sujoy Bhore
备注：28 pages, 11 figures

【4】VoMP: Predicting Volumetric Mechanical Property Fields
标题：VoMP：预测体积机械性能场
链接：https://arxiv.org/abs/2510.22975

作者：Rishit Dagli, Donglai Xiang, Vismay Modi, Charles Loop, Clement Fuji Tsang, Anka He Chen, Anita Hu, Gavriel State, David I.W. Levin, Maria Shugrina
备注：hi-res paper and other details at: this https URL

【5】SARNet: A Spike-Aware consecutive validation Framework for Accurate Remaining Useful Life Prediction
标题：SARNet：用于准确剩余使用寿命预测的尖峰感知连续验证框架
链接：https://arxiv.org/abs/2510.22955

作者：Junhao Fan, Wenrui Liang, Wei-Qiang Zhang
备注：5 pages, 2 figures, 3 tables. Equal contribution by Junhao Fan and Wenrui Liang. Corresponding author: Wei-Qiang Zhang. Submitted to ICASSP 2026

【6】Long-Term PM2.5 Forecasting Using a DTW-Enhanced CNN-GRU Model
标题：使用DTW增强CNN-GRU模型进行长期PM2.5预测
链接：https://arxiv.org/abs/2510.22863

作者：Amirali Ataee Naeini, Arshia Ataee Naeini, Fatemeh Karami Mohammadi, Omid Ghaffarpasand
备注：26 pages

【7】A Review of End-to-End Precipitation Prediction Using Remote Sensing Data: from Divination to Machine Learning
标题：利用遥感数据进行端到端降水预测回顾：从占卜到机器学习
链接：https://arxiv.org/abs/2510.22855

作者：Yugong Zeng, Jonathan Wu

【8】Air Quality Prediction Using LOESS-ARIMA and Multi-Scale CNN-BiLSTM with Residual-Gated Attention
标题：使用LOESS-ARIMA和具有剩余门控注意力的多尺度CNN-BiLSTM预测空气质量
链接：https://arxiv.org/abs/2510.22818

作者：Soham Pahari, Sandeep Chand Kumain

【9】NetBurst: Event-Centric Forecasting of Bursty, Intermittent Time Series
标题：NetBurst：以事件为中心的突发性、间歇性时间序列预测
链接：https://arxiv.org/abs/2510.22397

作者：Satyandra Guthula, Jaber Daneshamooz, Charles Fleming, Ashish Kundu, Walter Willinger, Arpit Gupta

【10】The Lossy Horizon: Error-Bounded Predictive Coding for Lossy Text Compression (Episode I)
标题：有损地平线：有损文本压缩的有误差预测编码（第一集）
链接：https://arxiv.org/abs/2510.22207

作者：Nnamdi Aghanya, Jun Li, Kewei Wang
备注：12 pages, 7 figures

【11】Geographic Transferability of Machine Learning Models for Short-Term Airport Fog Forecasting
标题：短期机场雾预测机器学习模型的地理可移植性
链接：https://arxiv.org/abs/2510.21819

作者：Marcelo Cerda Castillo
备注：21 pages, 8 tables, 2 figures. Uses publicly available ERA5 and METAR datasets

【12】A Physics-Guided AI Cascaded Corrector Model Significantly Extends Madden-Julian Oscillation Prediction Skill
标题：物理引导的人工智能级联修正器模型显着扩展了Madden-Julian振荡预测技能
链接：https://arxiv.org/abs/2510.21796

作者：Xiao Zhou, Yuze Sun, Jie Wu, Xiaomeng Huang

【13】A Multi-Component AI Framework for Computational Psychology: From Robust Predictive Modeling to Deployed Generative Dialogue
标题：计算心理学的多组件人工智能框架：从稳健的预测建模到部署的生成对话
链接：https://arxiv.org/abs/2510.21720

作者：Anant Pareek

【14】OEUVRE: OnlinE Unbiased Variance-Reduced loss Estimation
标题：OEUVRE：在线无偏方差-降低损失估计
链接：https://arxiv.org/abs/2510.22744

作者：Kanad Pardeshi, Bryan Wilder, Aarti Singh

【15】Beyond Isotonization: Scalable Non-Crossing Quantile Estimation via Neural Networks for Student Growth Percentiles
标题：超越等序化：通过神经网络对学生成长百分位数进行可扩展的非交叉分位数估计
链接：https://arxiv.org/abs/2510.22419

作者：Kaihua Chang
备注：15 pages, 2 tables, 1 code listing

【16】TraceTrans: Translation and Spatial Tracing for Surgical Prediction
标题：TraceTrans：手术预测的翻译和空间跟踪
链接：https://arxiv.org/abs/2510.22379

作者：Xiyu Luo, Haodong LI, Xinxing Cheng, He Zhao, Yang Hu, Xuan Song, Tianyang Zhang

其他神经网络|深度学习|模型|建模(56篇)

【1】Variational Masked Diffusion Models
标题：变分掩蔽扩散模型
链接：https://arxiv.org/abs/2510.23606

作者：Yichi Zhang, Alex Schwing, Zhizhen Zhao
备注：Project Page: this https URL

【2】Learning Linearity in Audio Consistency Autoencoders via Implicit Regularization
标题：基于隐式正则化的音频一致性自编码器线性学习
链接：https://arxiv.org/abs/2510.23530

作者：Bernardo Torres, Manuel Moussallam, Gabriel Meseguer-Brocal

【3】Toward Carbon-Neutral Human AI: Rethinking Data, Computation, and Learning Paradigms for Sustainable Intelligence
标题：迈向碳中性人类人工智能：重新思考可持续智能的数据，计算和学习范式
链接：https://arxiv.org/abs/2510.23524

作者：KC Santosh, Rodrigue Rizk, Longwei Wang
备注：9 pages, 3 figures

【4】Towards Deep Physics-Informed Kolmogorov-Arnold Networks
标题：迈向深度物理知情的Kolmogorov-Arnold网络
链接：https://arxiv.org/abs/2510.23501

作者：Spyros Rigas, Fotios Anagnostopoulos, Michalis Papachristou, Georgios Alexandridis
备注：73 pages, 22 figures

【5】PrivacyGuard: A Modular Framework for Privacy Auditing in Machine Learning
标题：PrivacyGuard：机器学习中隐私审计的模块化框架
链接：https://arxiv.org/abs/2510.23427

作者：Luca Melis, Matthew Grange, Iden Kalemaj, Karan Chadha, Shengyuan Hu, Elena Kashtelyan, Will Bullock

【6】Floating-Point Neural Network Verification at the Software Level
标题：软件级的浮点神经网络验证
链接：https://arxiv.org/abs/2510.23389

作者：Edoardo Manino, Bruno Farias, Rafael Sá Menezes, Fedor Shmarov, Lucas C. Cordeiro
备注：Pre-print before submission to peer review

【7】ZeroFlood: A Geospatial Foundation Model for Data-Efficient Flood Susceptibility Mapping
标题：ZeroFlood：数据高效的洪水易感性绘制的地理空间基础模型
链接：https://arxiv.org/abs/2510.23364

作者：Hyeongkyun Kim, Orestis Oikonomou
备注：Preprint submitted to EUSAR 2026 (under review)

【8】Robust Iterative Learning Hidden Quantum Markov Models
标题：鲁棒迭代学习隐量子马尔可夫模型
链接：https://arxiv.org/abs/2510.23237

作者：Ning Ning
备注：Quantum Computing, Bayesian Inference, Spatiotemporal Analysis, Robust Learning

【9】The Benchmarking Epistemology: Construct Validity for Evaluating Machine Learning Models
标题：基准认识论：评估机器学习模型的结构有效性
链接：https://arxiv.org/abs/2510.23191

作者：Timo Freiesleben, Sebastian Zezulka

【10】Neural Emulator Superiority: When Machine Learning for PDEs Surpasses its Training Data
标题：神经模拟器优势：当用于PDEs的机器学习超过其训练数据时
链接：https://arxiv.org/abs/2510.23111

作者：Felix Koehler, Nils Thuerey
备注：Accepted at NeurIPS 2025: this https URL

【11】Smaller Models, Smarter Rewards: A Two-Sided Approach to Process and Outcome Rewards
标题：更小的模型，更聪明的奖励：流程和结果奖励的双向方法
链接：https://arxiv.org/abs/2510.23083

作者：Jan Niklas Groeneveld, Xi Qin, Alexander Schaefer, Yaad Oren
备注：Accepted and to be presented at NeurIPS 2025 Workshop: Foundations of Reasoning in Language Models

【12】Equivariant Neural Networks for General Linear Symmetries on Lie Algebras
标题：李代数上一般线性对称的等变神经网络
链接：https://arxiv.org/abs/2510.22984

作者：Chankyo Kim (1), Sicheng Zhao (1), Minghan Zhu (1 and 2), Tzu-Yuan Lin (3), Maani Ghaffari (1) ((1) University of Michigan, (2) University of Pennsylvania, (3) Massachusetts Institute of Technology)
备注：23 pages, 5 figures

【13】Hankel Singular Value Regularization for Highly Compressible State Space Models
标题：高度可压缩状态空间模型的Hankel奇异值正则化
链接：https://arxiv.org/abs/2510.22951

作者：Paul Schwerdtner, Jules Berman, Benjamin Peherstorfer
备注：Accepted at NeurIPS 2025

【14】On the Anisotropy of Score-Based Generative Models
标题：基于分数的生成模型的各向异性
链接：https://arxiv.org/abs/2510.22899

作者：Andreas Floros, Seyed-Mohsen Moosavi-Dezfooli, Pier Luigi Dragotti

【15】Distributed Multi-Agent Bandits Over Erdős-Rényi Random Networks
标题：Erdjells-Rényi随机网络上的分布式多智能体盗贼
链接：https://arxiv.org/abs/2510.22811

作者：Jingyuan Liu, Hao Qiu, Lin Yang, Mengfan Xu

【16】Centrum: Model-based Database Auto-tuning with Minimal Distributional Assumptions
标题：Centrum：基于模型的数据库自动调整，具有最小分布假设
链接：https://arxiv.org/abs/2510.22734

作者：Yuanhao Lai, Pengfei Zheng, Chenpeng Ji, Yan Li, Songhan Zhang, Rutao Zhang, Zhengang Wang, Yunfei Du
备注：26 pages

【17】UCB-type Algorithm for Budget-Constrained Expert Learning
标题：预算约束专家学习的UCB型算法
链接：https://arxiv.org/abs/2510.22654

作者：Ilgam Latypov, Alexandra Suvorikova, Alexey Kroshnin, Alexander Gasnikov, Yuriy Dorn

【18】A Framework for Quantifying How Pre-Training and Context Benefit In-Context Learning
标题：量化预训练和上下文如何有利于上下文学习的框架
链接：https://arxiv.org/abs/2510.22594

作者：Bingqing Song, Jiaxiang Li, Rong Wang, Songtao Lu, Mingyi Hong

【19】Approximate Gradient Coding for Distributed Learning with Heterogeneous Stragglers
标题：具有异类离散器的分布式学习的近似梯度编码
链接：https://arxiv.org/abs/2510.22539

作者：Heekang Song, Wan Choi

【20】Smart Sensor Placement: A Correlation-Aware Attribution Framework (CAAF) for Real-world Data Modeling
标题：智能传感器放置：用于现实世界数据建模的相关性感知归因框架（CAAF）
链接：https://arxiv.org/abs/2510.22517

作者：Sze Chai Leung, Di Zhou, H. Jane Bae

【21】Transitive RL: Value Learning via Divide and Conquer
标题：传递性RL：通过分而治之的价值学习
链接：https://arxiv.org/abs/2510.22512

作者：Seohong Park, Aditya Oberai, Pranav Atreya, Sergey Levine

【22】CANDI: Hybrid Discrete-Continuous Diffusion Models
标题：CANDI：混合离散-连续扩散模型
链接：https://arxiv.org/abs/2510.22510

作者：Patrick Pynadath, Jiaxin Shi, Ruqi Zhang

【23】Learning Local Stackelberg Equilibria from Repeated Interactions with a Learning Agent
标题：从与学习代理的重复交互中学习本地Stackelberg均衡
链接：https://arxiv.org/abs/2510.22471

作者：Nivasini Ananthakrishnan, Yuval Dagan, Kunhe Yang

【24】Knowledge-guided Continual Learning for Behavioral Analytics Systems
标题：行为分析系统的知识引导持续学习
链接：https://arxiv.org/abs/2510.22405

作者：Yasas Senarath, Hemant Purohit
备注：This is a preprint of the accepted author manuscript that has been accepted for publication at IEEE CogMI 2025 - The 7th IEEE International Conference on Cognitive Machine Intelligence

【25】Dynamic Dropout: Leveraging Conway's Game of Life for Neural Networks Regularization
标题：动态辍学：利用康威的生命游戏进行神经网络正规化
链接：https://arxiv.org/abs/2510.22383

作者：David Freire-Obregón, José Salas-Cáceres, Modesto Castrillón-Santana
备注：Accepted for presentation at the 5th International Conference on Computing and Machine Intelligence (ICMI 2026)

【26】Stable neural networks and connections to continuous dynamical systems
标题：稳定的神经网络和与连续动态系统的连接
链接：https://arxiv.org/abs/2510.22299

作者：Matthias J. Ehrhardt, Davide Murari, Ferdia Sherry

【27】Machine Learning Enabled Early Warning System For Financial Distress Using Real-Time Digital Signals
标题：使用实时数字信号的机器学习支持财务困境预警系统
链接：https://arxiv.org/abs/2510.22287

作者：Laxmi pant, Syed Ali Reza, Md Khalilor Rahman, MD Saifur Rahman, Shamima Sharmin, Md Fazlul Huq Mithu, Kazi Nehal Hasnain, Adnan Farabi, Mahamuda khanom, Raisul Kabir

【28】Epistemic Deep Learning: Enabling Machine Learning Models to Know When They Do Not Know
标题：认识深度学习：使机器学习模型能够在不知道的情况下知道
链接：https://arxiv.org/abs/2510.22261

作者：Shireen Kudukkil Manchingal

【29】Dopamine-driven synaptic credit assignment in neural networks
标题：神经网络中多巴胺驱动的突触信用分配
链接：https://arxiv.org/abs/2510.22178

作者：Saranraj Nambusubramaniyan, Shervin Safavi, Raja Guru, Andreas Knoblauch

【30】Tractable Shapley Values and Interactions via Tensor Networks
标题：通过张量网络的易驾驭Shapley价值观和相互作用
链接：https://arxiv.org/abs/2510.22138

作者：Farzaneh Heidari, Chao Li, Farzaneh Heidari

【31】Automatic Assessment of Students' Classroom Engagement with Bias Mitigated Multi-task Model
标题：利用偏见缓解多任务模型自动评估学生课堂参与度
链接：https://arxiv.org/abs/2510.22057

作者：James Thiering, Tarun Sethupat Radha Krishna, Dylan Zelkin, Ashis Kumer Biswas
备注：13 pages, 12 figures, and 1 table

【32】Energy-Efficient Domain-Specific Artificial Intelligence Models and Agents: Pathways and Paradigms
标题：节能领域特定人工智能模型和代理：路径和范式
链接：https://arxiv.org/abs/2510.22052

作者：Abhijit Chatterjee, Niraj K. Jha, Jonathan D. Cohen, Thomas L. Griffiths, Hongjing Lu, Diana Marculescu, Ashiqur Rasul, Keshab K. Parhi

【33】Towards Error-Centric Intelligence II: Energy-Structured Causal Models
标题：迈向以错误为中心的智力II：能量结构因果模型
链接：https://arxiv.org/abs/2510.22050

作者：Marcus Thomas

【34】Generalized Top-k Mallows Model for Ranked Choices
标题：排序选择的广义Top-k Malows模型
链接：https://arxiv.org/abs/2510.22040

作者：Shahrzad Haddadan, Sara Ahmadian

【35】K-DAREK: Distance Aware Error for Kurkova Kolmogorov Networks
标题：K-DAREK：Kurkova Kolmogorov网络的距离感知误差
链接：https://arxiv.org/abs/2510.22021

作者：Masoud Ataei, Vikas Dhiman, Mohammad Javad Khojasteh
备注：Accepted at IEEE ACSSC, 9 pages and 3 figures

【36】An Introductory Guide to Koopman Learning
标题：库普曼学习入门指南
链接：https://arxiv.org/abs/2510.22002

作者：Matthew J. Colbrook, Zlatko Drmač, Andrew Horning

【37】From Black-box to Causal-box: Towards Building More Interpretable Models
标题：从黑箱到计算机箱：构建更可解释的模型
链接：https://arxiv.org/abs/2510.21998

作者：Inwoo Hwang, Yushu Pan, Elias Bareinboim
备注：NeurIPS 2025

【38】Is Temporal Difference Learning the Gold Standard for Stitching in RL?
标题：时间差异学习是RL缝合的黄金标准吗？
链接：https://arxiv.org/abs/2510.21995

作者：Michał Bortkiewicz, Władysław Pałucki, Mateusz Ostaszewski, Benjamin Eysenbach
备注：The first two authors contributed equally. Project website: this https URL

【39】Deep Jump Gaussian Processes for Surrogate Modeling of High-Dimensional Piecewise Continuous Functions
标题：用于多维分段连续函数替代建模的深跳高斯过程
链接：https://arxiv.org/abs/2510.21974

作者：Yang Xu, Chiwoo Park

【40】Parallel Sampling from Masked Diffusion Models via Conditional Independence Testing
标题：通过条件独立性测试对掩蔽扩散模型进行平行抽样
链接：https://arxiv.org/abs/2510.21961

作者：Iskander Azangulov, Teodora Pandeva, Niranjani Prasad, Javier Zazo, Sushrut Karmalkar

【41】Generalization Bounds for Rank-sparse Neural Networks
标题：排序稀疏神经网络的广义界
链接：https://arxiv.org/abs/2510.21945

作者：Antoine Ledent, Rodrigo Alves, Yunwen Lei
备注：Accepted at NeurIPS 2025

【42】The Principles of Diffusion Models
标题：扩散模型的原则
链接：https://arxiv.org/abs/2510.21890

作者：Chieh-Hsin Lai, Yang Song, Dongjun Kim, Yuki Mitsufuji, Stefano Ermon

【43】Privacy-preserving Decision-focused Learning for Multi-energy Systems
标题：多能源系统的隐私保护以决策为中心的学习
链接：https://arxiv.org/abs/2510.21858

作者：Yangze Zhou, Ruiyang Yao, Dalin Qin, Yixiong Jia, Yi Wang
备注：10 pages, 7 figures

【44】It Takes Two to Tango: Two Parallel Samplers Improve Quality in Diffusion Models for Limited Steps
标题：探戈需要两个人：两个并行采样器在有限步骤内提高扩散模型的质量
链接：https://arxiv.org/abs/2510.21802

作者：Pedro Cisneros-Velarde

【45】Direct Debiased Machine Learning via Bregman Divergence Minimization
标题：通过Bregman分歧最小化直接去偏机器学习
链接：https://arxiv.org/abs/2510.23534

作者：Masahiro Kato

【46】The First Star-by-star $N$-body/Hydrodynamics Simulation of Our Galaxy Coupling with a Surrogate Model
标题：与代理模型相结合的银河系第一个逐星$N$-体/流体动力学模拟
链接：https://arxiv.org/abs/2510.23330

作者：Keiya Hirashima, Michiko S. Fujii, Takayuki R. Saitoh, Naoto Harada, Kentaro Nomura, Kohji Yoshikawa, Yutaka Hirai, Tetsuro Asano, Kana Moriwaki, Masaki Iwasawa, Takashi Okamoto, Junichiro Makino
备注：12 pages, 7 figures, 7 tables, IEEE/ACM Supercomputing Conference (SC25)

【47】Provable test-time adaptivity and distributional robustness of in-context learning
标题：可证明的测试时自适应性和上下文学习的分布鲁棒性
链接：https://arxiv.org/abs/2510.23254

作者：Tianyi Ma, Tengyao Wang, Richard J. Samworth
备注：44 pages

【48】Physics-informed diffusion models for extrapolating crystal structures beyond known motifs
标题：用于外推已知基序以外的晶体结构的物理信息扩散模型
链接：https://arxiv.org/abs/2510.23181

作者：Andrij Vasylenko, Federico Ottomano, Christopher M. Collins, Rahul Savani, Matthew S. Dyer, Matthew J. Rosseinsky

【49】Complexity Dependent Error Rates for Physics-informed Statistical Learning via the Small-ball Method
标题：通过小球方法实现基于物理信息的统计学习的复杂性相关错误率
链接：https://arxiv.org/abs/2510.23149

作者：Diego Marcondes

【50】AQCat25: Unlocking spin-aware, high-fidelity machine learning potentials for heterogeneous catalysis
标题：AQCat 25：释放用于异类催化的旋转感知、高保真机器学习潜力
链接：https://arxiv.org/abs/2510.22938

作者：Omar Allam, Brook Wander, Aayush R. Singh
备注：32 pages, 17 figures

【51】Block Coordinate Descent for Neural Networks Provably Finds Global Minima
标题：神经网络块坐标下降可证明找到全局极小值
链接：https://arxiv.org/abs/2510.22667

作者：Shunta Akiyama
备注：32 pages, 4 figures

【52】Multi-Modal Masked Autoencoders for Learning Image-Spectrum Associations for Galaxy Evolution and Cosmology
标题：多模态掩蔽自编码器用于星系演化和宇宙学的图像-频谱关联学习
链接：https://arxiv.org/abs/2510.22527

作者：Morgan Himes, Samiksha Krishnamurthy, Andrew Lizarraga, Srinath Saikrishnan, Vikram Seenivasan, Jonathan Soriano, Ying Nian Wu, Tuan Do
备注：8 pages, 3 figures, 1 table, accepted to NeurIPS 2025 Workshop ML4PS

【53】Bridging the Perceptual - Statistical Gap in Dysarthria Assessment: Why Machine Learning Still Falls Short
标题：弥合味觉障碍评估中的统计差距：为什么机器学习仍然落后
链接：https://arxiv.org/abs/2510.22237

作者：Krishna Gurugubelli

【54】HPC-Driven Modeling with ML-Based Surrogates for Magnon-Photon Dynamics in Hybrid Quantum Systems
标题：混合量子系统中磁振子-子动力学的基于ML的MPC驱动建模
链接：https://arxiv.org/abs/2510.22221

作者：Jialin Song, Yingheng Tang, Pu Ren, Shintaro Takayoshi, Saurabh Sawant, Yujie Zhu, Jia-Mian Hu, Andy Nonaka, Michael W. Mahoney, Benjamin Erichson, Zhi (Jackie)Yao

【55】MMbeddings: Parameter-Efficient, Low-Overfitting Probabilistic Embeddings Inspired by Nonlinear Mixed Models
标题：MMbeddings：受非线性混合模型启发的参数高效、低过拟的概率嵌入
链接：https://arxiv.org/abs/2510.22198

作者：Giora Simchoni, Saharon Rosset

【56】Frequency-Spatial Interaction Driven Network for Low-Light Image Enhancement
标题：用于弱光图像增强的频率-空间交互驱动网络
链接：https://arxiv.org/abs/2510.22154

作者：Yunhong Tao, Wenbing Tao, Xiang Xiang

其他(71篇)

【1】Lookahead Anchoring: Preserving Character Identity in Audio-Driven Human Animation
标题：前瞻锚定：在音频驱动的人类动画中保留角色身份
链接：https://arxiv.org/abs/2510.23581

作者：Junyoung Seo, Rodrigo Mira, Alexandros Haliassos, Stella Bounareli, Honglie Chen, Linh Tran, Seungryong Kim, Zoe Landgraf, Jie Shen
备注：Project page: this https URL

【2】RobotArena $\infty$: Scalable Robot Benchmarking via Real-to-Sim Translation
链接：https://arxiv.org/abs/2510.23571

作者：Yash Jangir, Yidi Zhang, Kashu Yamazaki, Chenyu Zhang, Kuan-Hsun Tu, Tsung-Wei Ke, Lei Ke, Yonatan Bisk, Katerina Fragkiadaki
备注：Website: this https URL

【3】ReCode: Unify Plan and Action for Universal Granularity Control
标题：ReCode：统一通用粒度控制的计划和行动
链接：https://arxiv.org/abs/2510.23564

作者：Zhaoyang Yu, Jiayi Zhang, Huixue Su, Yufan Zhao, Yifan Wu, Mingyi Deng, Jinyu Xiang, Yizhang Lin, Lingxiao Tang, Yingchao Li, Yuyu Luo, Bang Liu, Chenglin Wu

【4】Sequential Multi-Agent Dynamic Algorithm Configuration
标题：顺序多代理动态算法配置
链接：https://arxiv.org/abs/2510.23535

作者：Chen Lu, Ke Xue, Lei Yuan, Yao Wang, Yaoyuan Wang, Sheng Fu, Chao Qian
备注：NeurIPS 2025

【5】Mixed Precision Training of Neural ODEs
标题：神经ODE的混合精确训练
链接：https://arxiv.org/abs/2510.23498

作者：Elena Celledoni, Brynjulf Owren, Lars Ruthotto, Tianjiao Nicole Yang
备注：Code available at this https URL 26 pages, 4 figures

【6】Eigen-Value: Efficient Domain-Robust Data Valuation via Eigenvalue-Based Approach
标题：特征值：基于特征值的有效域鲁棒数据估值方法
链接：https://arxiv.org/abs/2510.23409

作者：Youngjun Choi, Joonseong Kang, Sungjun Lim, Kyungwoo Song

【7】Opinion Mining Based Entity Ranking using Fuzzy Logic Algorithmic Approach
标题：基于模糊逻辑算术方法的意见挖掘实体排名
链接：https://arxiv.org/abs/2510.23384

作者：Pratik N. Kalamkar, A.G. Phakatkar
备注：8 pages, 4 figures, Conference Paper

【8】Towards a Generalizable AI for Materials Discovery: Validation through Immersion Coolant Screening
标题：迈向材料发现的可通用人工智能：通过沉浸式收件箱筛选进行验证
链接：https://arxiv.org/abs/2510.23371

作者：Hyunseung Kim, Dae-Woong Jeong, Changyoung Park, Won-Ji Lee, Ha-Eun Lee, Ji-Hye Lee, Rodrigo Hormazabal, Sung Moon Ko, Sumin Lee, Soorin Yim, Chanhui Lee, Sehui Han, Sang-Ho Cha, Woohyung Lim
备注：16 pages, 4 figures

【9】Robust Non-negative Proximal Gradient Algorithm for Inverse Problems
标题：反问题的鲁棒非负近梯度算法
链接：https://arxiv.org/abs/2510.23362

作者：Hanzhang Wang, Zonglin Liu, Jingyi Xu, Chenyang Wang, Zhiwei Zhong, Qiangqiang Shen

【10】Block-Diagonal LoRA for Eliminating Communication Overhead in Tensor Parallel LoRA Serving
标题：块对角LoRA消除张量并行LoRA服务中的通信负担
链接：https://arxiv.org/abs/2510.23346

作者：Xinyu Wang, Jonas M. Kübler, Kailash Budhathoki, Yida Wang, Matthäus Kleindessner

【11】Toward Interpretable Evaluation Measures for Time Series Segmentation
标题：走向时间序列分割的可解释评估措施
链接：https://arxiv.org/abs/2510.23261

作者：Félix Chavelli, Paul Boniol, Michaël Thomazo

【12】Rethinking GSPO: The Perplexity-Entropy Equivalence
标题：重新思考GSPO：困惑-熵等效
链接：https://arxiv.org/abs/2510.23142

作者：Chi Liu
备注：10 pages, 2 figures

【13】Sampling from Energy distributions with Target Concrete Score Identity
标题：从具有目标混凝土分数同一性的能量分布中进行抽样
链接：https://arxiv.org/abs/2510.23106

作者：Sergei Kholkin, Francisco Vargas, Alexander Korotin

【14】Advantage Shaping as Surrogate Reward Maximization: Unifying Pass@K Policy Gradients
标题：代理人奖励最大化的优势塑造：统一Pass@K政策推动者
链接：https://arxiv.org/abs/2510.23049

作者：Christos Thrampoulidis, Sadegh Mahdavi, Wenlong Deng

【15】Softmax is $1/2$-Lipschitz: A tight bound across all $\ell_p$ norms
链接：https://arxiv.org/abs/2510.23012

作者：Pravin Nair
备注：Under review

【16】How Muon's Spectral Design Benefits Generalization: A Study on Imbalanced Data
标题：Muon的光谱设计如何有利于推广：不平衡数据的研究
链接：https://arxiv.org/abs/2510.22980

作者：Bhavya Vasudeva, Puneesh Deora, Yize Zhao, Vatsal Sharan, Christos Thrampoulidis
备注：32 pages, 28 figures

【17】Manifold Approximation leads to Robust Kernel Alignment
标题：总管逼近导致稳健的核对齐
链接：https://arxiv.org/abs/2510.22953

作者：Mohammad Tariqul Islam, Du Liu, Deblina Sarkar
备注：9 pages, 5 figures + supplementary

【18】Hazard-Responsive Digital Twin for Climate-Driven Urban Resilience and Equity
标题：应对危险的数字双胞胎，实现气候驱动的城市韧性和公平
链接：https://arxiv.org/abs/2510.22941

作者：Zhenglai Shen, Hongyu Zhou
备注：52 pages, 9 figures

【19】Limits of Generative Pre-Training in Structured EMR Trajectories with Irregular Sampling
标题：不规则抽样的结构化电子病历轨迹生成预训练的局限性
链接：https://arxiv.org/abs/2510.22878

作者：Nicholas I-Hsien Kuo, Blanca Gallego, Louisa Jorm

【20】Toward Agents That Reason About Their Computation
标题：走向为其计算推理的代理
链接：https://arxiv.org/abs/2510.22833

作者：Adrian Orenstein, Jessica Chen, Gwyneth Anne Delos Santos, Bayley Sapara, Michael Bowling

【21】FairJudge: MLLM Judging for Social Attributes and Prompt Image Alignment
标题：FairJudge：MLLM对社会属性和快速图像对齐的判断
链接：https://arxiv.org/abs/2510.22827

作者：Zahraa Al Sahili, Maryam Fetanat, Maimuna Nowaz, Ioannis Patras, Matthew Purver

【22】Last Iterate Analyses of FTRL in Stochasitc Bandits
标题：随机盗贼FTRL的最后迭代分析
链接：https://arxiv.org/abs/2510.22819

作者：Jingxin Zhan, Yuze Han, Zhihua Zhang

【23】SAO-Instruct: Free-form Audio Editing using Natural Language Instructions
标题：SAO-Instruct：使用自然语言指令进行自由格式音频编辑
链接：https://arxiv.org/abs/2510.22795

作者：Michael Ungersböck, Florian Grötschla, Luca A. Lanzendörfer, June Young Yi, Changho Choi, Roger Wattenhofer
备注：Accepted at NeurIPS 2025

【24】SeeDNorm: Self-Rescaled Dynamic Normalization
标题：SeeDNorm：自我重新调整的动态规范化
链接：https://arxiv.org/abs/2510.22777

作者：Wenrui Cai, Defa Zhu, Qingjie Liu, Qiyang Min

【25】ATLAS: Actor-Critic Task-Completion with Look-ahead Action Simulation
标题：ATLAS：通过前瞻动作模拟完成演员评论家任务
链接：https://arxiv.org/abs/2510.22732

作者：Jiali Cheng, Anjishnu Kumar, Roshan Lal, Rishi Rajasekaran, Hani Ramezani, Omar Zia Khan, Oleg Rokhlenko, Sunny Chiu-Webster, Gang Hua, Hadi Amiri
备注：9 pages, NeurIPS 2025 Workshop on Language Agents and World Models

【26】S-Chain: Structured Visual Chain-of-Thought For Medicine
标题：S-Chain：医学的结构化视觉思维链
链接：https://arxiv.org/abs/2510.22728

作者：Khai Le-Duc, Duy M. H. Nguyen, Phuong T. H. Trinh, Tien-Phat Nguyen, Nghiem T. Diep, An Ngo, Tung Vu, Trinh Vuong, Anh-Tien Nguyen, Mau Nguyen, Van Trung Hoang, Khai-Nguyen Nguyen, Hy Nguyen, Chris Ngo, Anji Liu, Nhat Ho, Anne-Christin Hauschild, Khanh Xuan Nguyen, Thanh Nguyen-Tang, Pengtao Xie, Daniel Sonntag, James Zou, Mathias Niepert, Anh Totti Nguyen
备注：First version

【27】If You Want to Be Robust, Be Wary of Initialization
标题：如果你想变得坚强，就要小心失败
链接：https://arxiv.org/abs/2510.22652

作者：Sofiane Ennadir, Johannes F. Lutzeyer, Michalis Vazirgiannis, El Houcine Bergou
备注：Accepted at NeurIPS 2024

【28】Variational Polya Tree
标题：变异波利亚树
链接：https://arxiv.org/abs/2510.22651

作者：Lu Xu, Tsai Hor Chan, Kwok Fai Lam, Lequan Yu, Guosheng Yin

【29】Environment-aware Motion Matching
标题：环保运动匹配
链接：https://arxiv.org/abs/2510.22632

作者：Jose Luis Ponton, Sheldon Andrews, Carlos Andujar, Nuria Pelechano
备注：Published in ACM TOG and presented in SIGGRAPH ASIA 2025. Project webpage: this https URL

【30】Multi-Scale Finite Expression Method for PDEs with Oscillatory Solutions on Complex Domains
标题：复杂区域上具有振动解的偏头痛方程的多尺度有限表达方法
链接：https://arxiv.org/abs/2510.22497

作者：Gareth Hardwick, Haizhao Yang

【31】Low-Precision Streaming PCA
标题：低精度流媒体PCA
链接：https://arxiv.org/abs/2510.22440

作者：Sanjoy Dasgupta, Syamantak Kumar, Shourya Pandey, Purnamrita Sarkar

【32】Bias Begins with Data: The FairGround Corpus for Robust and Reproducible Research on Algorithmic Fairness
标题：偏见始于数据：对数学公平性进行稳健且可重复研究的FairGround数据库
链接：https://arxiv.org/abs/2510.22363

作者：Jan Simson, Alessandro Fabris, Cosima Fröhner, Frauke Kreuter, Christoph Kern
备注：Website: this https URL

【33】Monitoring State Transitions in Markovian Systems with Sampling Cost
标题：具有抽样成本的马尔科夫系统中的状态转变监控
链接：https://arxiv.org/abs/2510.22327

作者：Kumar Saurav, Ness B. Shroff, Yingbin Liang
备注：6 pages, 4 figures

【34】Adapting Noise-Driven PUF and AI for Secure WBG ICS: A Proof-of-Concept Study
标题：调整噪音驱动的PFA和人工智能以实现安全WBG ICS：概念验证研究
链接：https://arxiv.org/abs/2510.22283

作者：Devon A. Kelly, Christiana Chamon

【35】You Don't Need Prompt Engineering Anymore: The Prompting Inversion
标题：你不再需要快速工程：预算倒置
链接：https://arxiv.org/abs/2510.22251

作者：Imran Khan (Independent Researcher)
备注：17 pages, 1 figure, 6 tables. Code and experimental data available at this https URL

【36】Taming Silent Failures: A Framework for Verifiable AI Reliability
标题：驯服无声故障：可验证人工智能可靠性的框架
链接：https://arxiv.org/abs/2510.22224

作者：Guan-Yan Yang, Farn Wang
备注：This preprint has been accepted by IEEE Reliability Magazine. 10 pages, 3 figures

【37】Multi-dataset Joint Pre-training of Emotional EEG Enables Generalizable Affective Computing
标题：多数据集情感联合预训练实现脑电可推广情感计算
链接：https://arxiv.org/abs/2510.22197

作者：Qingzhu Zhang, Jiani Zhong, Zongsheng Li, Xinke Shen, Quanying Liu

【38】Quantitative Bounds for Sorting-Based Permutation-Invariant Embeddings
标题：基于排序的排列不变嵌入的量化界限
链接：https://arxiv.org/abs/2510.22186

作者：Nadav Dym, Matthias Wellershoff, Efstratios Tsoukanis, Daniel Levy, Radu Balan
备注：26 pages, 1 figure, 2 tables

【39】HandPass: A Wi-Fi CSI Palm Authentication Approach for Access Control
标题：HandPass：一种用于访问控制的Wi-Fi SI Palm认证方法
链接：https://arxiv.org/abs/2510.22133

作者：Eduardo Fabricio Gomes Trindade, Felipe Silveira de Almeida, Gioliano de Oliveira Braga, Rafael Pimenta de Mattos Paixão, Pedro Henrique dos Santos Rocha, Lourenco Alves Pereira Jr
备注：9 pages, 4 figures, 3 tables

【40】Neural Index Policies for Restless Multi-Action Bandits with Heterogeneous Budgets
标题：预算不均匀的不安多行动盗贼的神经指数策略
链接：https://arxiv.org/abs/2510.22069

作者：Himadri S. Pandey, Kai Wang, Gian-Gabriel P. Garcia

【41】Deep Gaussian Processes for Functional Maps
标题：功能地图的深高斯过程
链接：https://arxiv.org/abs/2510.22068

作者：Matthew Lowery, Zhitong Xu, Da Long, Keyan Chen, Daniel S. Johnson, Yang Bai, Varun Shankar, Shandian Zhe
备注：10 pages + 9 page appendix, 5 figures

【42】Massive Memorization with Hundreds of Trillions of Parameters for Sequential Transducer Generative Recommenders
标题：具有数百万亿个参数的序列传感器生成推荐器的大规模并行化
链接：https://arxiv.org/abs/2510.22049

作者：Zhimin Chen, Chenyu Zhao, Ka Chun Mo, Yunjiang Jiang, Jane H. Lee, Shouwei Chen, Khushhall Chandra Mahajan, Ning Jiang, Kai Ren, Jinhui Li, Wen-Yun Yang

【43】Fast Non-Log-Concave Sampling under Nonconvex Equality and Inequality Constraints with Landing
标题：带着陆的非凸等式和不等式约束下的快速非log-凹凸采样
链接：https://arxiv.org/abs/2510.22044

作者：Kijung Jeon, Michael Muehlebach, Molei Tao
备注：62 pages

【44】Differentiable Constraint-Based Causal Discovery
标题：基于可区分约束的因果发现
链接：https://arxiv.org/abs/2510.22031

作者：Jincheng Zhou, Mengbo Wang, Anqi He, Yumeng Zhou, Hessam Olya, Murat Kocaoglu, Bruno Ribeiro

【45】Normalization in Attention Dynamics
标题：注意力动态的正常化
链接：https://arxiv.org/abs/2510.22026

作者：Nikita Karagodin, Shu Ge, Yury Polyanskiy, Philippe Rigollet
备注：39th Conference on Neural Information Processing Systems (NeurIPS 2025), 23 pages

【46】A Multimodal Human Protein Embeddings Database: DeepDrug Protein Embeddings Bank (DPEB)
标题：多模式人类蛋白质嵌入数据库：DeepDrug Protein Embeddings Bank（DPEB）
链接：https://arxiv.org/abs/2510.22008

作者：Md Saiful Islam Sajol, Magesh Rajasekaran, Hayden Gemeinhardt, Adam Bess, Chris Alvin, Supratik Mukhopadhyay

【47】Revisiting Orbital Minimization Method for Neural Operator Decomposition
标题：重新审视神经运算符分解的轨道最小化方法
链接：https://arxiv.org/abs/2510.21952

作者：J. Jon Ryu, Samuel Zhou, Gregory W. Wornell
备注：25 pages, 8 figures. To appear at NeurIPS 2025

【48】AutoSciDACT: Automated Scientific Discovery through Contrastive Embedding and Hypothesis Testing
标题：AutoSciDACT：通过对比嵌入和假设测试自动化科学发现
链接：https://arxiv.org/abs/2510.21935

作者：Samuel Bright-Thonney, Christina Reissel, Gaia Grosso, Nathaniel Woodward, Katya Govorkova, Andrzej Novak, Sang Eon Park, Eric Moreno, Philip Harris
备注：Accepted at NeurIPS 2025; 32 pages, 16 figures

【49】SIGN: Schema-Induced Games for Naming
标题：SIGN：收件箱的阴谋引发游戏
链接：https://arxiv.org/abs/2510.21855

作者：Ryan Zhang, Herbert Woisetscläger
备注：AAAI 2026 Student Abstract (Oral). Code available ar this https URL

【50】Data-Driven Approach to Capitation Reform in Rwanda
标题：卢旺达以数据驱动的方式进行配额改革
链接：https://arxiv.org/abs/2510.21851

作者：Babaniyi Olaniyi, Ina Kalisa, Ana Fernández del Río, Jean Marie Vianney Hakizayezu, Enric Jané, Eniola Olaleye, Juan Francisco Garamendi, Ivan Nazarov, Aditya Rastogi, Mateo Diaz-Quiroz, África Periáñez, Regis Hitimana

【51】Beyond Point Matching: Evaluating Multiscale Dubuc Distance for Time Series Similarity
标题：超越点匹配：评估多尺度Dubuc距离的时间序列相似性
链接：https://arxiv.org/abs/2510.21824

作者：Azim Ahmadzadeh, Mahsa Khazaei, Elaina Rohlfing

【52】VITA-E: Natural Embodied Interaction with Concurrent Seeing, Hearing, Speaking, and Acting
标题：VIA-E：自然的互动，同时看、听、说和表演
链接：https://arxiv.org/abs/2510.21817

作者：Xiaoyu Liu, Chaoyou Fu, Chi Yan, Chu Wu, Haihan Gao, Yi-Fan Zhang, Shaoqi Dong, Cheng Qian, Bin Luo, Xiuyong Yang, Guanwu Li, Yusheng Cai, Yunhang Shen, Deqiang Jiang, Haoyu Cao, Xing Sun, Caifeng Shan, Ran He
备注：Homepage: this https URL

【53】MARS-M: When Variance Reduction Meets Matrices
标题：MARS-M：当方差缩减满足矩阵时
链接：https://arxiv.org/abs/2510.21800

作者：Yifeng Liu, Angela Yuan, Quanquan Gu

【54】AI-Boosted Video Annotation: Assessing the Process Enhancement
标题：人工智能增强视频注释：评估流程增强
链接：https://arxiv.org/abs/2510.21798

作者：Juan Gutiérrez, Ángel Mora, Pablo Regodón, Silvia Rodriguez, José Luis Blanco

【55】What Causes Postoperative Aspiration?
标题：是什么导致术后抽吸？
链接：https://arxiv.org/abs/2510.21779

作者：Supriya Nagesh, Karina Covarrubias, Robert El-Kareh, Shiva Prasad Kasiviswanathan, Nina Mishra

【56】Chebyshev Moment Regularization (CMR): Condition-Number Control with Moment Shaping
标题：切比雪夫矩正规化（MCR）：具有矩整形的条件数控制
链接：https://arxiv.org/abs/2510.21772

作者：Jinwoo Baek
备注：15 pages

【57】Your Dense Retriever is Secretly an Expeditious Reasoner
标题：你的密集猎犬秘密地是一个快速推理者
链接：https://arxiv.org/abs/2510.21727

作者：Yichi Zhang, Jun Bai, Zhixin Cai, Shuhan Qin, Zhuofan Chen, Jinghua Guan, Wenge Rong
备注：16 pages, 11 figures

【58】From Authors to Reviewers: Leveraging Rankings to Improve Peer Review
标题：从作者到评论家：利用排名改善同行评审
链接：https://arxiv.org/abs/2510.21726

作者：Weichen Wang, Chengchun Shi

【59】Tighter CMI-Based Generalization Bounds via Stochastic Projection and Quantization
标题：通过随机投影和量化来更严格的基于CGI的概括界限
链接：https://arxiv.org/abs/2510.23485

作者：Milad Sefidgaran, Kimia Nadjahi, Abdellatif Zaidi
备注：Accepted for oral presentation at NeurIPS 2025

【60】Robust Decision Making with Partially Calibrated Forecasts
标题：通过部分校准的预测进行稳健决策
链接：https://arxiv.org/abs/2510.23471

作者：Shayan Kiyani, Hamed Hassani, George Pappas, Aaron Roth

【61】Benchmarking VQE Configurations: Architectures, Initializations, and Optimizers for Silicon Ground State Energy
标题：VQE NPS基准：硅接地状态能源的架构、初始化和优化器
链接：https://arxiv.org/abs/2510.23171

作者：Zakaria Boutakka, Nouhaila Innan, Muhammed Shafique, Mohamed Bennai, Z. Sakhi

【62】Coupled Flow Matching
标题：耦合流量匹配
链接：https://arxiv.org/abs/2510.23015

作者：Wenxi Cai, Yuheng Wang, Naichen Shi

【63】Clinic-Oriented Feasibility of a Sensor-Fused Wearable for Upper-Limb Function
标题：用于上肢功能的传感器熔断可穿戴设备面向临床的可行性
链接：https://arxiv.org/abs/2510.22913

作者：Thanyanee Srichaisak, Arissa Ieochai, Aueaphum Aueawattthanaphisut
备注：19 pages, 7 figures, 5 Tables

【64】Scalable Neural Decoders for Practical Real-Time Quantum Error Correction
标题：用于实际实时量子误差纠正的可扩展神经解码器
链接：https://arxiv.org/abs/2510.22724

作者：Changwon Lee, Tak Hur, Daniel K. Park
备注：10 pages, 5 figures

【65】qc-kmeans: A Quantum Compressive K-Means Algorithm for NISQ Devices
标题：qc-kmeans：用于NISQ设备的量子压缩K均值算法
链接：https://arxiv.org/abs/2510.22540

作者：Pedro Chumpitaz-Flores, My Duong, Ying Mao, Kaixun Hua
备注：10 pages, 3 figures, accepted to 2025 IEEE International Conference on Big Data (IEEE BigData 2025)

【66】An Analytic Theory of Quantum Imaginary Time Evolution
标题：量子虚时间演化的分析理论
链接：https://arxiv.org/abs/2510.22481

作者：Min Chen, Bingzhi Zhang, Quntao Zhuang, Junyu Liu
备注：35 pages, 8 figures

【67】Confidence Sets for Multidimensional Scaling
标题：多维缩放的置信集
链接：https://arxiv.org/abs/2510.22452

作者：Siddharth Vishwanath, Ery Arias-Castro
备注：62 pages, 5 figures

【68】Extragradient Method for $(L_0, L_1)$-Lipschitz Root-finding Problems
标题：$（L_0，L_1）$-Lipschitz寻根问题的外梯度方法
链接：https://arxiv.org/abs/2510.22421

作者：Sayantan Choudhury, Nicolas Loizou
备注：Published in NeurIPS 2025, 44 pages, 6 Figures

【69】Right Place, Right Time: Market Simulation-based RL for Execution Optimisation
标题：正确的地点，正确的时间：基于市场模拟的RL执行优化
链接：https://arxiv.org/abs/2510.22206

作者：Ollie Olby, Andreea Bacalum, Rory Baggott, Namid Stillman
备注：8 pages, 4 figures, accepted to ICAIF 2025

【70】Differentially Private High-dimensional Variable Selection via Integer Programming
标题：通过NPS规划的差异私有多维变量选择
链接：https://arxiv.org/abs/2510.22062

作者：Petros Prastakos, Kayhan Behdin, Rahul Mazumder
备注：NeurIPS 2025

【71】An AI enhanced approach to the tree unimodality conjecture
标题：树单峰猜想的人工智能增强方法
链接：https://arxiv.org/abs/2510.18826

作者：Eric Ramos, Sunny Sun
备注：V2 - Fixed typographical errors. Added a remark noting a private correspondence with Galvin and Bencs, who have shown the existence of trees with log concavity breakage at multiple indices

机器翻译由腾讯交互翻译提供，仅供参考

点击“阅读原文”获取带摘要的学术速递