arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

University of Southern California(南加州大学)

共收录 1288
2603.18257 2026-05-08 cs.LG cs.AI

Discovering What You Can Control: Interventional Boundary Discovery for Reinforcement Learning

发现你可以控制的东西:强化学习中的干预边界发现

Jiaxin Liu, Anzhe Cheng, Paul Bogdan

机构 * University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校) University of Southern California(南加州大学)

AI总结 本文提出干预边界发现方法,通过随机化动作生成干预对比,利用FDR校正的两样本检验确定观测维度,有效识别可控状态维度。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.01124 2026-05-08 cs.LG

ChronoSpike: An Adaptive Spiking Graph Neural Network for Dynamic Graphs

ChronoSpike: 一种自适应脉冲图神经网络用于动态图

Md Abrar Jahin, Taufikur Rahman Fuad, Jay Pujara, Craig Knoblock

机构 * University of Southern California(南加州大学) Islamic University of Technology(伊斯兰科技大学)

AI总结 本文提出ChronoSpike,结合可学习LIF神经元、多头空间注意力聚合和轻量Transformer时间编码器,实现动态图的高效表示学习,优于现有方法并在训练速度上更优。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.00407 2026-05-08 cs.LG

Fed-Listing: Federated Label Distribution Inference in Graph Neural Networks

Fed-Listing:图神经网络中的联邦标签分布推断

Suprim Nakarmi, Junggab Son, Yue Zhao, Zuobin Xiong

机构 * Department of Computer Science, University of Nevada Las Vegas(内华达大学拉斯维加斯分校计算机科学系) Department of Computer Science, University of Southern California(南加州大学计算机科学系)

AI总结 本文提出Fed-Listing,一种基于梯度的攻击方法,用于在联邦图神经网络中推断目标客户端的隐私标签统计信息,展示了其在非独立同分布场景下的优越性能。

Comments 9 pages, 3 figures, and 4 tables

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.00839 2026-05-05 cs.AI cs.LG

2026 Roadmap on Artificial Intelligence and Machine Learning for Smart Manufacturing

2026 年人工智能与机器学习在智能制造中的路线图

Jay Lee, Hanqi Su, Marco Macchi, Adalberto Polenghi, Wei Wu, Zhiheng Zhao, George Q. Huang, Kiva Allgood, Devendra Jain, Benedikt Gieger, Vibhor Pandhare, Soumyabrata Bhattacharjee, Ram Mohril, Lingbao Kong, Qiyuan Wang, Xinlan Tang, Sungjong Kim, Chan Hee Park, Byeng D. Youn, Guo Dong Goh, Xi Huang, Wai Yee Yeong, Yung C Shin, He Zhang, Zitong Wang, Fei Tao, Jagjit Singh Srai, Satyandra K. Gupta, Byung Gun Joung, Albin John, John W. Sutherland, Sang Won Lee, Olga Fink, Vinay Sharma, Faez Ahmed, Wei Chen, Mark Fuge, Arild Waaler, Martin G. Skjæveland, Dimitris Kyritsis, Wei Chen, VispiNevile Karkaria, Yi-Ping Chen, Ying-Kuan Tsai, Joseph Cohen, Xun Huan, Jing Lin, Liangwei Zhang, Gregory W. Vogl, Aaron W. Cornelius, Xiaodong Jia, Dai-Yan Ji, Takanobu Minami, Ruoxin Wang

机构 * Center for Industrial Artificial Intelligence, Department of Mechanical Engineering, University of Maryland, College Park(工业人工智能中心,机械工程系,马里兰大学College Park分校) Department of Management, Economics and Industrial Engineering, Politecnico di Milano(管理、经济与工业工程系,米兰理工学院) Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University(工业与系统工程系,香港理工大学) Centre for Advanced Manufacturing & Supply Chains, World Economic Forum(先进制造与供应链研究中心,世界经济论坛) Department of Mechanical Engineering, Indian Institute of Technology Indore(机械工程系,印度理工学院Indore分校) Future Information Innovative College, Fudan University(未来信息创新学院,复旦大学) Department of Mechanical Engineering, Seoul National University(机械工程系,首尔国立大学) Department of Mechanical and Information Engineering, University of Seoul(机械与信息工程系,首尔大学) Onepredict Corp.(Onepredict公司) School of Mechanical and Aerospace Engineering, Nanyang Technological University(机械与航空航天工程学院,南洋理工大学) Singapore Centre for 3D Printing, Nanyang Technological University(新加坡3D打印中心,南洋理工大学) Mechanical Engineering, Purdue University(机械工程系,普渡大学) Digital Twin International Research Center, International Institute for Interdisciplinary and Frontiers, Beihang University(数字孪生国际研究中心, interdisciplinary and Frontiers 国际研究院,北京航空航天大学) School of Automation Science and Electrical Engineering, Beihang University(自动化科学与电气工程学院,北京航空航天大学) Department of Engineering, University of Cambridge(工程系,剑桥大学) Center for Advanced Manufacturing, University of Southern California(先进制造中心,南加州大学) School of Sustainability Engineering and Environmental Engineering, Purdue University(可持续工程与环境工程系,普渡大学) School of Mechanical Engineering, Sungkyunkwan University(机械工程系,全南大学) Intelligent Maintenance and Operations Systems, EPFL(智能维护与运营系统,苏黎世联邦理工学院) Department of Mechanical Engineering, Massachusetts Institute of Technology(机械工程系,麻省理工学院) J. Mike Walker ’66 Department of Mechanical Engineering, Texas A&M University(J. Mike Walker ’66 机械工程系,德克萨斯A&M大学) Department of Mechanical and Process Engineering, ETH Zürich(机械与工艺工程系,苏黎世联邦理工学院)

AI总结 本文探讨人工智能与机器学习在智能制造中的发展现状与未来方向,涵盖基础理论、应用领域及新兴技术,旨在推动创新与产业应用。

Comments This paper has been accepted for publication in the Journal Machine Learning: Engineering

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.15823 2026-05-05 cs.LG cs.AI

CrispEdit: Low-Curvature Projections for Scalable Non-Destructive LLM Editing

CrispEdit:可扩展的非破坏性LLM编辑中的低曲率投影

Zarif Ikram, Arad Firouzkouhi, Stephen Tu, Mahdi Soltanolkotabi, Paria Rashidinejad

机构 * University of Southern California(南加州大学)

AI总结 CrispEdit通过低曲率子空间约束优化,实现LLM编辑中的能力保持,有效降低能力退化,提升编辑效果。

Comments ICML 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.09883 2026-05-05 cs.CL cs.LG

DELTA: Dynamic Layer-Aware Token Attention for Efficient Long-Context Reasoning

DELTA: 动态层感知令牌注意力机制用于高效的长上下文推理

Hossein Entezari Zarch, Lei Gao, Chaoyi Jiang, Murali Annavaram

机构 * University of Southern California(南加州大学)

AI总结 DELTA通过动态层感知令牌注意力机制提升长上下文推理效率,减少计算成本而不牺牲准确性,在AIME和GPQA-Diamond等基准上表现优异。

Comments Accepted to Findings of ACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.00136 2026-05-04 cs.AI

Are Tools All We Need? Unveiling the Tool-Use Tax in LLM Agents

工具是否足够?揭示LLM代理中的工具使用税

Kaituo Zhang, Zhen Xiong, Mingyu Zhong, Zhimeng Jiang, Zhouyuan Yuan, Zhecheng Li, Ying Lin

机构 * University of Houston(休斯顿大学) University of Southern California(南加州大学) New York University(纽约大学) Texas A&M University(德克萨斯农工大学) University of California, San Diego(加州大学圣地亚哥分校)

AI总结 本文研究了工具增强推理在LLM代理中的有效性,发现语义干扰下工具性能可能下降,提出Factorized Intervention Framework分析工具使用成本,并引入G-STEP缓解协议误差。

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.00087 2026-05-04 cs.NI cs.AI cs.CY cs.IR cs.LG

DeGenTWeb: A First Look at LLM-dominant Websites

DeGenTWeb:对以大语言模型为主导的网站的首次观察

Sichang Steven He, Calvin Ardi, Ramesh Govindan, Harsha V. Madhyastha

机构 * University of Southern California(南加州大学)

AI总结 本文通过DeGenTWeb系统识别出大量由大语言模型生成内容的网站,发现其在Web中的普及率持续增长,但准确识别此类网站面临挑战。

Comments 6 pages, 6 figures, 13 page total; in submission

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.00022 2026-05-04 cs.CL cs.AI cs.SD

Putting HUMANS first: Efficient LAM Evaluation with Human Preference Alignment

让人类优先:高效LAM评估的人类偏好对齐

Woody Haosheng Gan, William Held, Diyi Yang

机构 * University of Southern California(南加州大学) Stanford University(斯坦福大学) OpenAthena

AI总结 本文研究了如何通过最小子集高效评估LAM,发现50个样本即可达到高相关性,并通过回归模型提升预测效果,提出HUMANS基准作为高效评估代理。

Comments Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.01082 2026-05-04 cs.LG cs.NE

Discount Model Search for Quality Diversity Optimization in High-Dimensional Measure Spaces

折扣模型搜索用于高维度量空间中的质量多样性优化

Bryon Tjanaka, Henry Chen, Matthew C. Fontaine, Stefanos Nikolaidis

机构 * Thomas Lord Department of Computer Science, University of Southern California(美国南加州大学计算机科学系)

AI总结 本文提出折扣模型搜索(DMS)以解决高维度量空间中质量多样性优化的局限性,通过引入图像高维空间作为度量空间,提升算法性能。

Comments Accepted to ICLR 2026 (Oral presentation). Project page available at https://discount-models.github.io

详情

展开后加载摘要…

URL PDF HTML 收藏
2312.12339 2026-05-04 cs.LG cs.RO

Value Explicit Pretraining for Learning Transferable Representations

为学习可迁移的表示进行价值显式预训练

Kiran Lekkala, Henghui Bao, Sumedh A. Sontakke, Erdem Biyik, Laurent Itti

机构 * Thomas Lord Department of Computer Science at the University of Southern California(南加州大学汤姆·劳德计算机科学系)

AI总结 本文提出价值显式预训练(VEP),通过学习对环境动态和外观变化不变的表示,提升强化学习任务迁移能力。实验表明VEP在未见任务泛化能力、奖励和样本效率上均优于现有方法。

Comments Published in Robotics and Automation Letters (RA-L), January 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.28190 2026-05-01 cs.CV

Representation Fréchet Loss for Visual Generation

用于视觉生成的表示弗雷歇损失

Jiawei Yang, Zhengyang Geng, Xuan Ju, Yonglong Tian, Yue Wang

机构 * USC(美国南加州大学) CMU(卡内基梅隆大学) CUHK(香港中文大学) OpenAI

AI总结 本文提出FD-loss,通过分离FD估计的种群规模与梯度计算的批次规模,发现基于表示空间的FD优化能提升视觉质量,且多步生成器可转为强单步生成器,同时揭示FID可能误判视觉质量。

Comments Code and checkpoints are available at https://github.com/Jiawei-Yang/FD-loss

详情

展开后加载摘要…

URL PDF HTML 收藏
2306.10407 2026-05-01 cs.LG cs.AI physics.bio-ph q-bio.CB

FP-IRL: Fokker--Planck Inverse Reinforcement Learning -- A Physics-Constrained Approach to Markov Decision Processes

FP-IRL:Fokker--Planck反强化学习——一种受物理约束的马尔可夫决策过程方法

Chengyang Huang, Siddhartha Srivastava, Kenneth K. Y. Ho, Kathy E. Luker, Gary D. Luker, Xun Huan, Krishna Garikipati

机构 * Department of Mechanical Engineering, University of Michigan(密歇根大学机械工程系) Department of Aerospace Engineering, Auburn University(奥本大学航空航天工程系) Department of Radiology, University of Michigan(密歇根大学放射学系) Department of Aerospace and Mechanical Engineering, University of Southern California(南加州大学航空航天与机械工程系)

AI总结 本文提出FP-IRL,一种基于Fokker-Planck动态的反强化学习框架,通过轨迹数据直接推断奖励和转移函数,无需预先获取转移信息,提升未知动态下的学习效率与物理可解释性。

Journal ref Computer Methods in Applied Mechanics and Engineering, 458, 119010 (2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.27392 2026-05-01 cs.AI cs.CL cs.CY cs.HC

Leading Across the Spectrum of Human-AI Relationships: A Conceptual Framework for Increasingly Heterogeneous Teams

在人类与人工智能关系的光谱中引领:一种用于日益异质团队的框架

Alejandro R. Jadad

机构 * Keck Medical School, University of Southern California(usc凯克医学院) Vivenxia Group, LLC(Vivenxia集团)

AI总结 本文提出一个光谱框架,帮助领导者识别和管理人类与人工智能协作中的领导角色,强调共适应能力在异质团队中的重要性。

Comments 13 pages, 1 figure, 1 table, 1 appendix, 8 references

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.27356 2026-05-01 cs.LG cs.AI

TypeBandit: Type-Level Context Allocation and Reweighting for Effective Attribute Completion in Heterogeneous Graph Neural Networks

TypeBandit:基于类型级上下文分配与重加权的异构图神经网络中有效属性补全

Ta-Yang Wang, Rajgopal Kannan, Viktor Prasanna

机构 * University of Southern California(南加州大学) DEVCOM Army Research Office(陆军研究办公室)

AI总结 本文提出TypeBandit方法,通过类型级上下文分配与重加权提升异构图神经网络中属性补全效果,结合拓扑感知初始化、类型级老虎机采样和联合表征学习,实现高效且灵活的属性补全策略。

Comments 17 pages, 4 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.00937 2026-05-01 cs.RO cs.AI cs.CV cs.LG

CLAMP: Contrastive Learning for 3D Multi-View Action-Conditioned Robotic Manipulation Pretraining

CLAMP: 基于对比学习的3D多视角动作条件机器人操控预训练

I-Chun Arthur Liu, Krzysztof Choromanski, Sandy Huang, Connor Schenck

机构 * Google DeepMind(谷歌DeepMind) University of Southern California(南加州大学)

AI总结 CLAMP通过3D点云和机器人动作进行预训练,利用对比学习提升机器人操控精度与效率,优于现有基线方法。

Comments Accepted to the Robotics: Science and Systems (RSS) 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.26235 2026-04-30 cs.CR cs.AI cs.CL

LATTICE: Evaluating Decision Support Utility of Crypto Agents

LATTICE:评估加密代理决策支持效用的基准

Aaron Chan, Tengfei Li, Tianyi Xiao, Angela Chen, Junyi Du, Xiang Ren

机构 * Sahara AI University of Southern California(南加州大学)

AI总结 本文提出LATTICE基准,用于评估加密代理在真实用户场景中的决策支持能力,通过六个评估维度、16种任务类型和LLM评委,实现大规模可扩展评估,揭示不同代理在决策支持质量上的显著差异及权衡。

Comments 15 pages, 3 figures, 9 tables

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.18112 2026-04-30 cs.CL cs.MM

Retrieval-Augmented Multimodal Model for Fake News Detection

增强检索的多模态模型用于虚假新闻检测

Yiheng Li, Weihai Lu, Hanyi Yu, Yue Wang

机构 * University of International Business and Economics(国际商务经济大学) Peking University(北京大学) University of Southern California(南加州大学) Upstart Holdings, Inc.(Upstart Holdings公司)

AI总结 本文提出RAMM模型,通过多模态大语言模型和抽象叙述对齐模块,解决虚假新闻检测中跨实例叙述一致性缺失和领域知识不足的问题,实验验证了其有效性。

Comments Accepted to SIGIR 26

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.24977 2026-04-29 cs.CL cs.HC

A Survey on LLM-based Conversational User Simulation

基于大语言模型的对话用户模拟综述

Bo Ni, Leyao Wang, Yu Wang, Branislav Kveton, Franck Dernoncourt, Yu Xia, Hongjie Chen, Reuben Leura, Samyadeep Basu, Subhojyoti Mukherjee, Puneet Mathur, Nesreen Ahmed, Junda Wu, Li Li, Huixin Zhang, Ruiyi Zhang, Tong Yu, Sungchul Kim, Jiuxiang Gu, Zhengzhong Tu, Alexa Siu, Zichao Wang, David Seunghyun Yoon, Nedim Lipka, Namyong Park, Zihao Lin, Trung Bui, Yue Zhao, Tyler Derr, Ryan A. Rossi

机构 * Vanderbilt University(范德比大学) Adobe Research(Adobe研究) Yale University(耶鲁大学) University of Oregon(俄勒冈大学) University of California San Diego(加州大学圣地亚哥分校) Dolby Laboratories(Dolby实验室) University of California, Berkeley(加州大学伯克利分校) Cisco AI Research(思科人工智能研究) University of Southern California(南加州大学) Texas A&M University(德克萨斯阿姆斯特朗大学) UC Davis(加州大学戴维斯分校)

AI总结 本文综述了基于大语言模型的对话用户模拟最新进展,提出新的分类体系,分析核心技术和评估方法,旨在推动未来研究。

Comments Submitted in August 2025. MOD-81000 approved survey

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.12118 2026-04-29 cs.LG cs.DC

Cornserve: A Distributed Serving System for Any-to-Any Multimodal Models

Cornserve:一种用于任意到任意多模态模型的分布式服务系统

Jae-Won Chung, Jeff J. Ma, Jisang Ahn, Yizhuo Liang, Akshay Jajoo, Myungjin Lee, Mosharaf Chowdhury

机构 * University of Michigan(密歇根大学) University of Southern California(南加州大学) Cisco Research(思科研究)

AI总结 本文提出Cornserve,一种支持任意到任意多模态模型的分布式服务系统,通过灵活的任务抽象和组件解耦实现高效部署,提升了吞吐量和延迟性能。

Comments CAIS 2026 Demo track | Open source at https://github.com/cornserve-ai/cornserve | Demo video at https://www.youtube.com/watch?v=nb8R-vztLRg

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.03125 2026-04-29 stat.ML cs.LG

Provable Accelerated Bayesian Optimization with Knowledge Transfer

可证明的加速贝叶斯优化与知识转移

Haitao Lin, Boxin Zhao, Mladen Kolar, Chong Liu

机构 * University of Chicago(芝加哥大学) USC & MBZUAI(USC与MBZUAI) University at Albany, SUNY(纽约州立大学阿尔巴尼分校)

AI总结 本文提出DeltaBO算法,通过构建差分函数的不确定性量化方法,在不同RKHS中实现加速贝叶斯优化,理论证明其在转移学习中具有更优的 regret 上界。

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.12202 2026-04-29 cs.LG stat.ML

Near-Optimal Sample Complexities of Divergence-based S-rectangular Distributionally Robust Reinforcement Learning

基于分歧的S-矩形分布鲁棒强化学习的近优样本复杂性

Zhenghao Li, Shengbo Wang, Nian Si

机构 * HKUST(香港理工大学) University of Southern California(南加州大学)

AI总结 本文研究了基于分歧的S-矩形分布鲁棒强化学习的经验价值迭代算法,推导出近优样本复杂性界,并通过实验验证了算法在鲁棒库存控制中的高效性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.13006 2026-04-28 cs.CL cs.AI

One Token Away from Collapse: The Fragility of Instruction-Tuned Helpfulness

距离崩溃仅一步之遥:指令调制帮助性的脆弱性

Erfan Baghaei Potraghloo, Seyedarmin Azizi, Souvik Kundu, Massoud Pedram

机构 * University of Southern California(南加州大学) Intel AI(英特尔人工智能)

AI总结 研究探讨了指令调制大语言模型在简单约束下的帮助性稳定性,发现其在词汇约束下响应质量显著下降,揭示了指令调制对表面模板的依赖性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.06166 2026-04-28 cs.CV

B-FIRE: Binning-Free Diffusion Implicit Neural Representation for Hyper-Accelerated Motion-Resolved MRI

B-FIRE:无分组扩散隐式神经表示用于超加速运动分辨MRI

Di Xu, Hengjie Liu, Yang Yang, Mary Feng, Jin Ning, Xin Miao, Jessica E. Scholey, Alexandra E. Hotca-cho, William C. Chen, Michael Ohliger, Martina Descovich, Huiming Dong, Wensha Yang, Ke Sheng

机构 * Radiation Oncology, University of California, San Francisco, California(加州大学旧金山分校放射肿瘤学系) Radiology and Biomedical Imaging, University of California, San Francisco, California(加州大学旧金山分校放射学与生物医学成像系) Siemens Medical Solutions USA, Inc., Cleveland, Ohio(西门子医疗解决方案美国公司,克利夫兰,俄亥俄) Radiology at Children’s Hospital Los Angeles, Keck School of Medicine, University of Southern California, Los Angeles, California(洛杉矶儿童医院放射学,美国南加州大学凯克医学院,洛杉矶,加利福尼亚)

AI总结 B-FIRE通过扩散隐式神经表示框架实现超加速MRI重建,解决运动分辨信息模糊问题,提升3D腹部解剖的即时重建精度。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.23863 2026-04-28 cs.RO cs.SY eess.SY

Cooptimizing Safety and Performance Using Safety Value-Constrained Model Predictive Control

通过安全价值约束模型预测控制优化安全与性能

Hao Wang, Nam Nguyen, Armand Jordana, Ludovic Righetti, Somil Bansal

机构 * Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California(南加州大学明希德电气与计算机工程系) Department of Aeronautics and Astronautics, Stanford University(斯坦福大学航空与航天系) Electrical and Computer Engineering Department, New York University(纽约大学电气与计算机工程系)

AI总结 本文提出基于安全价值函数的终端约束MPC方法,实现安全与性能的协同优化,通过仿真和硬件实验验证了其在约束满足和鲁棒性方面的改进。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.23602 2026-04-28 cs.AR cs.LG

TimingLLM: A Two-Stage Retrieval-Augmented Framework for Pre-Synthesis Timing Prediction from Verilog

TimingLLM: 一种两阶段检索增强框架,用于从Verilog预合成时间预测

Armin Abdollahi, Negin Ashrafi, Mehdi Kamal, Massoud Pedram

机构 * University of Southern California(美国南加州大学)

AI总结 TimingLLM通过两阶段检索增强框架,利用Verilog直接预测最坏负松弛和总负松弛,提升预合成时间预测的准确性和效率。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.07410 2026-04-28 cs.RO cs.AI

Using Language Models as Closed-Loop High-Level Planners for Robotics Applications: A Brief Overview and Benchmarks

利用语言模型作为闭环高层规划器用于机器人应用:简要概述与基准测试

Hao Wang, Sathwik Karnik, Bea Lim, Somil Bansal

机构 * Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California(南加州大学明希德电气与计算机工程系) Department of Aeronautics and Astronautics, Stanford University(斯坦福大学航空航天工程系) Department of Mechanical Engineering, Stanford University(斯坦福大学机械工程系)

AI总结 本文研究了语言模型在机器人闭环高层规划中的应用,探讨了控制时间跨度和预热启动对性能的影响,并通过实验提供改进建议。

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.09603 2026-04-28 cs.CL

The Surprising Effectiveness of Membership Inference with Simple N-Gram Coverage

基于简单n-gram覆盖的成员推断的惊人效果

Skyler Hallinan, Jaehun Jung, Melanie Sclar, Ximing Lu, Abhilasha Ravichander, Sahana Ramnath, Yejin Choi, Sai Praneeth Karimireddy, Niloofar Mireshghallah, Xiang Ren

机构 * University of Southern California(南加州大学) University of Washington(华盛顿大学) Stanford University(斯坦福大学)

AI总结 本文提出基于文本输出的成员推断攻击,通过n-gram重叠度评估模型生成文本与真实后缀的相似性,验证了其在黑盒模型上的有效性,并发现攻击成功率随计算预算增加而提升。

Comments CoLM 2025. v2: update citation

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.23141 2026-04-28 cs.CR cs.AI

UNSEEN: A Cross-Stack LLM Unlearning Defense against AR-LLM Social Engineering Attacks

UNSEEN: 一种跨栈LLM去学习防御对抗AR-LLM社会工程攻击

Tianlong Yu, Yang Yang, Xiao Luo, Lihong Liu, Fudu Xing, Zui Tao, Kailong Wang, Gaoyang Liu, Ting Bi

机构 * Hubei University(湖北大学) University of Southern California(南加州大学) Huazhong University of Science and Technology(华中科技大学)

AI总结 针对AR-LLM社会工程攻击,提出UNSEEN跨栈防御机制,结合AR访问控制层、基于F-RMU的LLM去学习和运行时代理防护,通过用户研究验证其有效性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.22096 2026-04-27 cs.CR cs.LG cs.SE

Who Audits the Auditor? Tamper-Proof Fraud Detection with Blockchain-Anchored Explainable ML

谁审计审计师?基于区块链的可篡改欺诈检测可解释机器学习

Zhaohui Wang

机构 * Viterbi School of Engineering(维特比工程学院) University of Southern California(南加州大学) Los Angeles, CA, USA(美国加利福尼亚州洛杉矶)

AI总结 本文提出一种基于区块链的欺诈检测系统,通过智能合约确保审批流程的不可篡改性,提升审计可追溯性与监管合规性。

Comments Accepted to IEEE COMPSAC 2026 (Paper ID 9376, SEPT Symposium). This is the de-anonymized camera-ready version. Code is available at: https://github.com/GeoffreyWang1117/fraud-detection-chain

详情

展开后加载摘要…

URL PDF HTML 收藏