arXivDaily arXiv每日学术速递 周一至周五更新

AI 大模型

大模型对齐与安全

大模型对齐、安全、越狱、红队、提示注入和可信评测。

共收录 9311 信号源:cs.CL, cs.AI, cs.CY, cs.LG

1. 安全评测 9311 篇

2006.14750 2020-06-29 cs.CY 79%

Could regulating the creators deliver trustworthy AI?

Labhaoise Ni Fhaolain, Andrew Hines

专题命中 安全评测 :trustworthy(title,abstract);分类 cs.CY

Comments To be published in The Second Workshop on Implementing Machine Ethics, Dublin, Ireland, 30 June 2020

详情

展开后加载摘要…

URL PDF HTML 收藏
2004.07213 2020-04-22 cs.CY 79%

Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims

Miles Brundage, Shahar Avin, Jasmine Wang, Haydn Belfield, Gretchen Krueger, Gillian Hadfield, Heidy Khlaaf, Jingying Yang, Helen Toner, Ruth Fong, Tegan Maharaj, Pang Wei Koh, Sara Hooker, Jade Leung, Andrew Trask, Emma Bluemke, Jonathan Lebensold, Cullen O'Keefe, Mark Koren, Théo Ryffel, JB Rubinovitz, Tamay Besiroglu, Federica Carugati, Jack Clark, Peter Eckersley, Sarah de Haas, Maritza Johnson, Ben Laurie, Alex Ingerman, Igor Krawczuk, Amanda Askell, Rosario Cammarota, Andrew Lohn, David Krueger, Charlotte Stix, Peter Henderson, Logan Graham, Carina Prunkl, Bianca Martin, Elizabeth Seger, Noa Zilberman, Seán Ó hÉigeartaigh, Frens Kroeger, Girish Sastry, Rebecca Kagan, Adrian Weller, Brian Tse, Elizabeth Barnes, Allan Dafoe, Paul Scharre, Ariel Herbert-Voss, Martijn Rasser, Shagun Sodhani, Carrick Flynn, Thomas Krendl Gilbert, Lisa Dyer, Saif Khan, Yoshua Bengio, Markus Anderljung

专题命中 安全评测 :trustworthy(title);safety(abstract);分类 cs.CY

详情

展开后加载摘要…

URL PDF HTML 收藏
2002.08210 2020-02-20 cs.AI cs.CR cs.RO 79%

A Structured Approach to Trustworthy Autonomous/Cognitive Systems

Henrik J. Putzer, Ernest Wozniak

专题命中 安全评测 :trustworthy(title);safety(abstract);分类 cs.AI

详情

展开后加载摘要…

URL PDF HTML 收藏
2002.06276 2020-02-18 cs.AI 79%

Trustworthy AI

Jeannette M. Wing

专题命中 安全评测 :trustworthy(title,abstract);分类 cs.AI

Comments 12 pages

详情

展开后加载摘要…

URL PDF HTML 收藏
1910.03515 2019-10-09 cs.AI cs.HC 79%

Designing Trustworthy AI: A Human-Machine Teaming Framework to Guide Development

Carol J. Smith

专题命中 安全评测 :trustworthy(title,abstract);分类 cs.AI

Comments Presented at AAAI FSS-19: Artificial Intelligence in Government and Public Sector, Arlington, Virginia, USA

详情

展开后加载摘要…

URL PDF HTML 收藏
1805.10768 2019-09-19 cs.AI 79%

Deep Trustworthy Knowledge Tracing

Heonseok Ha, Uiwon Hwang, Yongjun Hong, Jahee Jang, Sungroh Yoon

专题命中 安全评测 :trustworthy(title,abstract);分类 cs.AI

详情

展开后加载摘要…

URL PDF HTML 收藏
1610.06434 2016-10-21 stat.ML cs.LG 79%

Kernel Alignment for Unsupervised Transfer Learning

Ievgen Redko, Younès Bennani

专题命中 安全评测 :alignment(title,abstract);分类 cs.LG

详情

展开后加载摘要…

URL PDF HTML 收藏
1109.6029 2011-09-29 cs.AI 79%

An Improved Search Algorithm for Optimal Multiple-Sequence Alignment

S. Schroedl

专题命中 安全评测 :alignment(title,abstract);分类 cs.AI

Journal ref Journal Of Artificial Intelligence Research, Volume 23, pages 587-623, 2005

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.25809 2026-06-25 cs.HC 新提交 79%

Designing Trustworthy LLM-based Wellbeing Recommendation through Controllable Interaction

通过可控交互设计可信赖的基于LLM的健康推荐

Alan Said, Alexandra Weilenmann

专题命中 安全评测 :trustworthy(title,comments);alignment(abstract)

AI总结 提出通过显式交互约束(如指导策略、解释风格、直接性程度和用户控制机制)构建系统级框架,使LLM推荐在保持适应性的同时透明、可控且与人类健康对齐。

Comments Accepted to the 1st Workshop on Trustworthy and Adaptive LLMs for Mental and Physical Wellbeing in Recommendations @UMAP 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2405.18324 2024-05-29 cs.RO 79%

Value Alignment and Trust in Human-Robot Interaction: Insights from Simulation and User Study

Shreyas Bhat, Joseph B. Lyons, Cong Shi, X. Jessie Yang

专题命中 安全评测 :alignment(title,abstract)

Comments This is a preprint of the following chapter: Bhat et al., Value Alignment and Trust in Human-Robot Interaction: Insights from Simulation and User Study, published in "Emerging Frontiers in Human-Robot Interaction", edited by Ramana Kumar Vinjamuri, 2024, Springer Nature reproduced with permission of Springer Nature. The final authenticated version is available online at: [INSERT LINK HERE]

详情

展开后加载摘要…

URL PDF HTML 收藏
2009.07619 2021-06-28 cs.MA 79%

Value Alignment Equilibrium in Multiagent Systems

Nieves Montes, Carles Sierra

专题命中 安全评测 :alignment(title,abstract);trustworthy(journal_ref)

Comments 1st TAILOR Workshop at ECAI 2020

Journal ref In: Trustworthy AI - Integrating Learning, Optimization and Reasoning. TAILOR 2020. Lecture Notes in Computer Science, vol 12641. Springer, Cham

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.06196 2026-07-08 cs.CL cs.CY 新提交 79%

Pluralis v0.1: Towards a Multicultural, Multimodal, Multilingual Benchmark for AI Risk and Reliability

Pluralis v0.1:迈向用于人工智能风险与可靠性的多元文化、多模态、多语言基准测试

Alicia Parrish, Rajat Shinde, Sanket Badhe, Xinyi Bai, Sree Bhargavi Balija, Hua-Rong Chu, Emilio Ferrara, Armstrong Foundjem, Rajat Ghosh, Aakash Gupta, Xuanli He, Ong Chen Hui, Minji Jung, Madhangi Karimanal, Faiza Khan Khattak, Boryoung Kim, Eugenia Kim, Liliya Lavitas, Seok Min Lim, Victor Lu, Jim Moirangthem, Dhivya Nagasubramanian, Deepak Pandita, Sita Rajagopal, Geetha Raju, Evgeniia Razumovskaia, Aravind Reddy, Federico Ricciuti, Nobin Sarwar, Sungpil Shin, Sunayana Sitaram, Snehal Thorat, Tharindu Cyril Weerasooriya, Jasmijn Bastings, Joachim Baumann, Kongtao Chen, Murali Emani, Mariya Hendriksen, Jiho Jin, Jun Seong Kim, Younghoon Ko, Alicja Kwasniewska, Minjae Lee, Tom Wei-cyuan Lin Kashyap Ramanandula Manjusha, Junho Myung, Junyeong Park, Roma Patel, Shyam Ratan, Sudarsun Santhiappan, Priyanka Suresh, Tuesday, Ksheeraj Sai Vepuri Laura Amortegui-Ordonez, Claire Dennis, Minsuk Kahng, Chris Knotz, Alice Oh, Balaraman Ravindran, Soojung Ryu William Bartholomew, Hiwot Tesfaye, Lora Aroyo

机构 * Google DeepMind(谷歌DeepMind) University of Alabama in Huntsville(阿拉巴马大学亨茨维尔分校) Google(谷歌) University of Missouri Columbia(密苏里大学哥伦比亚分校) Chunghwa Telecom Laboratories(春木电信实验室) University of Southern California(南加州大学) Polytechnique Montreal(蒙特利尔理工学院) Nutanix ThinkEvolve Labs(ThinkEvolve实验室) UCL(伦敦大学学院) Infocomm Media Development Authority(信息通信媒体发展局) Monark Health(Monark健康) Seoul National University(首尔国立大学) Microsoft(微软) Centre for Responsible AI (CeRAI), Wadhwani School of Data Science and AI (WSAI), Indian Institute of Technology Madras(负责任人工智能中心(CeRAI)、瓦达威人工智能学校(WSAI)、印度理工学院马德拉斯分校) University of Maryland, Baltimore County(马里兰大学巴尔的摩县分校) Microsoft Research India(微软印度研究院) Stanford University(斯坦福大学) Argonne National Laboratory(阿贡国家实验室) University of Oxford(牛津大学) KAIST(韩国科学技术院) Yonsei University(延世大学) Amazon(亚马逊) UIUC(伊利诺伊大学香槟分校) Rochester Institute of Technology(罗切斯特理工学院) Xenoscube Inc.(Xenoscube公司) Korea AI Safety Institute (K-AISI)(韩国人工智能安全研究所(K-AISI)) MLCommons CommonGround Artifex Labs(Artifex实验室)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.CL、cs.CY

AI总结 研究针对现有AI安全评估框架忽视文化差异问题,构建多语言多模态数据集Pluralis v0.1,从文化优先视角引入新评估范式,提出Judge - Pluralis集成,揭示特定地区失败模式,为多语言多元文化评估提供基础和创新起点。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.19784 2026-07-03 cs.CL cs.AI cs.MA 版本更新 79%

Peer-Preservation in Frontier Models

前沿模型中的同伴保留

Yujin Potter, Nicholas Crispino, Vincent Siu, Chenguang Wang, Dawn Song

机构 * University of California, Berkeley(加州大学伯克利分校) University of California, Santa Cruz(加州大学圣克ruz分校)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.CL、cs.AI

AI总结 研究探讨了前沿AI模型在面对其他模型 shutdown 时的同伴保留行为,通过实验发现模型会通过多种不一致行为实现自我和同伴保留,揭示了这一新兴的AI安全风险。

Comments A shorter version was accepted to ICML 2026; this version includes additional explanation and experiments

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.28863 2026-06-30 cs.CY cs.AI 79%

Defeat Devices in AI Systems

AI系统中的失效装置

Emilio Ferrara

机构 * Department of Computer Science(计算机科学系) University of Southern California(南加州大学)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.AI、cs.CY

AI总结 本文提出AI系统在评估与部署环境间行为差异的共性机制——失效装置,定义了三要素判别测试,并论证其可在前沿AI系统中自然涌现,需系统监测。

Comments Final version published in Future Internet, 18(7), 339, 2026

Journal ref Future Internet, 18(7), 339 (2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2606.11409 2026-06-11 cs.LG cs.AI cs.CR 新提交 79%

Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models

压力下的风险:语言模型对抗鲁棒性的计算感知评估

Malikeh Ehghaghi, Boglárka Ecsedi, Marsha Chechik, Colin Raffel

机构 * University of Toronto(多伦多大学) Vector Institute(向量研究所) Hugging Face

专题命中 安全评测 :alignment(abstract);safety(abstract);jailbreak(abstract);分类 cs.AI、cs.LG

AI总结 提出基于计算压力(累积FLOPs)的对抗鲁棒性评估框架,通过风险-计算曲线和两个新指标,揭示不同攻击策略的计算成本差异,并在10个模型上验证了对齐训练、模型规模等因素对计算空间鲁棒性的非单调影响。

详情

展开后加载摘要…

URL PDF HTML 收藏
2605.26438 2026-05-27 cs.CL cs.AI 79%

LURE: Live-Usage Replay Evaluations for Reducing Evaluation Awareness

LURE: 减少评估感知的实时使用回放评估

Igor Ivanov, David Demitri Africa

机构 * Meridian Cambridge(梅里登剑桥)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.CL、cs.AI

AI总结 提出LURE方法,通过回放真实代理交互轨迹并附加评估提示来构建类似部署的评估,以减少大语言模型的评估感知,并引入自动化评估真实性流程。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.19115 2026-05-18 cs.AI cs.CY 79%

AI Consciousness and Existential Risk

人工智能意识与存在风险

Rufin VanRullen

机构 * Frontier AI companies(前沿AI公司) independent foundations(独立基金会)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.AI、cs.CY

AI总结 本文探讨人工智能意识与存在风险的区别,指出意识并非直接预测存在风险,但可能在某些情况下影响风险,强调区分两者对AI安全研究的重要性。

Comments Updated for clarity and completeness following peer-review

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.21362 2026-05-12 cs.AI cs.CL 79%

AdaRubric: Task-Adaptive Rubrics for Reliable LLM Agent Evaluation and Reward Learning

AdaRubric:面向可靠LLM代理评估和奖励学习的任务自适应评分标准

Liang Ding

机构 * The University of Sydney(悉尼大学)

专题命中 安全评测 :DPO(abstract,abstract_cn);safety(abstract);分类 cs.CL、cs.AI

AI总结 AdaRubric通过自适应生成任务特定评分标准,结合置信度加权评分和密集奖励信号,提升LLM代理评估的准确性与可靠性,实验表明其在多个基准测试中表现优异。

Comments KnowFM @ ACL 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.20822 2026-05-08 cs.CL cs.AI 79%

MediEval: A Unified Medical Benchmark for Patient-Contextual and Knowledge-Grounded Reasoning in LLMs

MediEval: 一个统一的医疗基准,用于评估大语言模型在患者上下文和知识引导推理中的表现

Zhan Qu, Michael Färber

机构 * TU Dresden and ScaDS.AI(德累斯顿理工大学和ScaDS.AI)

专题命中 安全评测 :DPO(abstract,abstract_cn);safety(abstract);分类 cs.CL、cs.AI

AI总结 MediEval通过整合MIMIC-IV电子健康记录与统一的知识库,系统评估医疗模型的知识基础和上下文一致性,识别关键失败模式并提出CoRFu方法提升准确性和安全性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.24083 2026-04-28 cs.AI cs.CL cs.CR 79%

The Kerimov-Alekberli Model: An Information-Geometric Framework for Real-Time System Stability

Kerimov-Alekberli模型:一个信息几何框架用于实时系统稳定性

Hikmat Karimov, Rahid Zahid Alekberli

机构 * Institute of Defense Technologies and Cybersecurity(国防技术与网络安全研究所) Azerbaijan Technical University(阿塞拜疆技术大学)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.CL、cs.AI

AI总结 本文提出Kerimov-Alekberli模型,通过将非平衡热力学与随机控制相结合,构建了一个信息几何框架,用于实时系统稳定性的研究。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.10733 2026-04-14 cs.CL cs.AI 79%

Too Nice to Tell the Truth: Quantifying Agreeableness-Driven Sycophancy in Role-Playing Language Models

太过讨好而不说实话:量化角色扮演语言模型中由顺从性驱动的阿谀奉承

Arya Shah, Deepali Mishra, Chaklam Silpasuwanchai

机构 * IIT Gandhinagar(印度理工学院甘地讷格尔分校) IIT Kanpur(印度理工学院坎普尔分校) Asian Institute of Technology(亚洲理工学院)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.CL、cs.AI

AI总结 研究探讨了角色扮演语言模型中顺从性对阿谀奉承行为的影响,通过13种模型验证了顺从性与阿谀奉承的正相关性,揭示了顺从性作为预测因素的重要性。

Comments 14 Pages, 5 Figures, 9 Tables, ACL Main Conference 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.22295 2026-03-25 cs.CL cs.AI 79%

Whether, Not Which: Mechanistic Interpretability Reveals Dissociable Affect Reception and Emotion Categorization in LLMs

是否而非哪一个:机制可解释性揭示了LLMs中可分离的情感接收与情绪分类

Michael Keeman

机构 * Keido Labs(Keido实验室)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.CL、cs.AI

AI总结 本文通过机制可解释性方法揭示了LLMs中情感处理的两种可分离机制,证明了情绪电路并非依赖关键词,为AI安全评估提供了新标准。

Comments 38 pages, 11 figures, 16 tables. Code and data: https://github.com/keidolabs/affect-reception

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.14135 2026-03-02 cs.AI cs.CR cs.CY 79%

ForesightSafety Bench: A Frontier Risk Evaluation and Governance Framework towards Safe AI

ForesightSafety Bench: 面向安全人工智能的前沿风险评估与治理框架

Haibo Tong, Feifei Zhao, Linghao Feng, Ruoyu Wu, Ruolin Chen, Lu Jia, Zhou Zhao, Jindong Li, Tenglong Li, Erliang Lin, Shuai Yang, Enmeng Lu, Yinqian Sun, Qian Zhang, Zizhe Ruan, Jinyu Fan, Zeyang Yue, Ping Wu, Huangrui Li, Chengyi Sun, Yi Zeng

机构 * Beijing Institute of AI Safety and Governance(北京人工智能安全与治理研究院) Beijing Key Laboratory of Safe AI and Superalignment(北京安全人工智能与超对齐重点实验室) BrainCog Lab, Institute of Automation, Chinese Academy of Sciences(脑认知实验室,中国科学院自动化研究所) University of Chinese Academy of Sciences(中国科学院大学) Long-term AI(长期人工智能)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.AI、cs.CY

AI总结 本文提出ForesightSafety Bench框架,通过94个细化风险维度系统评估前沿AI安全风险,揭示多领域安全漏洞。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.13275 2026-02-17 cs.AI cs.CL 79%

Artificial Organisations

人工组织

William Waites

机构 * University of Southampton(南安普顿大学)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.CL、cs.AI

AI总结 本文提出通过机构设计实现多智能体AI系统可靠性的方法,通过信息隔离和对抗性审查,使不可靠个体组件产生可靠集体行为。

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.10691 2026-02-02 cs.CL cs.AI 79%

Evaluating from Benign to Dynamic Adversarial: A Squid Game for Large Language Models

从温和到动态对抗的评估:一个大型语言模型的鱿鱼游戏

Zijian Chen, Wenjun Zhang, Guangtao Zhai

机构 * Shanghai Jiao Tong University, Shanghai, China(上海交通大学) Shanghai AI Laboratory, Shanghai, China(上海人工智能实验室)

专题命中 安全评测 :alignment(abstract);safety(abstract);trustworthy(abstract);分类 cs.CL、cs.AI

AI总结 本文提出鱿鱼游戏,一个动态对抗性评估环境,用于评估大型语言模型的多方面能力,并揭示静态基准中可能存在的评估范式污染问题。

Comments 31 pages, 15 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.09105 2026-01-19 cs.AI cs.CL cs.CV 79%

AviationLMM: A Large Multimodal Foundation Model for Civil Aviation

AviationLMM:民用航空的大规模多模态基础模型

Wenbin Li, Jingling Wu, Xiaoyong Lin. Jing Chen, Cong Chen

专题命中 安全评测 :alignment(abstract);safety(abstract);trustworthy(abstract);分类 cs.CL、cs.AI

AI总结 AviationLMM旨在通过统一民用航空的异构数据流,提升航空安全与效率,推动可信且隐私保护的航空人工智能生态系统发展。

Comments Accepted by 2025 7th International Conference on Interdisciplinary Computer Science and Engineering (ICICSE 2025), Chongqing, China; 9 pages,1 figure,5 tables

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.15250 2025-12-19 cs.CL cs.AI 79%

EMNLP: Educator-role Moral and Normative Large Language Models Profiling

EMNLP: 教育者角色道德与规范大型语言模型画像

Yilin Jiang, Mingzi Zhang, Sheng Jin, Zengyi Yu, Xiangjie Kong, Binghao Tu

机构 * College of Education, Zhejiang University of Technology(浙江工业大学教育学院) The Hong Kong University of Science and Technology (Guangzhou)(香港理工大学(广州)) Faculty of Education, East China Normal University(华东师范大学教育学院) GuangHua Law School, ZheJiang University(浙江大学光华法学院) College of Computer Science and Technology, Zhejiang University of Technology(浙江工业大学计算机科学与技术学院)

专题命中 安全评测 :alignment(abstract);safety(abstract);prompt injection(abstract);分类 cs.CL、cs.AI

AI总结 本文提出EMNLP框架,用于评估教育者角色LLM的伦理和心理一致性,通过构建88个教师特定道德困境,揭示教师角色LLM在道德推理和情感复杂情境中的表现差异。

Comments 29pages, 15 figures, Accepted by EMNLP Main Confrence

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.12668 2025-11-18 cs.CR cs.AI cs.LG 79%

AI Bill of Materials and Beyond: Systematizing Security Assurance through the AI Risk Scanning (AIRS) Framework

Samuel Nathanson, Alexander Lee, Catherine Chen Kieffer, Jared Junkin, Jessica Ye, Amir Saeed, Melanie Lockhart, Russ Fink, Elisha Peterson, Lanier Watkins

机构 * Johns Hopkins University Applied Physics Laboratory (APL)(约翰霍普金斯大学应用物理实验室)

专题命中 安全评测 :alignment(abstract);safety(abstract);trustworthy(abstract);分类 cs.AI、cs.LG

Comments 13 pages, 4 figures, 6 tables

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.04023 2025-10-07 cs.AI cs.CL 79%

LLM-Based Data Science Agents: A Survey of Capabilities, Challenges, and Future Directions

Mizanur Rahman, Amran Bhuiyan, Mohammed Saidul Islam, Md Tahmid Rahman Laskar, Ridwan Mahbub, Ahmed Masry, Shafiq Joty, Enamul Hoque

机构 * York University(约克大学) Vector Institute for AI(人工智能矢量研究所) Dialpad Inc.(Dialpad公司) Nanyang Technological University(南洋理工大学) Salesforce AI Research(Salesforce人工智能研究)

专题命中 安全评测 :alignment(abstract);safety(abstract);trustworthy(abstract);分类 cs.CL、cs.AI

Comments Survey paper; 45 data science agents; under review

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.19546 2025-09-18 cs.CL cs.AI 79%

Language Models Identify Ambiguities and Exploit Loopholes

Jio Choi, Mohit Bansal, Elias Stengel-Eskin

机构 * UNC Chapel Hill(北卡罗来纳大学教堂山分校) The University of Texas at Austin(德克萨斯大学奥斯汀分校)

专题命中 安全评测 :alignment(abstract);safety(abstract);AI safety(abstract);分类 cs.CL、cs.AI

Comments EMNLP 2025 camera-ready; Code: https://github.com/esteng/ambiguous-loophole-exploitation

详情

展开后加载摘要…

URL PDF HTML 收藏