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。

2025-10-15 至 2025-10-15 共收录 3 篇信号源：cs.CL, cs.AI, cs.CY, cs.LG

1. 越狱攻击 3 篇

2507.07146 2025-10-15 cs.LG cs.CL 84%

Attention-Aware GNN-based Input Defense against Multi-Turn LLM Jailbreak

Zixuan Huang, Kecheng Huang, Lihao Yin, Bowei He, Huiling Zhen, Mingxuan Yuan, Zili Shao

机构 * The Chinese University of Hong Kong（香港中文大学）； Noah’s Ark Lab, Huawei（华为诺亚实验室）； City University of Hong Kong（香港城市大学）

专题命中越狱攻击：jailbreak(title,abstract)；safety(abstract)；分类 cs.CL、cs.LG

2402.10601 2025-10-15 cs.CL cs.AI 62%

When "Competency" in Reasoning Opens the Door to Vulnerability: Jailbreaking LLMs via Novel Complex Ciphers

Divij Handa, Zehua Zhang, Amir Saeidi, Shrinidhi Kumbhar, Md Nayem Uddin, Aswin RRV, Chitta Baral

机构 * Arizona State University（亚利桑那州立大学）

专题命中越狱攻击：safety(abstract)；分类 cs.CL、cs.AI

Comments Published in Reliable ML from Unreliable Data workshop @ NeurIPS 2025

2510.11837 2025-10-15 cs.CR cs.AI 61%

Countermind: A Multi-Layered Security Architecture for Large Language Models

Dominik Schwarz

机构 * Independent Researcher（独立研究者）

专题命中越狱攻击：prompt injection(abstract,comments)；分类 cs.AI

Comments 33 pages, 3 figures, 6 tables. Keywords: LLM security; defense-in-depth; prompt injection; activation steering; multimodal sandbox; threat modeling