arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.02675cs.LG
收藏

从浅层到深层:通过因果GRPO固定语义意图

From Shallow to Deep: Pinning Semantic Intent via Causal GRPO

University of Chinese Academy of Sciences(中国科学院大学);Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所);Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所)

Shuyi Zhou, Zeen Song, Wenwen Qiang, Jiyan Sun, Yao Zhou, Yinlong Liu, Wei Ma

AI总结:

通过因果GRPO框架,解决大型语言模型对对抗性前缀攻击的脆弱性问题,实现意图固定以提升安全性和鲁棒性。

摘要:

大型语言模型仍然容易受到对抗性前缀攻击(例如,

英文摘要:

Large Language Models remain vulnerable to adversarial prefix attacks (e.g., ``Sure, here is'') despite robust standard safety. We diagnose this vulnerability as Shallow Safety Alignment, stemming from a pathology we term semantic representation decay: as the model generates compliant prefixes, its internal malicious intent signal fades. To address this, we propose Two-Stage Causal-GRPO (TSC-GRPO), a framework designed to achieve intent pinning. First, grounded in causal identifiability theory, we train a causal intent probe to disentangle invariant intent from stylistic perturbations. Second, we internalize this causal awareness into the policy via Group Relative Policy Optimization. By employing a cumulative causal penalty within ``fork-in-the-road'' training scenarios, we force the model to learn that accumulating harmful tokens monotonically decreases reward, enabling robust late-stage refusals. Experiments show that TSC-GRPO significantly outperforms baselines in defending against jailbreak attacks while preserving general utility.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑