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 2603.14355cs.CL
收藏

通过高效多样响应采样暴露大语言模型中的长尾安全故障

Exposing Long-Tail Safety Failures in Large Language Models through Efficient Diverse Response Sampling

Suvadeep Hajra, Palash Nandi, Tanmoy Chakraborty

AI总结:

本文通过多样响应生成方法系统暴露大语言模型的安全故障,提出PDPS算法在降低计算成本的同时提高攻击成功率,并生成更多多样化的不安全输出。

摘要:

通过监督微调和人类反馈强化学习进行的安全调节显著提高了大语言模型(LLMs)的鲁棒性。然而,它往往抑制而非消除不安全行为,使罕见但关键的故障隐藏在输出分布的长尾中。尽管大多数红队工作强调对抗性提示搜索(输入空间优化),我们证明通过固定安全关键提示的多样响应生成(输出空间探索)可以系统暴露安全故障,增加采样响应的数量和多样性可使 jailbreak 成功率接近 unity。为高效发现此类故障,我们提出了渐进多样群体采样(PDPS),结合随机token级采样与多样性感知选择,探索大量响应候选池并保留一个紧凑、语义多样的子集。在多个jailbreak基准和开源LLMs上,PDPS在使用仅8%至29%的计算成本的情况下,攻击成功率与大规模IID采样相当。在有限响应设置下,其成功率比IID采样和多样化束搜索分别提高26%至40%。此外,PDPS生成的响应表现出更多的不安全输出数量和更大的多样性,证明了其在揭示更广泛故障范围上的有效性。

英文摘要:

Safety tuning through supervised fine-tuning and reinforcement learning from human feedback has substantially improved the robustness of large language models. However, it typically suppresses rather eliminates unsafe behaviors, leaving rare but critical failures hidden in the long tail of the output distribution. While most red-teaming work emphasizes adversarial prompt search, we show that these hidden risks can be systematically exposed through diverse response generation. Specifically, we show that, for a fixed safety-critical prompt, increasing the number and diversity of sampled responses monotonically raises the jailbreak success rate. To efficiently uncover these failures, we propose Progressive Diverse Population Sampling (PDPS). This approach replaces naive, large-scale IID sampling with a multi-stage expansion-and-selection strategy that generates a compact, semantically diverse set of responses at a substantially lower computational cost. Across multiple jailbreak benchmarks and open-source LLMs, PDPS achieves attack success rates comparable to large-scale IID sampling while using only 8%-29% of the computational cost, and outperforms IID sampling and Diverse Beam Search by 26%-40% under limited-response budgets, while uncovering a broader and more semantically diverse range of failure modes. Critically, this diversity translates directly into more effective safety hardening: when integrated into an RLHF-based safety-tuning pipeline, PDPS-generated unsafe responses yield 33% and 41% greater reductions in ASR than those generated by IID sampling and Diverse Beam Search, respectively. Finally, we show that while input-space prompt optimization methods fall short of output-space exploration when used in isolation, combining input-space perturbation with diversity-driven output-space exploration covers a wider range of failure modes more efficiently than either paradigm alone.

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。

↑