Abstract

Safety tuning through supervised fine-tuning and reinforcement learning from human feedback has substantially improved the robustness of large language models (LLMs). However, it typically suppresses rather than 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 (input-space search), we show that these hidden risks can be systematically exposed through diverse response generation (output-space search). 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.