Skip to content

DACE: Diversity-Driven Adversarial Co-Evolution for Robust LLM Safety Alignment

DACE: Diversity-Driven Adversarial Co-Evolution for Robust LLM Safety Alignment

arXiv: 其他 AI 安全论文· arXiv:2611.22278· NeurIPS 2026· Peng Yu, Xiaoyu Wen, Zhida He, Ziyuan Zhou, Han Qi, Shao Zhang, Qiaosheng Zhang, Ying Wen, Chaochao Lu·Indexed by this site · 2 days ago· Originally publishedDefenses· Its place in the scan record →
AI summary

DACE is a diversity-driven adversarial co-evolution framework that targets attack strategy collapse and defensive adversarial forgetting to enhance the robustness of LLM safety alignment. On the attack side, it uses a two-dimensional strategy space of risk category × attack style together with a marginal coverage gain reward; on the defense side, it uses Thompson sampling to maintain a Bayesian adversarial replay pool. On standard safety, automated attacker, and general capability benchmarks, DACE improves robustness to out-of-distribution attacks, preserves general capabilities, and achieves broader attack strategy coverage.

Read originalNeurIPS 2026