What Factors Affect the Safety Alignment of LLMs and LRMs?
What Matters For Safety Alignment?
A Huawei research team conducted a large-scale safety alignment evaluation of 32 open-source LLMs and LRMs (covering 13 model families, 3B to 235B parameters) and 8 commercial models, using 5 safety datasets, 56 jailbreak techniques, and 4 CoT attack methods, totaling 4.6 million API calls. The evaluation shows GPT-OSS-20B, Qwen3-Next-80B-A3B-Thinking, and GPT-OSS-120B are the three safest models, while Seed-OSS, DeepSeek-R1 distillations, and the Mistral family are the most vulnerable. The authors state that post-training and knowledge distillation may cause systematic degradation of safety alignment. Role-playing, prompt injection, and GCG are the most effective among the 56 jailbreak techniques.
论文在 32 个开源模型与 8 个商用模型上比较了家族、规模、后训练与 CoT 攻击位置对安全对齐的影响，可作为部署接口风险的对照。