OpenAI News·· 2026-03-05AI 评分44
OpenAI:推理模型难以控制其思维链,而这反而是好事
Reasoning models struggle to control their chains of thought, and that’s good
AI 导读
OpenAI 推出 CoT-Control,发现推理模型难以控制自己的思维链,这反而强化了可监控性作为 AI 安全保障的价值。
来源:OpenAI News · openai.com
Reasoning models struggle to control their chains of thought, and that’s good
OpenAI 推出 CoT-Control,发现推理模型难以控制自己的思维链,这反而强化了可监控性作为 AI 安全保障的价值。
来源:OpenAI News · openai.com