arXiv 论文：推理模型对思维链的可控性高于此前评测结果
DAIR.AI 转发一篇 arXiv 论文（arxiv.org/abs/2610.23257）的发现：推理模型对自身思维链的控制能力比早期测试显示的更强。此前评测把约束放在用户提示词中，得到很低的可控性；研究者将同一约束作为自我指令放进思维链后，模型遵循约束的频率大幅提高。加入一条提醒模型遵守用户约束的通用提示也有明显帮助，且该自我指令可仅从用户提示词触发，无需访问推理通道。
Interesting finding about reasoning models and CoT.
They find that reasoning models can control their chain of thought much better than earlier tests suggested.
Earlier evaluations put the constraint in the user prompt and found very low controllability. Researchers put the same constraint inside the chain of thought as a self-instruction, and models followed it far more often.
A generic reminder to follow the user's constraint also helped a lot. The self-instruction can be triggered from the user prompt alone, so no access to the reasoning channel is needed.
Paper: arxiv.org/abs/2610.23257
Chat with Paper: academy.dair.ai/papers/reaso…
来源：dair_ai · x.com