清华、牛津与斯坦福论文发现 LLM 关闭思考后仍会继续推理
New Tsinghua, Oxford, and Stanford paper finds that LLMs keep reasoning out loud even with thinking turned off, especially on open-ended questions.
清华大学、牛津大学与斯坦福大学的新论文发现，LLM 在关闭思考后仍会继续输出推理过程，在开放式问题上尤其明显。关闭思考时，DeepSeek-V4-Flash 在 99.9% 的开放式回答中仍写出了推理内容，缺少 think 标签或回复简短并不能证明模型跳过了推理。对 5 个模型强制只给答案可将开放式任务的遵从率提升到约 40%，但准确率下降约 15 个百分点。论文地址为 arxiv.org/abs/2610.11765。
正文
当前显示原文，中文全文译文尚未就绪。
New Tsinghua, Oxford, and Stanford paper finds that LLMs keep reasoning out loud even with thinking turned off, especially on open-ended questions.
With thinking disabled, DeepSeek-V4-Flash still wrote out reasoning in 99.9% of open-ended answers. A missing think tag or a short reply does not prove the model skipped reasoning.
Yes/no questions are easy to answer directly, multiple-choice questions sit in the middle, and open-ended ones keep pulling the model back into reasoning.
Forcing answer-only replies on open-ended tasks raised compliance to about 40% across 5 models but cut accuracy by about 15 points.
– arxiv. org/abs/2610.11765