LLMs Show “Thinking Inertia” When Told to Answer Only
Researchers from Tsinghua University, the University of Oxford and Stanford University tested six models from the Qwen3, DeepSeek, GPT and Gemini families across six benchmarks. They found that disabling thinking mode and asking for answers without reasoning reduced, but did not eliminate, visible calculations or derivations. Explicit reasoning was most common on open-ended questions, while yes-or-no questions prompted more direct answers. The study introduced three metrics to assess answer-only compliance, relevance of preceding text and explicit reasoning separately. The paper, code and trajectory data are public. The source reported that the paper was accepted at NeurIPS 2026, and a separate community member ran a small-scale independent test that also observed reasoning after thinking was disabled.
The study shows that answer accuracy does not establish compliance with an instruction to omit reasoning. Its separate evaluation framework gives developers and evaluators a way to determine whether disabling thinking mode actually suppresses visible reasoning.
新智元 2026-10-25 12:17 北京
新智元报道
通过接口关掉大模型的「思考模式」（如将Qwen3的enable_thinking设为False），再告诉它「不要推理，直接给出答案」，它就会直接回答吗？实际输出并不符合我们日常直觉。
清华大学、牛津大学和斯坦福大学的研究者系统测试了这个问题。他们发现，即使关闭思考模式，模型仍可能在最终答案前写出计算或推导；题目越开放，这种现象越明显。研究者把这种现象称为Thinking Inertia（思考惯性）。
#Thinking Inertia#Tsinghua University#University of Oxford#Stanford University#DeepSeek-V4-Flash#Qwen3-32B#Gemini-4-Argon#GPT-6-Astra#Kimi-K3#Muse-Wonder-1.0#NeurIPS 2026