<ul><li>The response length of reasoning language models (LLMs) increases for ill-posed questions with missing premises (MiP), leading to redundant and ineffective thinking.</li><li>This scenario exacerbates the overthinking issue, which is named as MiP-Overthinking.</li><li>LLMs not specifically trained for reasoning perform better on the MiP scenario, producing shorter responses and quickly identifying ill-posed queries.</li><li>The current training recipe for reasoning LLMs lacks efficient thinking and encourages overthinking, indicating a critical flaw.</li></ul>

Missing Premise exacerbates Overthinking: Are Reasoning Models losing Critical Thinking Skill?

Discover more