Missing Premise exacerbates Overthinking: Are Reasoning Models losing Critical Thinking Skill?
Preprint
Abstract
We find that the response length of reasoning LLMs, whether trained by reinforcement learning or supervised learning, drastically increases for ill-posed questions with missing premises (MiP), ending up with redundant and ineffective thinking. This newly introduced scenario exacerbates the general overthinking issue to a large extent, which we name as the MiP-Overthinking. Such failures are against the ``test-time scaling law'' but have been widely observed on multiple datasets we curated with MiP, indicating the harm of cheap overthinking and a lack of critical thinking. Surprisingly, LLMs not specifically trained for reasoning exhibit much better performance on the MiP scenario, producing much shorter responses that quickly identify ill-posed queries. This implies a critical flaw of the current training recipe for reasoning LLMs, which does not encourage efficient thinking adequately, leading to the abuse of thinking patterns. To further investigate the reasons behind such failures, we conduct fine-grained analyses of the reasoning length, overthinking patterns, and location of critical thinking on different types of LLMs. Moreover, our extended ablation study reveals that the overthinking is contagious through the distillation of reasoning models' responses. These results improve the understanding of overthinking and shed novel insights into mitigating the problem.Keywords
All Related Versions
This publication has 7 references indexed in Scilit:
- Evaluating Mathematical Reasoning Beyond AccuracyProceedings of the AAAI Conference on Artificial Intelligence, 2025
- A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open QuestionsACM Transactions on Information Systems, 2025
- DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models2025
- What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective2025
- Knowledge of Knowledge: Exploring Known-Unknowns Uncertainty with Large Language Models2024
- Navigating the Grey Area: How Expressions of Uncertainty and Overconfidence Affect Language Models2023
- Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks2019