FlashThink: An Early Exit Method For Efficient Reasoning
This addresses efficiency issues in reasoning tasks for users of large language models, though it is incremental as it builds on existing early exit techniques.
The paper tackles the problem of large language models generating excessively long reasoning content, which increases computational overhead, by introducing FlashThink, an early exit method that reduces reasoning length by over 77% for models like Deepseek-R1 and QwQ-32B while maintaining accuracy.
Large Language Models (LLMs) have shown impressive performance in reasoning tasks. However, LLMs tend to generate excessively long reasoning content, leading to significant computational overhead. Our observations indicate that even on simple problems, LLMs tend to produce unnecessarily lengthy reasoning content, which is against intuitive expectations. Preliminary experiments show that at a certain point during the generation process, the model is already capable of producing the correct solution without completing the full reasoning content. Therefore, we consider that the reasoning process of the model can be exited early to achieve the purpose of efficient reasoning. We introduce a verification model that identifies the exact moment when the model can stop reasoning and still provide the correct answer. Comprehensive experiments on four different benchmarks demonstrate that our proposed method, FlashThink, effectively shortens the reasoning content while preserving the model accuracy. For the Deepseek-R1 and QwQ-32B models, we reduced the length of reasoning content by 77.04% and 77.47%, respectively, without reducing the accuracy.