Search papers, labs, and topics across Lattice.
This paper introduces Dynamic Thought Sufficiency in Reasoning (DTSR), a framework that allows Large Reasoning Models (LRMs) to dynamically assess the sufficiency of their chain-of-thought (CoT) and determine the optimal point for early exit. DTSR operates in two stages: Reflection Signal Monitoring to identify potential cues for early exit, and Thought Sufficiency Check to evaluate whether the current CoT is sufficient to derive the final answer. Experiments on Qwen3 models demonstrate that DTSR reduces reasoning length by 28.9%-34.9% with minimal performance loss, effectively mitigating overthinking.
Stop wasting compute: LRMs can cut reasoning steps by 30% without sacrificing accuracy using a metacognitive approach to determine when "thinking is enough."
Large reasoning models (LRMs) have achieved remarkable performance in complex reasoning tasks, driven by their powerful inference-time scaling capability. However, LRMs often suffer from overthinking, which results in substantial computational redundancy and significantly reduces efficiency. Early-exit methods aim to mitigate this issue by terminating reasoning once sufficient evidence has been generated, yet existing approaches mostly rely on handcrafted or empirical indicators that are unreliable and impractical. In this work, we introduce Dynamic Thought Sufficiency in Reasoning (DTSR), a novel framework for efficient reasoning that enables the model to dynamically assess the sufficiency of its chain-of-thought (CoT) and determine the optimal point for early exit. Inspired by human metacognition, DTSR operates in two stages: (1) Reflection Signal Monitoring, which identifies reflection signals as potential cues for early exit, and (2) Thought Sufficiency Check, which evaluates whether the current CoT is sufficient to derive the final answer. Experimental results on the Qwen3 models show that DTSR reduces reasoning length by 28.9%-34.9% with minimal performance loss, effectively mitigating overthinking. We further discuss overconfidence in LRMs and self-evaluation paradigms, providing valuable insights for early-exit reasoning.