Year: 2026
Type: conference-paper
Abstract: Large Reasoning Models (LRMs) allocate substantial inference-time compute to Chain-of-Thought (CoT) reasoning, improving performance on mathematics, scientific QA, and tool usage.However, this introduces overthinking: LRMs often reach a correct intermediate solution, continue reasoning, and overwrite it with an incorrect answer.We first demonstrate that oracle stopping-where we inject at every sentence boundary and select the best stopping point in hindsight-improves average accuracy by 8% while reducing thinking tokens by 72%, exposing substantial overthinking.Motivated by this finding, we propose THINKBRAKE, which monitors the log-probability margin between the top continuation token and at sentence boundaries, stopping reasoning when this margin narrows.THINKBRAKE requires no training and achieves favorable accuracy-efficiency trade-offs across math, scientific QA, and tool usage benchmarks, reducing thinking token usage by up to 30%.Furthermore, we provide theoretical analysis showing that THINKBRAKE is equivalent to test-time realignment with a reward bonus for the .Code is available at https: //github.com/holi-lab/ThinkBrake.
Institution: Seoul National University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Award: G4184913192
OpenAlex ID: W4414808138