DART: Draft-Agreement Routing for Training-Free Adaptive Thinking Budgets in Hybrid Reasoning Models

• URL: http://arxiv.org/abs/2606.23181v1
• Date: Mon, 22 Jun 2026 11:24:59 GMT
• Title: DART: Draft-Agreement Routing for Training-Free Adaptive Thinking Budgets in Hybrid Reasoning Models
• Abstract summary: We introduce DART, a training-free routing framework that samples two cheap no-think drafts.<n>DART accepts direct answering when the drafts agree, and predicts a thinking budget from draft entropy when they disagree.<n>On math reasoning, accuracy improves by up to $+$9.0 points on Olympiad-level problems while thinking tokens drop 15-69%.<n>On code reasoning under execution-based equivalence, accuracy improves by up to +22.5 points while thinking tokens drop 51-63%.
• Score: 45.71412371299232
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Hybrid reasoning models can answer directly or spend extra tokens on extended thinking. A practical router should choose between these modes for each query, so easy problems avoid unnecessary reasoning and hard problems receive enough budget to finish the answer. Existing routers move in this direction, but they typically require labeled training data or fix thinking budgets up front, ignoring answer-level evidence from the model itself. We introduce DART, a training-free routing framework that samples two cheap no-think drafts, accepts direct answering when the drafts agree, and predicts a thinking budget from draft entropy when they disagree. Across the main comparisons, DART preserves or improves always-thinking accuracy in most settings while reducing thinking-token use. On math reasoning, accuracy improves by up to $+$9.0 points on Olympiad-level problems while thinking tokens drop 15-69%. On code reasoning under execution-based equivalence, accuracy improves by up to +22.5 points while thinking tokens drop 51-63%. The Stage~1 signal extends across model scales (0.6B-32B), model families, and API-only hosted settings, with no labeled data and no gradient updates required.

Related papers

• CEPO: RLVR Self-Distillation using Contrastive Evidence Policy Optimization [50.59956036193097]
  A model produces a correct solution under reinforcement learning with verifiable rewards (RLVR)<n>Every token receives the same reward signal regardless of whether it was a decisive reasoning step or a grammatical filler.<n>We propose Contrastive Evidence Policy Optimization (CEPO)<n>CEPO asks a sharper question at every token: "does the correct answer favor this token?" but "does the correct answer favor it while the wrong answer disfavors it?"
  arXiv  Detail & Related papers  (2026-05-19T06:46:19Z)
• Hint Tuning: Less Data Makes Better Reasoners [30.274192211828723]
  Hint Tuning is a data-efficient approach that teaches models to calibrate reasoning depth.<n>By testing what the instruct model can solve with varying guidance, we automatically construct training data across three states.<n>We achieve superior efficiency through simple alignment with the instruct model's capabilities.
  arXiv  Detail & Related papers  (2026-05-09T04:07:16Z)
• e1: Learning Adaptive Control of Reasoning Effort [88.51897900019485]
  Increasing the thinking budget of AI models can significantly improve accuracy, but not all questions warrant the same amount of reasoning.<n>Users may prefer to allocate different amounts of reasoning effort depending on how they value output quality versus latency and cost.<n>We propose Adaptive Effort Control, a self-adaptive reinforcement learning method that trains models to use a user-specified fraction of tokens.
  arXiv  Detail & Related papers  (2025-10-30T23:12:21Z)
• Think Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression [68.69801176669843]
  We propose an online post-training RL method that prunes redundant steps and estimates difficulty.<n> TRAAC (Think Right with Adaptive, Attentive Compression) achieves an average absolute accuracy gain of 8.4%.<n>Although our models are trained on math datasets, they show accuracy and efficiency gains on out-of-distribution non-math datasets.
  arXiv  Detail & Related papers  (2025-10-02T02:00:20Z)
• Entropy After $\langle \texttt{/Think} \rangle$ for reasoning model early exiting [38.93424884988798]
  We show that large reasoning models overthink, continuing to revise answers even after reaching the correct solution.<n>We propose Entropy After /Think> (EAT) for monitoring and deciding whether to exit reasoning early.<n>EAT reduces token usage by 13 - 21% without harming accuracy.
  arXiv  Detail & Related papers  (2025-09-30T16:59:37Z)
• SynapseRoute: An Auto-Route Switching Framework on Dual-State Large Language Model [12.929385845055137]
  We show that approximately 58% of medical questions can be accurately answered by the non-thinking mode alone.<n>We propose SynapseRoute, a machine learning-based dynamic routing framework that intelligently assigns input queries to either thinking or non-thinking modes.
  arXiv  Detail & Related papers  (2025-07-03T17:33:58Z)
• Do Thinking Tokens Help or Trap? Towards More Efficient Large Reasoning Model [7.8354921036790275]
  Large Reasoning Models (LRMs) excel at solving complex problems but face an overthinking dilemma.<n>When handling simple tasks, they often produce verbose responses overloaded with thinking tokens.<n>These tokens trigger unnecessary high-level reasoning behaviors like reflection and backtracking, reducing efficiency.
  arXiv  Detail & Related papers  (2025-06-30T13:30:33Z)
• Steering LLM Thinking with Budget Guidance [48.65894557568655]
  Budget guidance is a method for steering the reasoning process of LLMs toward a target budget without requiring any fine-tuning.<n>Our approach introduces a lightweight predictor that models a Gamma distribution over the remaining thinking length.<n>This signal is then used to guide generation in a soft, token-level manner, ensuring that the overall reasoning trace adheres to the specified thinking budget.
  arXiv  Detail & Related papers  (2025-06-16T17:57:05Z)
• The Price of a Second Thought: On the Evaluation of Reasoning Efficiency in Large Language Models [54.88805865447848]
  We show that instruct models achieve higher efficiency overall, and problem difficulty affects efficiency.<n>We propose COTHINK, a simple two-stage pipeline: an instruct model drafts a brief outline, and a thinking model expands it.<n>On GSM8K, MATH500, and AIME24, COTHINK cuts token usage by 21.1% while keeping accuracy on four thinking models, and remains competitive with strong efficiency baselines.
  arXiv  Detail & Related papers  (2025-05-28T06:24:45Z)
• DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models [30.184895117009457]
  This paper introduces Difficulty-Adaptive Slow Thinking (DAST), a novel framework that enables models to autonomously adjust the length of Chain-of-Thought (CoT) based on problem difficulty.<n>Experiments on diverse datasets and model scales demonstrate that DAST effectively mitigates overthinking while preserving reasoning accuracy on complex problems.
  arXiv  Detail & Related papers  (2025-03-06T14:23:06Z)
• WR-ONE2SET: Towards Well-Calibrated Keyphrase Generation [57.11538133231843]
  Keyphrase generation aims to automatically generate short phrases summarizing an input document. The recently emerged ONE2SET paradigm generates keyphrases as a set and has achieved competitive performance. We propose WR-ONE2SET which extends ONE2SET with an adaptive instance-level cost Weighting strategy and a target Re-assignment mechanism.
  arXiv  Detail & Related papers  (2022-11-13T09:56:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.