Reinforcement Learning for Chain of Thought Compression with One-Domain-to-All Generalization

• URL: http://arxiv.org/abs/2601.06052v2
• Date: Wed, 21 Jan 2026 06:34:10 GMT
• Title: Reinforcement Learning for Chain of Thought Compression with One-Domain-to-All Generalization
• Abstract summary: Chain-of-thought reasoning in large language models can trigger an "overthinking trap"<n>We propose mastery-gated, sample-level, soft reinforcement learning compression.<n>It penalizes long rollouts only when the model already solves the problem and has produced a shorter rollout.
• Score: 27.13234369037689
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Chain-of-thought reasoning in large language models can trigger an "overthinking trap": longer rollouts raise cost and latency yet often yield unreliable accuracy gains. Existing methods use global, static controls that may suppress needed reasoning. We propose mastery-gated, sample-level, soft reinforcement learning compression that penalizes long rollouts only when the model already solves the problem and has produced a shorter rollout. Across benchmarks, it cuts response length by 20-40% with comparable or higher accuracy and generalizes across domains: a model trained on math spontaneously shortens unseen tasks (code, instruction following, general-knowledge QA) without hurting accuracy. We further show two-way transfer between non-agent CoT and tool-use agents: non-agent training reduces SWE-Bench Verified rounds by 13%, while compressing a thinking agent cuts SWE trajectories by 67% tokens and 52% rounds and shortens non-agent outputs by up to 44%. Compression is thus not cosmetic brevity, but an inherent computation policy -- what to keep, and what to forget.

Related papers

• ExpThink: Experience-Guided Reinforcement Learning for Adaptive Chain-of-Thought Compression [14.830831997228657]
  Large reasoning models (LRMs) achieve strong performance via extended chain-of-thought (CoT) reasoning, yet suffer from excessive token consumption and high inference latency.<n>We propose textbfExpThinkxspace, an RL framework that addresses both dimensions through two complementary mechanisms.<n> Experiments on multiple mathematical reasoning benchmarks demonstrate that textbfExpThinkxspace reduces average response length by up to 77% while simultaneously improving accuracy.
  arXiv  Detail & Related papers  (2026-05-08T09:37:30Z)
• Implicit Compression Regularization: Concise Reasoning via Internal Shorter Distributions in RL Post-Training [11.132427208920424]
  We propose an on-policy regularization method whose compression signal comes from a virtual shorter distribution induced by the shortest correct responses in rollout groups.<n>Experiments on three reasoning backbones and multiple mathematical and knowledge-intensive benchmarks show that ICR consistently shortens responses while preserving or improving accuracy.
  arXiv  Detail & Related papers  (2026-05-08T06:25:08Z)
• Stepwise Penalization for Length-Efficient Chain-of-Thought Reasoning [66.22060690012512]
  Large reasoning models improve with more test-time computation, but often overthink, producing unnecessarily long chains-of-thought that raise cost without improving accuracy.<n>We propose Step-wise Adaptive Penalization (SWAP), a fine-grained framework that allocates length reduction across steps based on intrinsic contribution.
  arXiv  Detail & Related papers  (2026-02-27T20:23:59Z)
• Correct, Concise and Complete: Multi-stage Training For Adaptive Reasoning [11.179446105672461]
  We propose a multi-stage efficient reasoning method that combines supervised fine-tuning and reinforcement learning.<n>Our approach reduces response length by an average of 28% for 8B models and 40% for 32B models.<n>It achieves a superior trade-off compared to more complex state-of-the-art efficient reasoning methods.
  arXiv  Detail & Related papers  (2026-01-06T12:31:51Z)
• TeaRAG: A Token-Efficient Agentic Retrieval-Augmented Generation Framework [62.66056331998838]
  TeaRAG is a token-efficient agentic RAG framework capable of compressing both retrieval content and reasoning steps.<n>Our reward function evaluates the knowledge sufficiency by a knowledge matching mechanism, while penalizing excessive reasoning steps.
  arXiv  Detail & Related papers  (2025-11-07T16:08:34Z)
• DLER: Doing Length pEnalty Right - Incentivizing More Intelligence per Token via Reinforcement Learning [134.03095505580276]
  Doing Length pEnalty Right (DLER) is a training recipe combining batch-wise reward normalization, higher clipping, dynamic sampling, and a simple truncation length penalty.<n>DLER achieves state-of-the-art accuracy--efficiency trade-offs, cutting output length by over 70 percent while surpassing all previous baseline accuracy.
  arXiv  Detail & Related papers  (2025-10-16T20:05:57Z)
• Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking [50.97239453902612]
  Large Reasoning Models (LRMs) have achieved impressive performance on challenging tasks, yet their deep reasoning often incurs substantial computational costs.<n>Inspired by Evidence Accumulation Models, we find that LRMs have accumulated sufficient information early in reasoning, making further reasoning steps redundant.<n>We propose Just-Enough Thinking (JET), which trains models to proactively terminate unnecessary reasoning.
  arXiv  Detail & Related papers  (2025-09-27T16:25:06Z)
• Reasoning Efficiently Through Adaptive Chain-of-Thought Compression: A Self-Optimizing Framework [10.148124073650349]
  Chain-of-Thought (CoT) reasoning enhances Large Language Models (LLMs)<n>Longer outputs increase latency, memory usage, and KV-cache demands.<n>We propose SEER (Self-Enhancing Efficient Reasoning), an adaptive framework that compresses CoT while preserving accuracy.
  arXiv  Detail & Related papers  (2025-09-17T15:33:44Z)
• R-Stitch: Dynamic Trajectory Stitching for Efficient Reasoning [80.104336426172]
  Chain-of-thought (CoT) enhances problem-solving ability of large language models.<n>CoT incurs substantial inference cost due to long autoregressive trajectories.<n>We introduce R-Stitch, a training-free hybrid decoding framework.
  arXiv  Detail & Related papers  (2025-07-23T08:14:36Z)
• Answer Convergence as a Signal for Early Stopping in Reasoning [7.51755942515969]
  Chain-of-thought (CoT) prompting enhances reasoning in large language models (LLMs)<n>We propose three inference-time strategies to improve efficiency: (1) early stopping via answer consistency, (2) boosting the probability of generating end-of-reasoning signals, and (3) a supervised method that learns when to stop based on internal activations.
  arXiv  Detail & Related papers  (2025-06-03T07:20:54Z)
• VeriThinker: Learning to Verify Makes Reasoning Model Efficient [52.74493506816969]
  Large Reasoning Models excel at complex tasks using Chain-of-Thought (CoT) reasoning.<n>Their tendency to overthinking leads to unnecessarily lengthy reasoning chains.<n>We introduce VeriThinker, a novel approach for CoT compression.
  arXiv  Detail & Related papers  (2025-05-23T14:17:56Z)
• Think Silently, Think Fast: Dynamic Latent Compression of LLM Reasoning Chains [24.805434364781306]
  We introduce Compressed Latent Reasoning (CoLaR), a novel framework that dynamically compresses reasoning processes in latent space.<n>CoLaR achieves 14.1% higher accuracy than latent-based baseline methods at comparable compression ratios.<n>Our RL-enhanced CoLaR demonstrates performance gains of up to 5.4% while dramatically reducing latent reasoning chain length by 82.8%.
  arXiv  Detail & Related papers  (2025-05-22T11:40:26Z)

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.