Fewer Tokens, Smaller Cache: Reward-Coordinated Efficient Reasoning

• URL: http://arxiv.org/abs/2608.04771v1
• Date: Wed, 05 Aug 2026 12:36:28 GMT
• Title: Fewer Tokens, Smaller Cache: Reward-Coordinated Efficient Reasoning
• Abstract summary: Large Reasoning Models excel on complex tasks through long chain-of-thought (CoT) reasoning.<n> KV-cache compression is a common solution, yet existing reasoning-oriented methods judge compression only by what it removes from the cache.<n>We propose ReCo, a framework in which a lightweight process-reward estimator scores each completed step and drives three components.
• Score: 15.652256738395018
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Reasoning Models (LRMs) excel on complex tasks through long chain-of-thought (CoT) reasoning, but their lengthy intermediate steps cause severe overthinking that inflates inference cost. KV-cache compression is a common solution, yet existing reasoning-oriented methods apply a uniform policy across the trajectory and judge compression only by what it removes from the cache. Two observations point the other way. First, a reasoning state's tolerance to context loss varies along the trajectory, and process reward tracks it: deleting tokens at high-reward steps preserves accuracy far better than deleting the same budget at random. Second, compression is not free on the generation side, since a smaller cache leads the model to generate more tokens, partly canceling the saving. Together these motivate coordinating both sides under a single process reward. We propose ReCo (Reward-Coordinated Compression), a step-wise framework in which a lightweight process-reward estimator scores each completed step and drives three components: (1) reward-adaptive KV-cache compression that shrinks the retained cache harder at high-reward steps and less at low-reward ones, (2) a reward-banded penalty on reflection tokens that curbs redundant generation, and (3) confidence-based early stopping that triggers when the reasoning is reliable. Across three reasoning models and six benchmarks, ReCo reduces generated tokens by 37%-65% and end-to-end latency by 2.08x-2.35x over Full CoT, all while largely preserving accuracy.

Related papers

• ShortOPD: Recovering Pruned LLMs with Short-to-Long On-Policy Distillation [72.54604107217669]
  textscpass@$k$ nearly vanishes after compression, yet textscpass@$k$ recovers substantially under repeated sampling.<n>Recovery should train on the compressed model's own on-policy states with dense token-level supervision.<n>We propose textbfshortopd, a short-to-long OPD schedule that detects teacher-confirmed repetitive suffixes.
  arXiv  Detail & Related papers  (2026-07-14T17:50:50Z)
• SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation [0.0]
  Large Language Models generate text autoregressively, relying on a key-value (KV) cache whose memory footprint grows linearly with context length.<n>Recent compression methods mitigate this cost via token merging.<n>We propose a training-free, dual-component framework for KV cache compression that addresses these limitations.
  arXiv  Detail & Related papers  (2026-05-19T17:44:56Z)
• 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)
• Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression [55.63153956934198]
  Chain-of-Thought (CoT) reasoning successfully enhances the reasoning capabilities of Large Language Models (LLMs)<n>Existing CoT compression methods often suffer from a critical loss of logical fidelity at high compression ratios.<n>We propose a novel EXTreme-RAtio Chain-of-Thought Compression framework, termed Extra-CoT, which aggressively reduces the token budget while preserving answer accuracy.
  arXiv  Detail & Related papers  (2026-02-09T06:57:15Z)
• CtrlCoT: Dual-Granularity Chain-of-Thought Compression for Controllable Reasoning [29.057579417751203]
  Chain-of-thought (CoT) prompting improves LLM reasoning but incurs high latency and memory cost due to verbose traces.<n>We propose textbfCtrlCoT, a dual-granularity CoT compression framework that harmonizes semantic abstraction and token-level pruning.
  arXiv  Detail & Related papers  (2026-01-28T10:38:49Z)
• Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models [56.055015597319674]
  Reinforcement learning with verifiable rewards (RLVR) is effective to improve the reasoning ability of large language models (LLMs)<n>Recent self-rewarding methods investigate a label-free alternative to unlock the reasoning capabilities of LLMs.<n>We propose textitCo-rewarding, a novel self-supervised RL framework that improves training stability by seeking complementary supervision from another views.
  arXiv  Detail & Related papers  (2025-08-01T08:09:14Z)
• 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)
• Krul: Efficient State Restoration for Multi-turn Conversations with Dynamic Cross-layer KV Sharing [41.792908098945766]
  We present Krul, a multi-turn LLM inference system that enables accurate and efficient KV cache restoration.<n>Krul selects compression strategies based on attention similarity across layer pairs and uses a recomputation-loading pipeline to restore the KV cache.<n>It achieves a 1.5x-2.68x reduction in time-to-first-token (TTFT) and a 1.33x-2.35x reduction in KV cache storage.
  arXiv  Detail & Related papers  (2025-07-10T01:51:17Z)
• Fractured Chain-of-Thought Reasoning [61.647243580650446]
  We introduce Fractured Sampling, a unified inference-time strategy that interpolates between full CoT and solution-only sampling.<n>We show that Fractured Sampling consistently achieves superior accuracy-cost trade-offs, yielding steep log-linear scaling gains in Pass@k versus token budget.
  arXiv  Detail & Related papers  (2025-05-19T11:30:41Z)
• Sparsity Forcing: Reinforcing Token Sparsity of MLLMs [40.93786579652003]
  We explicitly reinforce token sparsity in well-posed multimodal large language models (MLLMs) through a simple RL-based post-training framework named textitSparsity Forcing.<n>Our method explores the efficiency-accuracy trade-off by running multiple rollouts with different token budgets, where both efficiency (token reduction ratio) and performance (answer correctness) are formulated as joint rewards.
  arXiv  Detail & Related papers  (2025-04-23T01:45:55Z)
• CoT-Valve: Length-Compressible Chain-of-Thought Tuning [50.196317781229496]
  We introduce a new tuning and inference strategy named CoT-Valve, designed to allow models to generate reasoning chains of varying lengths.<n>We show that CoT-Valve successfully enables controllability and compressibility of the chain and shows better performance than the prompt-based control.
  arXiv  Detail & Related papers  (2025-02-13T18:52:36Z)

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.