mlsanityData to 7 Oct 2026

Short Chains, Deep Thoughts: Balancing Reasoning Efficiency and Intra-Segment Capability via Split-Merge Optimization

arXiv 2602.03141 ↗First posted 3 Feb 2026cs.CLcited by 1

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

While Large Reasoning Models (LRMs) have demonstrated impressive capabilities in solving complex tasks through the generation of long reasoning chains, this reliance on verbose generation results in significant latency and computational overhead. To address these challenges, we propose CoSMo (Consistency-Guided Split-Merge Optimization), a framework designed to eliminate structural redundancy rather than indiscriminately restricting token volume. Specifically, CoSMo utilizes a split-merge algorithm that dynamically refines reasoning chains by merging redundant segments and splitting logical gaps to ensure coherence. We then employ structure-aligned reinforcement learning with a novel segment-level budget to supervise the model in maintaining efficient reasoning structures throughout training. Extensive experiments across multiple benchmarks and backbones demonstrate that CoSMo achieves superior performance, improving accuracy by 3.3 points while reducing segment usage by 28.7% on average compared to reasoning efficiency baselines.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 26 of its 49 references that are in the map. First posted on arXiv 3 Feb 2026.

Builds on

26 of 49 references.

Closest ancestors · 12

Same direction, newest first

  1. 2025Making Slow Thinking Faster: Compressing LLM Chain-of-Thought via Step Entropy5 Aug
  2. 2025ReCUT: Balancing Reasoning Length and Accuracy in LLMs via Stepwise Trails and Preference Optimization12 Jun
  3. 2025Efficient Reasoning Models: A Survey15 Apr
  4. 2025ThinkPrune: Pruning Long Chain-of-Thought of LLMs via Reinforcement Learning2 Apr
  5. 2025Efficient Inference for Large Reasoning Models: A Survey29 Mar
  6. 2025Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models20 Mar
  7. 2025L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning6 Mar
  8. 2025DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models6 Mar
  9. 2025Self-Training Elicits Concise Reasoning in Large Language Models27 Feb
  10. 2025TokenSkip: Controllable Chain-of-Thought Compression in LLMs17 Feb
  11. 2025CoT-Valve: Length-Compressible Chain-of-Thought Tuning13 Feb
  12. 2024Training Language Models to Self-Correct via Reinforcement Learning19 Sep

Other references · 14

  1. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  2. 2024OpenAI o1 System Cardcited by 1,220
  3. 2023Gemini: A Family of Highly Capable Multimodal Modelscited by 810
  4. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 655
  5. 2023HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Modelscited by 329
  6. 2025Chain of Draft: Thinking Faster by Writing Lesscited by 290
  7. 2025Stepwise Perplexity-Guided Refinement for Efficient Chain-of-Thought Reasoning in Large Language Modelscited by 51
  8. 2024Rethinking Data Selection for Supervised Fine-Tuningcited by 35
  9. 2025Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignmentcited by 35
  10. 2024Break the Chain: Large Language Models Can be Shortcut Reasonerscited by 32
  11. 2025HyperTree Planning: Enhancing LLM Reasoning via Hierarchical Thinkingcited by 14
  12. 2025Bridging the Semantic-Action Gap in Visual Token Pruning for Efficient VLA Inferencecited by 14
2 more
  1. 2024Benchmarking End-To-End Performance of AI-Based Chip Placement Algorithmscited by 12
  2. 2025Hop, Skip, and Overthink: Diagnosing Why Reasoning Models Fumble during Multi-Hop Analysiscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026ThoughtFold: Folding Reasoning Chains via Introspective Preference Learningcited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026OS-Pruner: Pruning Chains-of-Thought of Reasoning Models via Optimal Stopping
  2. 2026The Art of Efficient Reasoning: Data, Reward, and Optimization
  3. 2026Batched Contextual Reinforcement: A Task-Scaling Law for Efficient Reasoning
  4. 2026SLAT: Segment-Level Adaptive Trimming for Efficient CoT Reasoning
  5. 2026ChainPrune: Evaluating and Reducing Redundancy in Long Chain-of-Thought Reasoning
  6. 2024Fine-Tuning on Diverse Reasoning Chains Drives Within-Inference CoT Refinement in LLMs
  7. 2025Why Does Reasoning Length Converge? Unveiling the Underfitting-Overfitting Trade-off in Chain-of-Thought
  8. 2026TERMINATOR: Learning Optimal Exit Points for Early Stopping in Chain-of-Thought Reasoning