mlsanityData to 7 Oct 2026

Divide-and-Conquer CoT: RL for Reducing Latency via Parallel Reasoning

arXiv 2601.23027 ↗First posted 30 Jan 2026cs.LGcited by 1

On the map: LLM reasoning → Prompting and chain-of-thought reasoning

Abstract (arXiv)

Long chain-of-thought reasoning (Long CoT) is now fundamental to state-of-the-art LLMs, especially in mathematical reasoning. However, LLM generation is highly sequential, and long CoTs lead to a high latency. We propose to train Divide-and-Conquer CoT (DC-CoT) to reduce the latency. With DC-CoT, the model can act as a director that identifies distinct subtasks that can be performed in parallel in its reasoning process, and then spawns workers to execute the subtasks. Our goal is to achieve high accuracy, with a low longest path length, which is a theoretical measure of the latency needed for the response. We start with a long CoT base model (DeepScaleR-1.5B-Preview), and first use SFT with a small curated demonstration set to initialize its ability to spawn workers in a certain format. Because SFT degrades the accuracy significantly, we design a multi-stage RL algorithm, with various data filtering strategies, to recover the accuracy while decreasing the longest path length. Across several benchmarks including AIME 2024 and HMMT 2025, DC-CoT achieves similar accuracy as DeepScaleR-1.5B-Preview while decreasing longest path length by 35-40%. Our code, SFT dataset and models are publicly available at https://github.com/amahankali10/DC_CoT_RL_for_Low_Latency_CoT_with_Parallel_Reasoning.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 59 of its 77 references that are in the map. First posted on arXiv 30 Jan 2026.

Builds on

59 of 77 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025PENCIL: Long Thoughts with Short Memory18 Mar
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Models21 Mar

Other references · 57

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  3. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  4. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,155
  5. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,678
  6. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 1,980
  7. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,928
  8. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
  9. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,615
  10. 2022Solving Quantitative Reasoning Problems with Language Modelscited by 1,359
  11. 2024OpenAI o1 System Cardcited by 1,220
  12. 2023Accelerating Large Language Model Decoding with Speculative Samplingcited by 975
38 more
  1. 2024Large Language Monkeys: Scaling Inference Compute with Repeated Samplingcited by 946
  2. 2024Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Headscited by 821
  3. 2025Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoningcited by 540
  4. 2025The Entropy Mechanism of Reinforcement Learning for Reasoning Language Modelscited by 463
  5. 2025GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learningcited by 449
  6. 2025OpenThoughts: Data Recipes for Reasoning Modelscited by 287
  7. 2025MathArena: Evaluating LLMs on Uncontaminated Math Competitionscited by 275
  8. 2025MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attentioncited by 248
  9. 2025MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agentscited by 226
  10. 2025ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Modelscited by 187
  11. 2024Imitate, Explore, and Self-Improve: A Reproduction Report on Slow-thinking Reasoning Systemscited by 162
  12. 2025Skywork Open Reasoner 1 Technical Reportcited by 161
  13. 2024Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8Bcited by 156
  14. 2025Scaling Long-Horizon LLM Agent via Context-Foldingcited by 136
  15. 2025OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automationcited by 129
  16. 2025Towards a Science of Scaling Agent Systemscited by 125
  17. 2023Skeleton-of-Thought: Prompting LLMs for Efficient Parallel Generationcited by 103
  18. 2025Multi-Agent Collaboration via Evolving Orchestrationcited by 102
  19. 2025The Art of Scaling Reinforcement Learning Compute for LLMscited by 91
  20. 2025Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoningcited by 82
  21. 2025Latent Collaboration in Multi-Agent Systemscited by 67
  22. 2025Parallel-R1: Towards Parallel Thinking via Reinforcement Learningcited by 59
  23. 2025Learning Adaptive Parallel Reasoning with Language Modelscited by 58
  24. 2025JustRL: Scaling a 1.5B LLM with a Simple RL Recipecited by 57
  25. 2025e3: Learning to Explore Enables Extrapolation of Test-Time Compute for LLMscited by 56
  26. 2025Part I: Tricks or Traps? A Deep Dive into RL for LLM Reasoningcited by 53
  27. 2025The Majority is not always right: RL training for solution aggregationcited by 46
  28. 2024Fine-Tuning and Prompt Optimization: Two Great Steps that Work Better Togethercited by 38
  29. 2025ParaThinker: Native Parallel Thinking as a New Paradigm to Scale LLM Test-time Computecited by 36
  30. 2025DLER: Doing Length pEnalty Right - Incentivizing More Intelligence per Token via Reinforcement Learningcited by 34
  31. 2025Learning to Reason Across Parallel Samples for LLM Reasoningcited by 32
  32. 2025Multiverse: Your Language Models Secretly Decide How to Parallelize and Merge Generationcited by 30
  33. 2025SFR-DeepResearch: Towards Effective Reinforcement Learning for Autonomously Reasoning Single Agentscited by 29
  34. 2025Single-agent or Multi-agent Systems? Why Not Both?cited by 28
  35. 2025Learning to Keep a Promise: Scaling Language Model Decoding Parallelism with Learned Asynchronous Decodingcited by 26
  36. 2025FastCuRL: Curriculum Reinforcement Learning with Stage-wise Context Scaling for Efficient Training R1-like Reasoning Modelscited by 26
  37. 2024APAR: LLMs Can Do Auto-Parallel Auto-Regressive Decodingcited by 23
  38. 2025BroRL: Scaling Reinforcement Learning via Broadened Explorationcited by 23

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026FluxLite: Inference-Time Proposal Control for Discrete Diffusion Modelscited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2025Efficient Long CoT Reasoning in Small Language Models
  2. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models
  3. 2025One Missing Piece for Open-Source Reasoning Models: A Dataset to Mitigate Cold-Starting Short CoT LLMs in RL
  4. 2025Demystifying Long Chain-of-Thought Reasoning in LLMs
  5. 2025Deconstructing Long Chain-of-Thought: A Structured Reasoning Optimization Framework for Long CoT Distillation
  6. 2025Ada-R1: Hybrid-CoT via Bi-Level Adaptive Reasoning Optimization
  7. 2025RedStar: Does Scaling Long-CoT Data Unlock Better Slow-Reasoning Systems?
  8. 2025Towards Thinking-Optimal Scaling of Test-Time Compute for LLM Reasoning