mlsanityData to 7 Oct 2026

DISC: Dynamic Decomposition Improves LLM Inference Scaling

arXiv 2502.16706 ↗First posted 23 Feb 2025cs.LG cs.AI cs.CLcited by 9

Not on the map. Why?

Abstract (arXiv)

Inference scaling methods for LLMs often rely on decomposing problems into steps (or groups of tokens), followed by sampling and selecting the best next steps. However, these steps and their sizes are often predetermined or manually designed based on domain knowledge. We propose dynamic decomposition, a method that adaptively and automatically partitions solution and reasoning traces into manageable steps during inference. By more effectively allocating compute -- particularly through subdividing challenging steps and prioritizing their sampling -- dynamic decomposition significantly improves inference efficiency. Experiments on benchmarks such as APPS, MATH, and LiveCodeBench demonstrate that dynamic decomposition outperforms static approaches, including token-level, sentence-level, and single-step decompositions, reducing the pass@10 error rate by 5.0%, 6.7%, and 10.5% respectively. These findings highlight the potential of dynamic decomposition to improve a wide range of inference scaling techniques.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 54 of its 77 references that are in the map. First posted on arXiv 23 Feb 2025.

Builds on

54 of 77 references.

References · 54

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  3. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  4. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  5. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  6. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  7. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  8. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  9. 2022Large Language Models are Zero-Shot Reasonerscited by 4,323
  10. 2023Let's Verify Step by Stepcited by 3,598
  11. 2021Program Synthesis with Large Language Modelscited by 3,509
  12. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,155
38 more
  1. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,928
  2. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 1,906
  3. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,397
  4. 2024A Survey on LLM-as-a-Judgecited by 1,384
  5. 2025s1: Simple test-time scalingcited by 1,311
  6. 2023Teaching Large Language Models to Self-Debugcited by 963
  7. 2024Large Language Monkeys: Scaling Inference Compute with Repeated Samplingcited by 946
  8. 2023Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotationscited by 895
  9. 2023Reasoning with Language Model is Planning with World Modelcited by 846
  10. 2024Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMscited by 564
  11. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
  12. 2023Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Modelscited by 551
  13. 2024Generative Verifiers: Reward Modeling as Next-Token Predictioncited by 465
  14. 2022CodeT: Code Generation with Generated Testscited by 369
  15. 2023Alphazero-like Tree-Search can Guide Large Language Model Decoding and Trainingcited by 291
  16. 2024To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoningcited by 257
  17. 2022Making Large Language Models Better Reasoners with Step-Aware Verifiercited by 231
  18. 2024Chain-of-Thought Reasoning Without Promptingcited by 208
  19. 2023Self-Evaluation Guided Beam Search for Reasoningcited by 195
  20. 2024Stream of Search (SoS): Learning to Search in Languagecited by 141
  21. 2024LLM Critics Help Catch LLM Bugscited by 138
  22. 2022Evolution through Large Modelscited by 131
  23. 2023EvoPrompting: Language Models for Code-Level Neural Architecture Searchcited by 129
  24. 2022Successive Prompting for Decomposing Complex Questionscited by 94
  25. 2024Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capabilitycited by 80
  26. 2024Planning In Natural Language Improves LLM Search For Code Generationcited by 79
  27. 2024Adaptive Inference-Time Compute: LLMs Can Predict if They Can Do Better, Even Mid-Generationcited by 64
  28. 2022Parsel: Algorithmic Reasoning with Language Models by Composing Decompositionscited by 62
  29. 2025Evolving Deeper LLM Thinkingcited by 58
  30. 2024Scaling of Search and Learning: A Roadmap to Reproduce o1 from Reinforcement Learning Perspectivecited by 49
  31. 2024Evolving Code with A Large Language Modelcited by 42
  32. 2024Improving LLM Reasoning through Scaling Inference Computation with Collaborative Verificationcited by 40
  33. 2023Fully Autonomous Programming with Large Language Modelscited by 31
  34. 2024Forking Paths in Neural Text Generationcited by 30
  35. 2024From Medprompt to o1: Exploration of Run-Time Strategies for Medical Challenge Problems and Beyondcited by 24
  36. 2020Divide-and-Conquer Monte Carlo Tree Search For Goal-Directed Planningcited by 19
  37. 2024Divide-and-Conquer Meets Consensus: Unleashing the Power of Functions in Code Generationcited by 19
  38. 2024Strategist: Self-improvement of LLM Decision Making via Bi-Level Tree Searchcited by 16

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 5; 2026: 4.

9 papers.

  1. 2025Efficient Reasoning Models: A Surveycited by 93
  2. 2025Every Rollout Counts: Optimal Resource Allocation for Efficient Test-Time Scalingcited by 15
  3. 2025TDRM: Smooth Reward Models with Temporal Difference for LLM RL and Inferencecited by 8
  4. 2025Fact-Augmented Lookahead Planning for LLM Agentscited by 7
  5. 2025DynScaling: Efficient Verifier-free Inference Scaling via Dynamic and Integrated Samplingcited by 7
  6. 2026Actor-Curator: Co-adaptive Curriculum Learning via Policy-Improvement Bandits for RL Post-Trainingcited by 7
  7. 2026More Bang for the Buck: Improving the Inference of Large Language Models at a Fixed Budget using Reset and Discard (ReD)cited by 1
  8. 2026Harness as a Language: A Minimalist Agent Framework With Maximal Expressivitycited by 0
  9. 2026Shockingly Simple Self-retrospection Improves Agentic Models Without RLcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Refining Over Resampling: Test-Time Self-Correction for LLM Reasoning
  2. 2025The Art of Scaling Test-Time Compute for Large Language Models
  3. 2026ThinkBooster: A Unified Framework for Seamless Test-Time Scaling of LLM Reasoning
  4. 2025From Long to Short: LLMs Excel at Trimming Own Reasoning Chains
  5. 2026Test-Time Scaling in Reasoning LLMs: Inference Regimes, Evaluation, and Reproducibility
  6. 2025A*-Decoding: Token-Efficient Inference Scaling
  7. 2025Prompting Test-Time Scaling Is A Strong LLM Reasoning Data Augmentation
  8. 2024Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Models