mlsanityData to 7 Oct 2026

OpenCodeReasoning-II: A Simple Test Time Scaling Approach via Self-Critique

arXiv 2507.09075 ↗First posted 11 Jul 2025cs.CLcited by 24

On the map: LLM reasoning → Reasoning for agents, planning and code

Abstract (arXiv)

Recent advancements in reasoning-based Large Language Models (LLMs), particularly their potential through test-time scaling, have created significant opportunities for distillation in code generation and critique. However, progress in both areas fundamentally depends on large-scale, high-quality datasets. In this work, we introduce OpenCodeReasoning-II, a dataset consists of 2.5M question-solution-critique triples (approx. 35K unique programming questions), making it nearly twice the size of the previous largest publicly available code reasoning dataset. In this work, we employ a two-stage supervised fine-tuning strategy. The first stage focuses on fine-tuning for code generation, while the second stage involves the joint training of models for both code generation and critique. Our resulting finetuned Qwen2.5-Instruct models achieve performance in code generation that either exceeds or equals the best prior open-weight distilled models. Notably, the integration of our code generation and critique models leads to significant improvements in competitive coding performance. Furthermore, we present an extension of the LiveCodeBench benchmark to specifically support the C++ programming language, thereby facilitating more comprehensive LLM evaluation using this benchmark.

Key facts. Cited by 24 papers in the map (data through 7 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 43 of its 66 references that are in the map. First posted on arXiv 11 Jul 2025.

Builds on

43 of 66 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025OpenCodeReasoning: Advancing Data Distillation for Competitive Coding2 Apr
  2. 2025KodCode: A Diverse, Challenging, and Verifiable Synthetic Dataset for Coding4 Mar
  3. 2025RefineCoder: Iterative Improving of Large Language Models via Adaptive Critique Refinement for Code Generation13 Feb
  4. 2025ACECODER: Acing Coder RL via Automated Test-Case Synthesis3 Feb
  5. 2022Competition-Level Code Generation with AlphaCode8 Feb

Other references · 38

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2024The Llama 3 Herd of Modelscited by 14,192
  3. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  4. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  5. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  6. 2021Program Synthesis with Large Language Modelscited by 3,509
  7. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,678
  8. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,057
  9. 2024Large Language Monkeys: Scaling Inference Compute with Repeated Samplingcited by 946
  10. 2023Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotationscited by 895
  11. 2024Generative Verifiers: Reward Modeling as Next-Token Predictioncited by 465
  12. 2025The Lessons of Developing Process Reward Models in Mathematical Reasoningcited by 402
26 more
  1. 2024Skywork-Reward: Bag of Tricks for Reward Modeling in LLMscited by 322
  2. 2025From System 1 to System 2: A Survey of Reasoning Large Language Modelscited by 319
  3. 2024Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Modelscited by 282
  4. 2025Inference-Time Scaling for Generalist Reward Modelingcited by 233
  5. 2024HelpSteer2: Open-source dataset for training top-performing reward modelscited by 232
  6. 2025A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?cited by 215
  7. 2023Rethinking Benchmark and Contamination for Language Models with Rephrased Samplescited by 192
  8. 2025AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning datasetcited by 165
  9. 2025Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMscited by 162
  10. 2025Llama-Nemotron: Efficient Reasoning Modelscited by 151
  11. 2023TACO: Topics in Algorithmic COde generation datasetcited by 150
  12. 2025RM-R1: Reward Modeling as Reasoningcited by 149
  13. 2024Generative Reward Modelscited by 148
  14. 2024Nemotron-4 340B Technical Reportcited by 143
  15. 2025LLMs Can Easily Learn to Reason from Demonstrations Structure, not content, is what matters!cited by 132
  16. 2024HelpSteer2-Preference: Complementing Ratings with Preferencescited by 129
  17. 2025Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuningcited by 121
  18. 2025Scaling Test-Time Compute Without Verification or RL is Suboptimalcited by 82
  19. 2024AceMath: Advancing Frontier Math Reasoning with Post-Training and Reward Modelingcited by 76
  20. 2025Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities?cited by 69
  21. 2025Critique Fine-Tuning: Learning to Critique is More Effective than Learning to Imitatecited by 61
  22. 2025OpenCodeInstruct: A Large-scale Instruction Tuning Dataset for Code LLMscited by 57
  23. 2024Are More LLM Calls All You Need? Towards Scaling Laws of Compound Inference Systemscited by 24
  24. 2024Genetic Instruct: Scaling up Synthetic Generation of Coding Instructions for Large Language Modelscited by 17
  25. 2025Scoring Verifiers: Evaluating Synthetic Verification for Code and Reasoningcited by 8
  26. 2024InverseCoder: Self-improving Instruction-Tuned Code LLMs with Inverse-Instructcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 8; 2026: 16.

24 papers.

Sort:

By area:

  1. 2025Klear-Reasoner: Advancing Reasoning Capability via Gradient-Preserving Clipping Policy Optimizationcited by 39
  2. 2025AgentMath: Empowering Mathematical Reasoning for Large Language Models via Tool-Augmented Agentcited by 17
  3. 2025Don't Pass@k: A Bayesian Framework for Large Language Model Evaluationcited by 15
  4. 2025From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligencecited by 13
  5. 2026F2LLM-v2: Inclusive, Performant, and Efficient Embeddings for a Multilingual Worldcited by 10
  6. 2026Distilled Reinforcement Learning for LLM Post-trainingcited by 7
  7. 2025MobileLLM-R1: Exploring the Limits of Sub-Billion Language Model Reasoners with Open Training Recipescited by 6
  8. 2026ASTER: Agentic Scaling with Tool-integrated Extended Reasoningcited by 5
  9. 2026Test-Time Scaling in Reasoning LLMs: Inference Regimes, Evaluation, and Reproducibilitycited by 4
  10. 2025When Reasoning Meets Its Lawscited by 3
  11. 2026Contextual Drag: How Errors in the Context Affect LLM Reasoningcited by 3
  12. 2025The Valley of Code Reasoning: Scaling Knowledge Distillation of Large Language Modelscited by 2
12 more
  1. 2026Learning Generative Selection for Best-of-Ncited by 2
  2. 2025Motif-2-12.7B-Reasoning: A Practitioner's Guide to RL Training Recipescited by 1
  3. 2026A Primer in Post-Training Reasoning Data: What We Know About How It Workscited by 1
  4. 2026Synthesizing File-Level Data for Unit Test Generation with Chain-of-Thoughts via Self-Debuggingcited by 0
  5. 2026The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignmentcited by 0
  6. 2026Length Value Model: Scalable Value Pretraining for Token-Level Length Modelingcited by 0
  7. 2026Primal Generation, Dual Judgment: Self-Training from Test-Time Scalingcited by 0
  8. 2026ML-Embed: Inclusive and Efficient Embeddings for a Multilingual Worldcited by 0
  9. 2026Post-training is (Massive) Supervised Learningcited by 0
  10. 2026The Generalization Spectrum: A Chromatographic Approach to Evaluating Learning Algorithmscited by 0
  11. 2026Don't Let Gains FADE: Breaking Down Policy Gradient Weights in RLcited by 0
  12. 2026Post-Training Language Models for Gold-Medal Performance in Coding Competitionscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Z1: Efficient Test-time Scaling with Code
  2. 2023Turbulence: Systematically and Automatically Testing Instruction-Tuned Large Language Models for Code
  3. 2025Infinite-Instruct: Synthesizing Scaling Code instruction Data with Bidirectional Synthesis and Static Verification
  4. 2025CodeChemist: Test-Time Scaling for Low-Resource Code Generation via Functional Knowledge Transfer
  5. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
  6. 2025A Code Comprehension Benchmark for Large Language Models for Code
  7. 2025CodeCriticBench: A Holistic Code Critique Benchmark for Large Language Models
  8. 2025rStar-Coder: Scaling Competitive Code Reasoning with a Large-Scale Verified Dataset