mlsanityData to 14 Oct 2026

rStar-Coder: Scaling Competitive Code Reasoning with a Large-Scale Verified Dataset

arXiv 2505.21297 ↗First posted 27 May 2025cs.CLcited by 51

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

Abstract (arXiv)

Advancing code reasoning in large language models (LLMs) is fundamentally limited by the scarcity of high-difficulty datasets, especially those with verifiable input-output test cases necessary for rigorous solution validation at scale. We introduce rStar-Coder, which significantly improves LLM code reasoning capabilities by constructing a large-scale, verified dataset of 418K competition-level code problems, 580K long-reasoning solutions along with rich test cases of varying difficulty. This is achieved through three core contributions: (1) we curate competitive programming code problems and oracle solutions to synthesize new, solvable problems; (2) we introduce a reliable input-output test case synthesis pipeline that decouples the generation into a three-step input generation method and a mutual verification mechanism for effective output labeling; (3) we augment problems with high-quality, test-case-verified long-reasoning solutions. Extensive experiments on Qwen models (1.5B-14B) across various code reasoning benchmarks demonstrate the superiority of rStar-Coder dataset, achieving leading performance comparable to frontier reasoning LLMs with much smaller model sizes. On LiveCodeBench, rStar-Coder improves Qwen2.5-7B from 17.4% to an impressive 57.3%, and Qwen2.5-14B from 23.3% to 62.5%, surpassing o3-mini (low) by3.1%. On the more challenging USA Computing Olympiad, our 7B model achieves an average pass@1 accuracy of 16.15%, outperforming the frontier-level QWQ-32B. Code and the dataset will be released at https://github.com/microsoft/rStar.

Key facts. Cited by 51 papers in the map (data through 14 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 15 of its 35 references that are in the map. First posted on arXiv 27 May 2025.

Builds on

15 of 35 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025OpenCodeReasoning: Advancing Data Distillation for Competitive Coding2 Apr
  2. 2025ACECODER: Acing Coder RL via Automated Test-Case Synthesis3 Feb
  3. 2024OpenAI o1 System Card21 Dec
  4. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code12 Mar
  5. 2023Enhancing Large Language Models in Coding Through Multi-Perspective Self-Consistency29 Sep

Other references · 10

  1. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  2. 2021Measuring Coding Challenge Competence With APPScited by 902
  3. 2022CodeT: Code Generation with Generated Testscited by 384
  4. 2025rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinkingcited by 313
  5. 2023TACO: Topics in Algorithmic COde generation datasetcited by 160
  6. 2025Phi-4-reasoning Technical Reportcited by 151
  7. 2024OpenCoder: The Open Cookbook for Top-Tier Code Large Language Modelscited by 130
  8. 2024Can Language Models Solve Olympiad Programming?cited by 72
  9. 2025SRPO: A Cross-Domain Implementation of Large-Scale Reinforcement Learning on LLMcited by 68
  10. 2024Key-Point-Driven Data Synthesis with its Enhancement on Mathematical Reasoningcited by 58

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 26; 2026: 25.

51 papers.

Sort:

By area:

  1. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 180
  2. 2026Embarrassingly Simple Self-Distillation Improves Code Generationcited by 40
  3. 2025LoongRL: Reinforcement Learning for Advanced Reasoning over Long Contextscited by 29
  4. 2025HardTests: Synthesizing High-Quality Test Cases for LLM Codingcited by 23
  5. 2025A Survey on Agentic Multimodal Large Language Modelscited by 23
  6. 2025RPG: A Repository Planning Graph for Unified and Scalable Codebase Generationcited by 19
  7. 2025From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligencecited by 15
  8. 2025OpenDataArena: A Fair and Open Arena for Benchmarking Post-Training Dataset Valuecited by 15
  9. 2025Xolver: Multi-Agent Reasoning with Holistic Experience Learning Just Like an Olympiad Teamcited by 13
  10. 2025Propose, Solve, Verify: Self-Play Through Formal Verificationcited by 13
  11. 2026Learn from Your Mistakes: Self-Correcting Masked Diffusion Modelscited by 13
  12. 2025Self-Jailbreaking: Language Models Can Reason Themselves Out of Safety Alignment After Benign Reasoning Trainingcited by 12
38 more
  1. 2025AutoCode: LLMs as Problem Setters for Competitive Programmingcited by 11
  2. 2025Pushing on Multilingual Reasoning Models with Language-Mixed Chain-of-Thoughtcited by 10
  3. 2026Golden Goose: A Simple Trick to Synthesize Unlimited RLVR Tasks from Unverifiable Internet Textcited by 10
  4. 2025Klear-CodeTest: Scalable Test Case Generation for Code Reinforcement Learningcited by 9
  5. 2025Hermes 4 Technical Reportcited by 8
  6. 2025Critique-Coder: Enhancing Coder Models by Critique Reinforcement Learningcited by 7
  7. 2026LaDi-RL: Latent Diffusion Reasoning Prevents Entropy Collapse in Reinforcement Learningcited by 7
  8. 2025Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Datacited by 5
  9. 2026CodeScaler: Scaling Code LLM Training and Test-Time Inference via Reward Modelscited by 5
  10. 2026EvolveCoder: Evolving Test Cases via Adversarial Verification for Code Reinforcement Learningcited by 5
  11. 2025Closing the Data Loop: Using OpenDataArena to Engineer Superior Training Datasetscited by 4
  12. 2026GrandCode: Achieving Grandmaster Level in Competitive Programming via Agentic Reinforcement Learningcited by 4
  13. 2026InCoder-32B-Thinking: Industrial Code World Model for Thinkingcited by 4
  14. 2025K2-V2: A 360-Open, Reasoning-Enhanced LLMcited by 3
  15. 2026Incentivizing In-depth Reasoning over Long Contexts with Process Advantage Shapingcited by 3
  16. 2026FrontierSmith: Synthesizing Open-Ended Coding Problems at Scalecited by 3
  17. 2025The Valley of Code Reasoning: Scaling Knowledge Distillation of Large Language Modelscited by 2
  18. 2025A Model Can Help Itself: Reward-Free Self-Training for LLM Reasoningcited by 2
  19. 2026Breaking Training Bottlenecks: Effective and Stable Reinforcement Learning for Coding Modelscited by 2
  20. 2026PaT: Planning-after-Trial for Efficient Test-Time Code Generationcited by 2
  21. 2026OctoLong: Mid-Training On Cross-Repository Code Contexts Enhances Long-Context Modelingcited by 2
  22. 2025UniCode: Augmenting Evaluation for Code Reasoningcited by 1
  23. 2025Policy Optimization Prefers The Path of Least Resistancecited by 1
  24. 2025Motif-2-12.7B-Reasoning: A Practitioner's Guide to RL Training Recipescited by 1
  25. 2026X-Coder: Advancing Competitive Programming with Synthetic Tasks, Solutions, and Testscited by 1
  26. 2026D3-Gym: Constructing Real-World Verifiable Environments for Data-Driven Discoverycited by 1
  27. 2026Themis: Training Robust Multilingual Code Reward Models for Flexible Multi-Criteria Scoringcited by 1
  28. 2026From Interface to Inference: Eliciting Any-Order Inference from Any-Order Modelscited by 1
  29. 2025SCoGen: Scenario-Centric Graph-Based Synthesis of Real-World Code Problemscited by 0
  30. 2025Adaptive Tool Generation with Models as Tools and Reinforcement Learningcited by 0
  31. 2025Generating Verifiable Chain of Thoughts from Exection-Tracescited by 0
  32. 2026Primal Generation, Dual Judgment: Self-Training from Test-Time Scalingcited by 0
  33. 2026Semantic Flow Regularization: Teaching LLMs to Generate Diverse Yet Coherent Responsescited by 0
  34. 2026The Generalization Spectrum: A Chromatographic Approach to Evaluating Learning Algorithmscited by 0
  35. 2026Reinforcement Learning without Ground-Truth Solutions can Improve LLMscited by 0
  36. 2026Confidence-Gated Transductive Test Generation for Code Rerankingcited by 0
  37. 2026Where the Model Changes Its Mind: Hindsight-Divergence Localization for Efficient Reinforcement Learning with Verifiable Rewardscited by 0
  38. 2026MagicBeanStalk: Escaping the RLVR Zero Advantage Trap with Recursive Data Synthesiscited by 0

50 most cited of 51 All 51 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Beyond Output Correctness: Benchmarking and Evaluating Large Language Model Reasoning in Coding Tasks
  2. 2025Evaluating the Generalization Capabilities of Large Language Models on Code Reasoning
  3. 2025Increasing LLM Coding Capabilities through Diverse Synthetic Coding Tasks
  4. 2025OJBench: A Competition Level Code Benchmark For Large Language Models
  5. 2026CodeContests-O: Powering LLMs via Feedback-Driven Iterative Test Case Generation
  6. 2023Competition-Level Problems are Effective LLM Evaluators
  7. 2025ICPC-Eval: Probing the Frontiers of LLM Reasoning with Competitive Programming Contests
  8. 2025CodeContests+: High-Quality Test Case Generation for Competitive Programming