mlsanityData to 7 Oct 2026

NaturalReasoning: Reasoning in the Wild with 2.8M Challenging Questions

arXiv 2502.13124 ↗First posted 18 Feb 2025cs.CLcited by 81

On the map: LLM reasoning → Mathematical reasoning

Abstract (arXiv)

Scaling reasoning capabilities beyond traditional domains such as math and coding is hindered by the lack of diverse and high-quality questions. To overcome this limitation, we introduce a scalable approach for generating diverse and challenging reasoning questions, accompanied by reference answers. We present NaturalReasoning, a comprehensive dataset comprising 2.8 million questions that span multiple domains, including STEM fields (e.g., Physics, Computer Science), Economics, Social Sciences, and more. We demonstrate the utility of the questions in NaturalReasoning through knowledge distillation experiments which show that NaturalReasoning can effectively elicit and transfer reasoning capabilities from a strong teacher model. Furthermore, we demonstrate that NaturalReasoning is also effective for unsupervised self-training using external reward models or self-rewarding. To foster future work, we publicly release NaturalReasoning at https://huggingface.co/datasets/facebook/natural_reasoning.

Key facts. Cited by 81 papers in the map (data through 7 Oct 2026). Placed in Mathematical reasoning, one of 17 directions in LLM reasoning. Builds on 34 of its 57 references that are in the map. First posted on arXiv 18 Feb 2025.

Builds on

34 of 57 references.

Closest ancestors · 8

Same direction, newest first

  1. 2025LIMO: Less is More for Reasoning5 Feb
  2. 2024Self-Consistency Preference Optimization6 Nov
  3. 2024OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data2 Oct
  4. 2024Metacognitive Capabilities of LLMs: An Exploration in Mathematical Problem Solving20 May
  5. 2024Iterative Reasoning Preference Optimization30 Apr
  6. 2024Common 7B Language Models Already Possess Strong Math Capabilities7 Mar
  7. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models5 Feb
  8. 2024Augmenting Math Word Problems via Iterative Question Composing17 Jan

Other references · 26

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  3. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  4. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  5. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  6. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  7. 2023Let's Verify Step by Stepcited by 3,598
  8. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,841
  9. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  10. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,674
  11. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
  12. 2025s1: Simple test-time scalingcited by 1,311
14 more
  1. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,057
  2. 2022STaR: Bootstrapping Reasoning With Reasoningcited by 895
  3. 2022Solving math word problems with process- and outcome-based feedbackcited by 871
  4. 2023MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Modelscited by 724
  5. 2023RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignmentcited by 672
  6. 2024Self-Rewarding Language Modelscited by 654
  7. 2024Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Modelscited by 552
  8. 2024DataComp-LM: In search of the next generation of training sets for language modelscited by 381
  9. 2023Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Modelscited by 325
  10. 2024Direct Nash Optimization: Teaching Language Models to Self-Improve with General Preferencescited by 176
  11. 2024The Impact of Reasoning Step Length on Large Language Modelscited by 152
  12. 2023Self-Alignment with Instruction Backtranslationcited by 149
  13. 2024MAmmoTH2: Scaling Instructions from the Webcited by 143
  14. 2024MAPO: Advancing Multilingual Reasoning through Multilingual Alignment-as-Preference Optimizationcited by 68

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 45; 2026: 36.

81 papers.

Sort:

By area:

  1. 2025WebThinker: Empowering Large Reasoning Models with Deep Research Capabilitycited by 320
  2. 2025OpenThoughts: Data Recipes for Reasoning Modelscited by 287
  3. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 175
  4. 2025Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domainscited by 150
  5. 2025From LLM Reasoning to Autonomous AI Agents: A Comprehensive Reviewcited by 146
  6. 2025Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivizationcited by 124
  7. 2025Large Language Models Often Know When They Are Being Evaluatedcited by 105
  8. 2025Big-Math: A Large-Scale, High-Quality Math Dataset for Reinforcement Learning in Language Modelscited by 104
  9. 2025A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibilitycited by 89
  10. 2025SPICE: Self-Play In Corpus Environments Improves Reasoningcited by 78
  11. 2025Reinforcing General Reasoning without Verifierscited by 76
  12. 2025Revisiting Reinforcement Learning for LLM Reasoning from A Cross-Domain Perspectivecited by 75
38 more
  1. 2025One Token to Fool LLM-as-a-Judgecited by 56
  2. 2025MegaScience: Pushing the Frontiers of Post-Training Datasets for Science Reasoningcited by 56
  3. 2025SeRL: Self-Play Reinforcement Learning for Large Language Models with Limited Datacited by 50
  4. 20251.4 Million Open-Source Distilled Reasoning Dataset to Empower Large Language Model Trainingcited by 49
  5. 2025Scaling Laws of Synthetic Data for Language Modelscited by 44
  6. 2025AgentSynth: Scalable Task Generation for Generalist Computer-Use Agentscited by 41
  7. 2025Planning with Reasoning using Vision Language World Modelcited by 36
  8. 2025CoT-Self-Instruct: Building high-quality synthetic prompts for reasoning and non-reasoning taskscited by 30
  9. 2025Nemotron-CrossThink: Scaling Self-Learning beyond Math Reasoningcited by 29
  10. 2025NaturalThoughts: Selecting and Distilling Reasoning Traces for General Reasoning Taskscited by 29
  11. 2025Base Models Know How to Reason, Thinking Models Learn Whencited by 25
  12. 2025Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Insteadcited by 24
  13. 2025VerifyBench: Benchmarking Reference-based Reward Systems for Large Language Modelscited by 19
  14. 2025NeuroStrike: Neuron-Level Attacks on Aligned LLMscited by 18
  15. 2025DeepDistill: Enhancing LLM Reasoning Capabilities via Large-Scale Difficulty-Graded Data Trainingcited by 17
  16. 2025Intelligence per Watt: Measuring Intelligence Efficiency of Local AIcited by 13
  17. 2026X-OPD: Cross-Modal On-Policy Distillation for Capability Alignment in Speech LLMscited by 13
  18. 2025AgentFrontier: Expanding the Capability Frontier of LLM Agents with ZPD-Guided Data Synthesiscited by 12
  19. 2026Autodata: An agentic data scientist to create high quality synthetic datacited by 12
  20. 2025Influence Functions for Efficient Data Selection in Reasoningcited by 10
  21. 2026F2LLM-v2: Inclusive, Performant, and Efficient Embeddings for a Multilingual Worldcited by 10
  22. 2026Rethinking Reward Supervision: Rubric-Conditioned Self-Distillationcited by 9
  23. 2025Webscale-RL: Automated Data Pipeline for Scaling RL Data to Pretraining Levelscited by 8
  24. 2026Lost in the Noise: How Reasoning Models Fail with Contextual Distractorscited by 8
  25. 2026SafeNeuron: Neuron-Level Safety Alignment for Large Language Modelscited by 8
  26. 2025Distilled Pretraining: A modern lens of Data, In-Context Learning and Test-Time Scalingcited by 7
  27. 2026HRM-Text: Efficient Pretraining Beyond Scalingcited by 7
  28. 2025Essential-Web v1.0: 24T tokens of organized web datacited by 6
  29. 2025DESIGNER: Design-Logic-Guided Multidisciplinary Data Synthesis for LLM Reasoningcited by 6
  30. 2025Context-Free Synthetic Data Mitigates Forgettingcited by 5
  31. 2025A Short Survey on Small Reasoning Models: Training, Inference, Applications and Research Directionscited by 4
  32. 2025Aligning Multilingual Reasoning with Verifiable Semantics from a High-Resource Expert Modelcited by 4
  33. 2025More Data or Better Data? A Critical Analysis of Data Selection and Synthesis for Mathematical Reasoningcited by 4
  34. 2025QueST: Incentivizing LLMs to Generate Difficult Problemscited by 4
  35. 2025CoSineVerifier: Tool-Augmented Answer Verification for Computation-Oriented Scientific Questionscited by 4
  36. 2026WildSci: Advancing Scientific Reasoning from In-the-Wild Literaturecited by 4
  37. 2026Improving Data and Reward Design for Scientific Reasoning in Large Language Modelscited by 4
  38. 2026When Models Judge Themselves: Unsupervised Self-Evolution for Multimodal Reasoningcited by 4

50 most cited of 81 All 81 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Climbing the Ladder of Reasoning: What LLMs Can-and Still Can't-Solve after SFT?
  2. 2026Ladders of Thought: A Self-Evolving Curriculum of Progressively Simplified Reasoning Traces
  3. 2026Native Reasoning Models: Training Language Models to Reason on Unverifiable Data
  4. 2024Knowledge Distillation Using Frontier Open-source LLMs: Generalizability and the Role of Synthetic Data
  5. 2022Faithful Reasoning Using Large Language Models
  6. 2025LongReason: A Synthetic Long-Context Reasoning Benchmark via Context Expansion
  7. 2025Follow the Path: Reasoning over Knowledge Graph Paths to Improve Large Language Model Factuality
  8. 2026Questioning the Questions: Sustaining Self-Evolution in Reasoning Models