数学推理论文趋势 2023–2026

定理与解题

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 1,085 篇被归入该主题的论文。

主要结论

每年论文数

各会议论文数

按年份和会议统计

数学推理论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
20231917111057
202452343912137
2025190614518314
202620817511381577

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

全部数学推理论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 314 篇，第 5 / 7 页

  1. NeurIPS 2025HELM: Hyperbolic Large Language Models via Mixture-of-Curvature Experts
  2. NeurIPS 2025Heterogeneous Adversarial Play in Interactive Environments
  3. NeurIPS 2025How Far Are We from Optimal Reasoning Efficiency?
  4. NeurIPS 2025Hybrid Latent Reasoning via Reinforcement Learning
  5. NeurIPS 2025Hybrid-Balance GFlowNet for Solving Vehicle Routing Problems
  6. NeurIPS 2025Improve Temporal Reasoning in Multimodal Large Language Models via Video Contrastive Decoding
  7. NeurIPS 2025Incentivizing Dual Process Thinking for Efficient Large Language Model Reasoning
  8. NeurIPS 2025Incentivizing LLMs to Self-Verify Their Answers
  9. NeurIPS 2025Incentivizing Reasoning for Advanced Instruction-Following of Large Language Models
  10. NeurIPS 2025Ineq-Comp: Benchmarking Human-Intuitive Compositional Reasoning in Automated Theorem Proving of Inequalities
  11. NeurIPS 2025IneqSearch: Hybrid Reasoning for Olympiad Inequality Proofs
  12. NeurIPS 2025Is PRM Necessary? Problem-Solving RL Implicitly Induces PRM Capability in LLMs
  13. NeurIPS 2025Knot So Simple: A Minimalistic Environment for Spatial Reasoning
  14. NeurIPS 2025KTAE: A Model-Free Algorithm to Key-Tokens Advantage Estimation in Mathematical Reasoning
  15. NeurIPS 2025Large Language Models Think Too Fast To Explore Effectively
  16. NeurIPS 2025Latent Refinement via Flow Matching for Training-free Linear Inverse Problem Solving
  17. NeurIPS 2025Learning Interestingness in Automated Mathematical Theory Formation
  18. NeurIPS 2025Learning to Solve Complex Problems via Dataset Decomposition
  19. NeurIPS 2025Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RL
  20. NeurIPS 2025Let LRMs Break Free from Overthinking via Self-Braking Tuning
  21. NeurIPS 2025LIMOPro: Reasoning Refinement for Efficient and Effective Test-time Scaling
  22. NeurIPS 2025Look-Ahead Reasoning on Learning Platforms
  23. NeurIPS 2025Lost in Transmission: When and Why LLMs Fail to Reason Globally
  24. NeurIPS 2025MathArena: Evaluating LLMs on Uncontaminated Math Competitions
  25. NeurIPS 2025Measuring the Faithfulness of Thinking Drafts in Large Reasoning Models
  26. NeurIPS 2025Mind the Gap: Bridging Thought Leap for Improved Chain-of-Thought Tuning
  27. NeurIPS 2025miniF2F-Lean Revisited: Reviewing Limitations and Charting a Path Forward
  28. NeurIPS 2025MINT-CoT: Enabling Interleaved Visual Tokens in Mathematical Chain-of-Thought Reasoning
  29. NeurIPS 2025Mitigating Overthinking in Large Reasoning Models via Manifold Steering
  30. NeurIPS 2025MM-Agent: LLM as Agents for Real-world Mathematical Modeling Problem
  31. NeurIPS 2025More Thinking, Less Seeing? Assessing Amplified Hallucination in Multimodal Reasoning Models
  32. NeurIPS 2025Multipole Attention for Efficient Long Context Reasoning
  33. NeurIPS 2025NaturalReasoning: Reasoning in the Wild with 2.8M Challenging Questions
  34. NeurIPS 2025No Experts, No Problem: Avoidance Learning from Bad Demonstrations
  35. NeurIPS 2025OMEGA: Can LLMs Reason Outside the Box in Math? Evaluating Exploratory, Compositional, and Transformative Generalization
  36. NeurIPS 2025Omni-R1: Reinforcement Learning for Omnimodal Reasoning via Two-System Collaboration
  37. NeurIPS 2025On Learning Verifiers and Implications to Chain-of-Thought Reasoning
  38. NeurIPS 2025On the Bias of Next-Token Predictors Toward Systematically Inefficient Reasoning: A Shortest-Path Case Study
  39. NeurIPS 2025Once Upon an Input: Reasoning via Per-Instance Program Synthesis
  40. NeurIPS 2025One Token Embedding Is Enough to Deadlock Your Large Reasoning Model
  41. NeurIPS 2025Optimizing Anytime Reasoning via Budget Relative Policy Optimization
  42. NeurIPS 2025Optimizing Chain-of-Thought Reasoners via Gradient Variance Minimization in Rejection Sampling and RL
  43. NeurIPS 2025PolyMath: Evaluating Mathematical Reasoning in Multilingual Contexts
  44. NeurIPS 2025Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning
  45. NeurIPS 2025ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models
  46. NeurIPS 2025QFFT, Question-Free Fine-Tuning for Adaptive Reasoning
  47. NeurIPS 2025QiMeng-CodeV-R1: Reasoning-Enhanced Verilog Generation
  48. NeurIPS 2025QuestBench: Can LLMs ask the right question to acquire information in reasoning tasks?
  49. NeurIPS 2025R-KV: Redundancy-aware KV Cache Compression for Reasoning Models
  50. NeurIPS 2025R$^2$ec: Towards Large Recommender Models with Reasoning

在 Claude、Cursor 里直接问这个趋势

接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。

可以直接复制这句话问 AI

用 paperlist 查数学推理在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？

AI 会调用的工具：get_topic_stats {"topic": "math-reasoning"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

常见问题

2023–2026 年四大会议一共有多少篇数学推理论文？
共 1,085 篇，按年份分别是 2023 年 57 篇、2024 年 137 篇、2025 年 314 篇、2026 年 577 篇。
NeurIPS 2026 有多少篇数学推理论文？
NeurIPS 2026 有 208 篇数学推理论文，上一届是 190 篇。
ICLR 2026 有多少篇数学推理论文？
ICLR 2026 有 175 篇数学推理论文，上一届是 61 篇。
ICML 2026 有多少篇数学推理论文？
ICML 2026 有 113 篇数学推理论文，上一届是 45 篇。
CVPR 2026 有多少篇数学推理论文？
CVPR 2026 有 81 篇数学推理论文，上一届是 18 篇。
哪个会议的数学推理论文最多？
NeurIPS，2023–2026 年共 469 篇。

「神经符号」下的其他主题