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 4 / 7 页

  1. NeurIPS 2025Benchmarking Spatiotemporal Reasoning in LLMs and Reasoning Models: Capabilities and Challenges
  2. NeurIPS 2025Beyond Chemical QA: Evaluating LLM's Chemical Reasoning with Modular Chemical Operations
  3. NeurIPS 2025Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning
  4. NeurIPS 2025Beyond Verifiable Rewards: Scaling Reinforcement Learning in Language Models to Unverifiable Data
  5. NeurIPS 2025Bootstrapping Hierarchical Autoregressive Formal Reasoner with Chain-of-Proxy-Autoformalization
  6. NeurIPS 2025BREAD: Branched Rollouts from Expert Anchors Bridge SFT & RL for Reasoning
  7. NeurIPS 2025C3PO: Optimized Large Language Model Cascades with Probabilistic Cost Constraints for Reasoning
  8. NeurIPS 2025Can MLLMs Absorb Math Reasoning Abilities from LLMs as Free Lunch?
  9. NeurIPS 2025Causal Sufficiency and Necessity Improves Chain-of-Thought Reasoning
  10. NeurIPS 2025Causal-R: A Causal-Reasoning Geometry Problem Solver for Optimized Solution Exploration
  11. NeurIPS 2025Chain of Execution Supervision Promotes General Reasoning in Large Language Models
  12. NeurIPS 2025CLAWS:Creativity detection for LLM-generated solutions using Attention Window of Sections
  13. NeurIPS 2025Co-PatcheR: Collaborative Software Patching with Component-specific Small Reasoning Models
  14. NeurIPS 2025Cognitive Mirrors: Exploring the Diverse Functional Roles of Attention Heads in LLM Reasoning
  15. NeurIPS 2025Composing Global Solutions to Reasoning Tasks via Algebraic Objects in Neural Nets
  16. NeurIPS 2025Compositional Neural Network Verification via Assume-Guarantee Reasoning
  17. NeurIPS 2025Compositional Reasoning with Transformers, RNNs, and Chain of Thought
  18. NeurIPS 2025Continuous Thought Machines
  19. NeurIPS 2025Contrastive Representations for Temporal Reasoning
  20. NeurIPS 2025Controlling Thinking Speed in Reasoning Models
  21. NeurIPS 2025Counterfactual reasoning: an analysis of in-context emergence
  22. NeurIPS 2025CPPO: Accelerating the Training of Group Relative Policy Optimization-Based Reasoning Models
  23. NeurIPS 2025Creativity or Brute Force? Using Brainteasers as a Window into the Problem-Solving Abilities of Large Language Models
  24. NeurIPS 2025Deliberation on Priors: Trustworthy Reasoning of Large Language Models on Knowledge Graphs
  25. NeurIPS 2025Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in LLM Reasoning
  26. NeurIPS 2025DisCO: Reinforcing Large Reasoning Models with Discriminative Constrained Optimization
  27. NeurIPS 2025Diversity-Aware Policy Optimization for Large Language Model Reasoning
  28. NeurIPS 2025DLoFT: Gradient-Decoupled Fine-Tuning for Generalizable Long Chain-of-Thought Reasoning
  29. NeurIPS 2025Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
  30. NeurIPS 2025Don’t Think Longer, Think Wisely: Optimizing Thinking Dynamics for Large Reasoning Models
  31. NeurIPS 2025DreamPRM: Domain-reweighted Process Reward Model for Multimodal Reasoning
  32. NeurIPS 2025DSRF: A Dynamic and Scalable Reasoning Framework for Solving RPMs
  33. NeurIPS 2025DuetGraph: Coarse-to-Fine Knowledge Graph Reasoning with Dual-Pathway Global-Local Fusion
  34. NeurIPS 2025DyFlow: Dynamic Workflow Framework for Agentic Reasoning
  35. NeurIPS 2025DynaAct: Large Language Model Reasoning with Dynamic Action Spaces
  36. NeurIPS 2025Efficiently Scaling LLM Reasoning Programs with Certaindex
  37. NeurIPS 2025Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzles
  38. NeurIPS 2025EvaLearn: Quantifying the Learning Capability and Efficiency of LLMs via Sequential Problem Solving
  39. NeurIPS 2025Evaluating the Inductive Abilities of Large Language Models: Why Chain-of-Thought Reasoning Sometimes Hurts More Than Helps
  40. NeurIPS 2025Evolutionary Reasoning Does Not Arise in Standard Usage of Protein Language Models
  41. NeurIPS 2025Execution Guided Line-by-Line Code Generation
  42. NeurIPS 2025ExPO: Unlocking Hard Reasoning with Self-Explanation-Guided Reinforcement Learning
  43. NeurIPS 2025Far from the Shallow: Brain-Predictive Reasoning Embedding through Residual Disentanglement
  44. NeurIPS 2025First SFT, Second RL, Third UPT: Continual Improving Multi-Modal LLM Reasoning via Unsupervised Post-Training
  45. NeurIPS 2025Flexible Realignment of Language Models
  46. NeurIPS 2025From Euler to AI: Unifying Formulas for Mathematical Constants
  47. NeurIPS 2025Generalizable Reasoning through Compositional Energy Minimization
  48. NeurIPS 2025GPO: Learning from Critical Steps to Improve LLM Reasoning
  49. NeurIPS 2025GRIP: A Graph-Based Reasoning Instruction Producer
  50. NeurIPS 2025HARDMath2: A Benchmark for Applied Mathematics Built by Students as Part of a Graduate Class

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