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 1,085 篇，第 5 / 22 页

  1. ICLR 2026Reasoning Language Model Inference Serving Unveiled: An Empirical Study
  2. ICLR 2026Reasoning Models Can be Accurately Pruned Via Chain-of-Thought Reconstruction
  3. ICLR 2026Reasoning Scaffolding: Distilling the Flow of Thought from LLMs
  4. ICLR 2026ReForm: Reflective Autoformalization with Prospective Bounded Sequence Optimization
  5. ICLR 2026Reinforced Latent Reasoning for LLM-based Recommendation
  6. ICLR 2026Reinforcement Mid-Training
  7. ICLR 2026Reliable Fine-Grained Evaluation of Natural Language Math Proofs
  8. ICLR 2026Rethinking LLM Reasoning: From Explicit Trajectories to Latent Representations
  9. ICLR 2026Rewriting Pre-Training Data Boosts LLM Performance in Math and Code
  10. ICLR 2026RFEval: Benchmarking Reasoning Faithfulness under Counterfactual Reasoning Intervention in Large Reasoning Models
  11. ICLR 2026RLAD: Training LLMs to Discover Abstractions for Solving Reasoning Problems
  12. ICLR 2026RLP: Reinforcement as a Pretraining Objective
  13. ICLR 2026RLVMR: Reinforcement Learning with Verifiable Meta-Reasoning Rewards for Robust Long-Horizon Agents
  14. ICLR 2026RM-R1: Reward Modeling as Reasoning
  15. ICLR 2026Sample Smart, Not Hard: Correctness-First Decoding for Better Reasoning in LLMs
  16. ICLR 2026Scaf-GRPO: Scaffolded Group Relative Policy Optimization for Enhancing LLM Reasoning
  17. ICLR 2026Scalable Chain of Thoughts via Elastic Reasoning
  18. ICLR 2026Scheduling Your LLM Reinforcement Learning with Reasoning Trees
  19. ICLR 2026Segment-Level Attribution for Selective Learning of Long Reasoning Traces
  20. ICLR 2026Selection, Reflection and Self-Refinement: Revisit Reasoning Tasks via a Causal Lens
  21. ICLR 2026Self-Jailbreaking: Language Models Can Reason Themselves Out of Safety Alignment After Benign Reasoning Training
  22. ICLR 2026Self-Refining Vision Language Model for Robotic Failure Detection and Reasoning
  23. ICLR 2026SIM-CoT: Supervised Implicit Chain-of-Thought
  24. ICLR 2026SLM-MUX: Orchestrating Small Language Models for Reasoning
  25. ICLR 2026Slow-Fast Policy Optimization: Reposition-Before-Update for LLM Reasoning
  26. ICLR 2026Smarter Not Harder: Generative Process Evaluation with Intrinsic-Signal Driving and Ability‑Adaptive Reward Shaping
  27. ICLR 2026Soft Tokens, Hard Truths
  28. ICLR 2026Solving the 2-norm k-hyperplane clustering problem via multi-norm formulations
  29. ICLR 2026Stabilizing Policy Gradients for Sample-Efficient Reinforcement Learning in LLM Reasoning
  30. ICLR 2026STITCH: Simultaneous Thinking and Talking with Chunked Reasoning for Spoken Language Models
  31. ICLR 2026Strategic Obfuscation of Deceptive Reasoning in Language Models
  32. ICLR 2026StreamingThinker: Large Language Models Can Think While Reading
  33. ICLR 2026SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs
  34. ICLR 2026T1: Tool-integrated Verification for Test-time Compute Scaling in Small Language Models
  35. ICLR 2026TaTToo: Tool-Grounded Thinking PRM for Test-Time Scaling in Tabular Reasoning
  36. ICLR 2026The First Impression Problem: Internal Bias Triggers Overthinking in Reasoning Models
  37. ICLR 2026The Open Proof Corpus: A Large-Scale Study of LLM-Generated Mathematical Proofs
  38. ICLR 2026THE PATH OF LEAST RESISTANCE: GUIDING LLM REASONING TRAJECTORIES WITH PREFIX CONSENSUS
  39. ICLR 2026Think in Parallel, Answer as One: Logit Averaging for Open-Ended Reasoning
  40. ICLR 2026Thinking on the Fly: Test-Time Reasoning Enhancement via Latent Thought Policy Optimization
  41. ICLR 2026ThinkMorph: Emergent Properties in Multimodal Interleaved Chain-of-Thought Reasoning
  42. ICLR 2026Token Hidden Reward: Steering Exploration-Exploitation in Group Relative Deep Reinforcement Learning
  43. ICLR 2026TokUR: Token-Level Uncertainty Estimation for Large Language Model Reasoning
  44. ICLR 2026Towards Quantization-Aware Training for Ultra-Low-Bit Reasoning LLMs
  45. ICLR 2026Towards Safe Reasoning in Large Reasoning Models via Corrective Intervention
  46. ICLR 2026Tracing the Traces: Latent Temporal Signals for Efficient and Accurate Reasoning
  47. ICLR 2026Training Large Language Models To Reason In Parallel With Global Forking Tokens
  48. ICLR 2026Training Large Reasoning Models Efficiently via Progressive Thought Encoding
  49. ICLR 2026TRIM: Hybrid Inference via Targeted Stepwise Routing in Multi-Step Reasoning Tasks
  50. ICLR 2026VERINA: Benchmarking Verifiable Code Generation

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