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 577 篇，第 8 / 12 页

  1. ICML 2026The Quality-Utility Paradox: Why High-Reward Data Impairs Small Model Mathematical Reasoning
  2. ICML 2026The Signal is in the Steps: Local Scoring for Reasoning Data Selection
  3. ICML 2026The Two-Hump Problem: Bridging the Difficulty Gap in Mathematical Reinforcement Learning
  4. ICML 2026Think Deep, Not Just Long: Measuring LLM Reasoning Effort via Deep-Thinking Tokens
  5. ICML 2026Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language Models
  6. ICML 2026Thinking in Flow: A Dissipative Stabilization Operator for Robust Autoregressive Reasoning
  7. ICML 2026Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression
  8. ICML 2026Towards Solving the Gilbert-Pollak Conjecture via Large Language Models
  9. ICML 2026Truthfulness Does Not Scale Like Reasoning: Why Polling Fails as a Proxy Verifier
  10. ICML 2026Turbo Connection: Reasoning as Information Flow from Higher to Lower Layers
  11. ICML 2026Turning Drift into Constraint: Robust Reasoning Alignment in Non-Stationary Multi-Stream Environments
  12. ICML 2026Uncovering the Gradient Geometry of Long CoT: A Spectral-guided Approach to Reasoning Distillation
  13. ICML 2026Unveiling the Entropy Dynamics of Chain-of-Thought Reasoning
  14. ICML 2026Verifying Meta-Awareness via Predictive Rewards in Reasoning Models
  15. ICML 2026Wait, Wait, Wait... Why Do Reasoning Models Loop?
  16. ICML 2026What Characterizes Effective Reasoning? Revisiting Length, Review, and Structure of CoT
  17. ICML 2026What If We Allocate Test-Time Compute Adaptively?
  18. ICML 2026What Really Improves Mathematical Reasoning: Structured Reasoning Signals Beyond Pure Code
  19. ICML 2026When to Trust the Cheap Check: Weak and Strong Verification for Reasoning
  20. NeurIPS 20262D Spatial Reasoning with Adaptive Neural Cellular Automata
  21. NeurIPS 2026A mathematical theory of balancing relational generalization and memorization
  22. NeurIPS 2026A Matter of Interest: Understanding Interestingness Judgments of Math Problems in Humans and Language Models
  23. NeurIPS 2026A Mechanistic Analysis of Looped Reasoning Language Models
  24. NeurIPS 2026A Model of Diverse Sampling from Language Models
  25. NeurIPS 2026A Theory of Online Learning with Autoregressive Chain-of-Thought Reasoning
  26. NeurIPS 2026A$^3$Bench: A Benchmark for Compositional Reasoning over Aggressive Interactions in Videos
  27. NeurIPS 2026Active Probabilistic Reasoning in Humans and LLMs
  28. NeurIPS 2026AdaKVQ: Adaptive Mixed-Precision KV Cache Quantization For Efficient Reasoning Models
  29. NeurIPS 2026Agentic Geometry Problem Solving via Human-like Parallel Bidirectional Reasoning
  30. NeurIPS 2026Agentic Multi-Turn Reasoning: A Fairness Approach
  31. NeurIPS 2026Allocentric Perceiver: Disentangling Allocentric Reasoning from Egocentric Visual Priors via Frame Instantiation
  32. NeurIPS 2026An Enigma of Artificial Reason: Investigating the Production-Evaluation Gap in Large Reasoning Models
  33. NeurIPS 2026Anchoring Reasoning Distillation via Syntactic Constraints
  34. NeurIPS 2026Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information
  35. NeurIPS 2026Argument Graph Uncertainty: Quantifying Uncertainty from the Logical Structure of Reasoning Chains
  36. NeurIPS 2026ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both
  37. NeurIPS 2026Auxiliary Clues Aware’s Geometry Problem Solving
  38. NeurIPS 2026AVSD: Adaptive-View Self-Distillation by Balancing Consensus and Teacher-Specific Privileged Signals
  39. NeurIPS 2026Bellman Contraction under MMD: A Unified Framework
  40. NeurIPS 2026Benchmarking Multimodal Mathematical Reasoning with Explicit Visual Dependency
  41. NeurIPS 2026Beyond Global Alignment: Structured Compositional Reasoning for Vision-Language Models
  42. NeurIPS 2026Beyond Real or Fake: A Dual-Channel Authenticity and Reasoning Protocol for Photographic Assessment
  43. NeurIPS 2026Beyond What Seems Necessary: Hidden Gains from Scaling Training-Time Reasoning Length under Outcome Supervision
  44. NeurIPS 2026Bias-Variance Optimized Preference Optimization for Large Reasoning Models
  45. NeurIPS 2026BoardGameArena: A Multi-Dimensional Benchmark for Strategic Reasoning of LLMs in Board Games
  46. NeurIPS 2026Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
  47. NeurIPS 2026Boosting LLM Reasoning via Human-Inspired Reward Shaping
  48. NeurIPS 2026Boule or Baguette? A Study on Task Topology, Length Generalization, and the Benefit of Reasoning Trace
  49. NeurIPS 2026Can Agents Price a Reaction? Evaluating LLMs on Chemical Cost Reasoning
  50. NeurIPS 2026Can Linguistic Reasoning Vectors Enhance Multimodal Reasoning Ability?

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