数学推理论文趋势 2023–2026

主要结论

• 数学推理论文从 2023 年的 57 篇增长到 2026 年的 577 篇，三年间扩大到约 10 倍。
• NeurIPS 是发数学推理论文最多的会议，2023–2026 年共 469 篇，占四个会议总量的 43%。
• 交叉方向里涨得最快的是强化学习：同时属于两个主题的论文从 2023 年的 3 篇增至 2026 年的 140 篇。
• 在「神经符号」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的可微推理是 2.5 倍，数学推理是 10 倍。

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

• NeurIPS 2026A mathematical theory of balancing relational generalization and memorization
• ICML 2026ASyMOB: Algebraic Symbolic Mathematical Operations Benchmark
• ICLR 2026AutoGPS: Automated Geometry Problem Solving via Multimodal Formalization and Deductive Reasoning
• CVPR 2026Affine Perspective-Three-Point Problem
• NeurIPS 2025Are Language Models Efficient Reasoners? A Perspective from Logic Programming
• ICML 2025Autoformulation of Mathematical Optimization Models Using LLMs
• ICLR 2025Advancing Mathematical Reasoning in Language Models: The Impact of Problem-Solving Data, Data Synthesis Methods, and Training Stages
• CVPR 2025Bridging Viewpoint Gaps: Geometric Reasoning Boosts Semantic Correspondence
• NeurIPS 2024A Careful Examination of Large Language Model Performance on Grade School Arithmetic
• ICML 2024Accelerating Legacy Numerical Solvers by Non-intrusive Gradient-based Meta-solving
• ICLR 2024Don't Trust: Verify -- Grounding LLM Quantitative Reasoning with Autoformalization
• CVPR 2024E-GPS: Explainable Geometry Problem Solving via Top-Down Solver and Bottom-Up Generator
• NeurIPS 2023Evaluating and Improving Tool-Augmented Computation-Intensive Math Reasoning
• ICML 2023Automated Search for Conjectures on Mathematical Constants using Analysis of Integer Sequences
• ICLR 2023Differentiable Mathematical Programming for Object-Centric Representation Learning
• CVPR 2023Abstract Visual Reasoning: An Algebraic Approach for Solving Raven's Progressive Matrices
• NeurIPS 2026A Mechanistic Analysis of Looped Reasoning Language Models
• ICML 2026Beyond Theorem Proving: Formulation, Framework and Benchmark for Formal Problem-Solving
• ICLR 2026CogFlow: Bridging Perception and Reasoning through Knowledge Internalization for Visual Mathematical Problem Solving
• CVPR 2026Hilbert-Geo Solving Solid Geometric Problems by Neural-Symbolic Reasoning

全部数学推理论文

共 61 篇，第 1 / 2 页

1. ICLR 2025Advancing Mathematical Reasoning in Language Models: The Impact of Problem-Solving Data, Data Synthesis Methods, and Training Stages
2. ICLR 2025Arithmetic Without Algorithms: Language Models Solve Math with a Bag of Heuristics
3. ICLR 2025B-STaR: Monitoring and Balancing Exploration and Exploitation in Self-Taught Reasoners
4. ICLR 2025Can LLMs Solve Longer Math Word Problems Better?
5. ICLR 2025CogCoM: A Visual Language Model with Chain-of-Manipulations Reasoning
6. ICLR 2025Conformal Language Model Reasoning with Coherent Factuality
7. ICLR 2025Distributional Associations vs In-Context Reasoning: A Study of Feed-forward and Attention Layers
8. ICLR 2025Don't Take Things Out of Context: Attention Intervention for Enhancing Chain-of-Thought Reasoning in Large Language Models
9. ICLR 2025DOTS: Learning to Reason Dynamically in LLMs via Optimal Reasoning Trajectories Search
10. ICLR 2025Dualformer: Controllable Fast and Slow Thinking by Learning with Randomized Reasoning Traces
11. ICLR 2025DynaMath: A Dynamic Visual Benchmark for Evaluating Mathematical Reasoning Robustness of Vision Language Models
12. ICLR 2025From Probability to Counterfactuals: the Increasing Complexity of Satisfiability in Pearl's Causal Hierarchy
13. ICLR 2025G-LLaVA: Solving Geometric Problem with Multi-Modal Large Language Model
14. ICLR 2025Generalizing Reasoning Problems to Longer Lengths
15. ICLR 2025GeoX: Geometric Problem Solving Through Unified Formalized Vision-Language Pre-training
16. ICLR 2025GReaTer: Gradients Over Reasoning Makes Smaller Language Models Strong Prompt Optimizers
17. ICLR 2025GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models
18. ICLR 2025HARDMath: A Benchmark Dataset for Challenging Problems in Applied Mathematics
19. ICLR 2025Hybrid Regularization Improves Diffusion-based Inverse Problem Solving
20. ICLR 2025Improving Complex Reasoning with Dynamic Prompt Corruption: A Soft Prompt Optimization Approach
21. ICLR 2025Improving Reasoning Performance in Large Language Models via Representation Engineering
22. ICLR 2025INFER: A Neural-symbolic Model For Extrapolation Reasoning on Temporal Knowledge Graph
23. ICLR 2025Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for LLM Problem-Solving
24. ICLR 2025Is Your Model Really A Good Math Reasoner? Evaluating Mathematical Reasoning with Checklist
25. ICLR 2025Layer Swapping for Zero-Shot Cross-Lingual Transfer in Large Language Models
26. ICLR 2025LeanAgent: Lifelong Learning for Formal Theorem Proving
27. ICLR 2025Learning to Plan Before Answering: Self-Teaching LLMs to Learn Abstract Plans for Problem Solving
28. ICLR 2025MathCoder2: Better Math Reasoning from Continued Pretraining on Model-translated Mathematical Code
29. ICLR 2025MathGAP: Out-of-Distribution Evaluation on Problems with Arbitrarily Complex Proofs
30. ICLR 2025MAVIS: Mathematical Visual Instruction Tuning with an Automatic Data Engine
31. ICLR 2025MIND: Math Informed syNthetic Dialogues for Pretraining LLMs
32. ICLR 2025MIRAGE: Evaluating and Explaining Inductive Reasoning Process in Language Models
33. ICLR 2025Mitigating Modality Prior-Induced Hallucinations in Multimodal Large Language Models via Deciphering Attention Causality
34. ICLR 2025MLPs Learn In-Context on Regression and Classification Tasks
35. ICLR 2025MR-GSM8K: A Meta-Reasoning Benchmark for Large Language Model Evaluation
36. ICLR 2025Multiagent Finetuning: Self Improvement with Diverse Reasoning Chains
37. ICLR 2025Mutual Reasoning Makes Smaller LLMs Stronger Problem-Solver
38. ICLR 2025Number Cookbook: Number Understanding of Language Models and How to Improve It
39. ICLR 2025Omni-MATH: A Universal Olympiad Level Mathematic Benchmark for Large Language Models
40. ICLR 2025OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data
41. ICLR 2025Physics of Language Models: Part 2.1, Grade-School Math and the Hidden Reasoning Process
42. ICLR 2025Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems
43. ICLR 2025Procedural Knowledge in Pretraining Drives Reasoning in Large Language Models
44. ICLR 2025Process Reward Model with Q-value Rankings
45. ICLR 2025Progress or Regress? Self-Improvement Reversal in Post-training
46. ICLR 2025Proving Olympiad Inequalities by Synergizing LLMs and Symbolic Reasoning
47. ICLR 2025Reasoning Elicitation in Language Models via Counterfactual Feedback
48. ICLR 2025Reasoning with Latent Thoughts: On the Power of Looped Transformers
49. ICLR 2025ReGenesis: LLMs can Grow into Reasoning Generalists via Self-Improvement
50. ICLR 2025SBSC: Step-by-Step Coding for Improving Mathematical Olympiad Performance

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。