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 57 篇，第 1 / 2 页

1. CVPR 2023Abstract Visual Reasoning: An Algebraic Approach for Solving Raven's Progressive Matrices
2. CVPR 2023ANetQA: A Large-Scale Benchmark for Fine-Grained Compositional Reasoning Over Untrimmed Videos
3. CVPR 2023DeepMAD: Mathematical Architecture Design for Deep Convolutional Neural Network
4. CVPR 2023Divide and Conquer: Answering Questions With Object Factorization and Compositional Reasoning
5. CVPR 2023Revisiting the P3P Problem
6. CVPR 2023Solving 3D Inverse Problems Using Pre-Trained 2D Diffusion Models
7. CVPR 2023Solving Oscillation Problem in Post-Training Quantization Through a Theoretical Perspective
8. CVPR 2023Solving Relaxations of MAP-MRF Problems: Combinatorial In-Face Frank-Wolfe Directions
9. CVPR 2023Super-CLEVR: A Virtual Benchmark To Diagnose Domain Robustness in Visual Reasoning
10. CVPR 2023Visual Programming: Compositional Visual Reasoning Without Training
11. ICLR 2023A Stable and Scalable Method for Solving Initial Value PDEs with Neural Networks
12. ICLR 2023A Unified Algebraic Perspective on Lipschitz Neural Networks
13. ICLR 2023Complexity-Based Prompting for Multi-step Reasoning
14. ICLR 2023Differentiable Mathematical Programming for Object-Centric Representation Learning
15. ICLR 2023Draft, Sketch, and Prove: Guiding Formal Theorem Provers with Informal Proofs
16. ICLR 2023Dual Algorithmic Reasoning
17. ICLR 2023Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning
18. ICLR 2023Language Models Are Greedy Reasoners: A Systematic Formal Analysis of Chain-of-Thought
19. ICLR 2023Language models are multilingual chain-of-thought reasoners
20. ICLR 2023Learning Math Reasoning from Self-Sampled Correct and Partially-Correct Solutions
21. ICLR 2023Learning to Solve Constraint Satisfaction Problems with Recurrent Transformer
22. ICLR 2023Least-to-Most Prompting Enables Complex Reasoning in Large Language Models
23. ICLR 2023Protein Representation Learning via Knowledge Enhanced Primary Structure Reasoning
24. ICLR 2023Scenario-based Question Answering with Interacting Contextual Properties
25. ICLR 2023Self-Consistency Improves Chain of Thought Reasoning in Language Models
26. ICLR 2023Solving Constrained Variational Inequalities via a First-order Interior Point-based Method
27. ICLR 2023UniKGQA: Unified Retrieval and Reasoning for Solving Multi-hop Question Answering Over Knowledge Graph
28. ICML 2023Alternating Local Enumeration (TnALE): Solving Tensor Network Structure Search with Fewer Evaluations
29. ICML 2023Automated Search for Conjectures on Mathematical Constants using Analysis of Integer Sequences
30. ICML 2023Generalization on the Unseen, Logic Reasoning and Degree Curriculum
31. ICML 2023GibbsDDRM: A Partially Collapsed Gibbs Sampler for Solving Blind Inverse Problems with Denoising Diffusion Restoration
32. ICML 2023Infusing Lattice Symmetry Priors in Attention Mechanisms for Sample-Efficient Abstract Geometric Reasoning
33. ICML 2023Neural Inverse Operators for Solving PDE Inverse Problems
34. ICML 2023Neural Prediction Errors enable Analogical Visual Reasoning in Human Standard Intelligence Tests
35. ICML 2023Specializing Smaller Language Models towards Multi-Step Reasoning
36. ICML 2023The Catalog Problem: Clustering and Ordering Variable-Sized Sets
37. ICML 2023The Monge Gap: A Regularizer to Learn All Transport Maps
38. ICML 2023Towards Constituting Mathematical Structures for Learning to Optimize
39. NeurIPS 2023Disentangled Counterfactual Learning for Physical Audiovisual Commonsense Reasoning
40. NeurIPS 2023Evaluating and Improving Tool-Augmented Computation-Intensive Math Reasoning
41. NeurIPS 2023Fitting trees to $\ell_1$-hyperbolic distances
42. NeurIPS 2023Globally solving the Gromov-Wasserstein problem for point clouds in low dimensional Euclidean spaces
43. NeurIPS 2023Grounding Neural Inference with Satisfiability Modulo Theories
44. NeurIPS 2023How does GPT-2 compute greater-than?: Interpreting mathematical abilities in a pre-trained language model
45. NeurIPS 2023Interactive Visual Reasoning under Uncertainty
46. NeurIPS 2023Landscape Surrogate: Learning Decision Losses for Mathematical Optimization Under Partial Information
47. NeurIPS 2023Learning Reliable Logical Rules with SATNet
48. NeurIPS 2023Mathematical Capabilities of ChatGPT
49. NeurIPS 2023MathNAS: If Blocks Have a Role in Mathematical Architecture Design
50. NeurIPS 2023Neural Algorithmic Reasoning Without Intermediate Supervision

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。