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 137 篇，第 3 / 3 页

101. NeurIPS 2024Deep Equilibrium Algorithmic Reasoning
102. NeurIPS 2024Diffusion of Thought: Chain-of-Thought Reasoning in Diffusion Language Models
103. NeurIPS 2024DMPlug: A Plug-in Method for Solving Inverse Problems with Diffusion Models
104. NeurIPS 2024Does Reasoning Emerge? Examining the Probabilities of Causation in Large Language Models
105. NeurIPS 2024Doing Experiments and Revising Rules with Natural Language and Probabilistic Reasoning
106. NeurIPS 2024Embedding Trajectory for Out-of-Distribution Detection in Mathematical Reasoning
107. NeurIPS 2024Evaluating Large Vision-and-Language Models on Children's Mathematical Olympiads
108. NeurIPS 2024Global Lyapunov functions: a long-standing open problem in mathematics, with symbolic transformers
109. NeurIPS 2024How Far Can Transformers Reason? The Globality Barrier and Inductive Scratchpad
110. NeurIPS 2024Initialization is Critical to Whether Transformers Fit Composite Functions by Reasoning or Memorizing
111. NeurIPS 2024Interpretable Concept-Based Memory Reasoning
112. NeurIPS 2024JiuZhang3.0: Efficiently Improving Mathematical Reasoning by Training Small Data Synthesis Models
113. NeurIPS 2024Latent Neural Operator for Solving Forward and Inverse PDE Problems
114. NeurIPS 2024Lean Workbook: A large-scale Lean problem set formalized from natural language math problems
115. NeurIPS 2024MACM: Utilizing a Multi-Agent System for Condition Mining in Solving Complex Mathematical Problems
116. NeurIPS 2024MARVEL: Multidimensional Abstraction and Reasoning through Visual Evaluation and Learning
117. NeurIPS 2024MathPile: A Billion-Token-Scale Pretraining Corpus for Math
118. NeurIPS 2024Meaningful Learning: Enhancing Abstract Reasoning in Large Language Models via Generic Fact Guidance
119. NeurIPS 2024Measuring Multimodal Mathematical Reasoning with MATH-Vision Dataset
120. NeurIPS 2024Metacognitive Capabilities of LLMs: An Exploration in Mathematical Problem Solving
121. NeurIPS 2024MindMerger: Efficiently Boosting LLM Reasoning in non-English Languages
122. NeurIPS 2024MLLM-CompBench: A Comparative Reasoning Benchmark for Multimodal LLMs
123. NeurIPS 2024Neuro-Symbolic Data Generation for Math Reasoning
124. NeurIPS 2024On the Inductive Bias of Stacking Towards Improving Reasoning
125. NeurIPS 2024Open-Book Neural Algorithmic Reasoning
126. NeurIPS 2024OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset
127. NeurIPS 2024Proving Olympiad Algebraic Inequalities without Human Demonstrations
128. NeurIPS 2024PutnamBench: Evaluating Neural Theorem-Provers on the Putnam Mathematical Competition
129. NeurIPS 2024PUZZLES: A Benchmark for Neural Algorithmic Reasoning
130. NeurIPS 2024RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold
131. NeurIPS 2024SciInstruct: a Self-Reflective Instruction Annotated Dataset for Training Scientific Language Models
132. NeurIPS 2024SELF-DISCOVER: Large Language Models Self-Compose Reasoning Structures
133. NeurIPS 2024Solving Inverse Problems via Diffusion Optimal Control
134. NeurIPS 2024ST$_k$: A Scalable Module for Solving Top-k Problems
135. NeurIPS 2024StrategyLLM: Large Language Models as Strategy Generators, Executors, Optimizers, and Evaluators for Problem Solving
136. NeurIPS 2024Understanding the Limits of Vision Language Models Through the Lens of the Binding Problem
137. NeurIPS 2024Unlocking the Capabilities of Thought: A Reasoning Boundary Framework to Quantify and Optimize Chain-of-Thought

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。