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
- 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。
各会议论文数
按年份和会议统计
| 年份 | NeurIPS | ICLR | ICML | CVPR | 合计 |
|---|---|---|---|---|---|
| 2023 | 19 | 17 | 11 | 10 | 57 |
| 2024 | 52 | 34 | 39 | 12 | 137 |
| 2025 | 190 | 61 | 45 | 18 | 314 |
| 2026 | 208 | 175 | 113 | 81 | 577 |
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 577 篇，第 4 / 12 页
- ICLR 2026Latent Veracity Inference for Identifying Errors in Stepwise Reasoning
- ICLR 2026Latent-Guided Reasoning: Empowering Small LLMs with Large-Model Thinking
- ICLR 2026LeanForPhysics: Comprehensive Reasoning Framework for University-level Physics in Lean4
- ICLR 2026Learn More with Less: Uncertainty Consistency Guided Query Selection for RLVR
- ICLR 2026Learning from Synthetic Data Improves Multi-hop Reasoning
- ICLR 2026Learning Global Hypothesis Space for Enhancing Synergistic Reasoning Chain
- ICLR 2026Learning to Reason in Structured In-context Environments with Reinforcement Learning
- ICLR 2026Learning to Reason over Continuous Tokens with Reinforcement Learning
- ICLR 2026Learning to Reason without External Rewards
- ICLR 2026Let's Explore Step by Step: Generating Provable Formal Statements with Deductive Exploration
- ICLR 2026LINGOLY-TOO: Disentangling Reasoning from Knowledge with Templatised Orthographic Obfuscation
- ICLR 2026LLMs are Single-threaded Reasoners: Demystifying the Working Mechanism of Soft Thinking
- ICLR 2026LoC-Decomp: LLM Autoformalization via Logical Concept Decomposition and Iterative Feedback Correction
- ICLR 2026LogicReward: Incentivizing LLM Reasoning via Step-Wise Logical Supervision
- ICLR 2026Long Chain-of-Thought Reasoning Across Languages
- ICLR 2026Math Blind: Failures in Diagram Understanding Undermine Reasoning in MLLMs
- ICLR 2026MATH-Beyond: A Benchmark for RL to Expand Beyond the Base Model
- ICLR 2026Mathesis: Towards Formal Theorem Proving from Natural Languages
- ICLR 2026MathFimer: Enhancing Mathematical Reasoning by Expanding Reasoning Steps through Fill-in-the-Middle Task
- ICLR 2026MATHMO: Automated Mathematical Modeling Through Adaptive Search
- ICLR 2026MathNet: A Global Multimodal Benchmark for Mathematical Reasoning and Retrieval
- ICLR 2026MSCR: Exploring the Vulnerability of LLMs’ Mathematical Reasoning Abilities Using Multi-Source Candidate Replacement
- ICLR 2026Native Reasoning Models: Training Language Models to Reason on Unverifiable Data
- ICLR 2026Nemotron-CC-Math: A 133 Billion-Token-Scale High Quality Math Pretraining Dataset
- ICLR 2026Neural Force Field: Few-shot Learning of Generalized Physical Reasoning
- ICLR 2026NFT: Bridging Supervised Learning and Reinforcement Learning in Math Reasoning
- ICLR 2026Not All Bits Are Equal: How Model Scale Changes Memory-Optimal Reasoning
- ICLR 2026Nudging the Boundaries of LLM Reasoning
- ICLR 2026Off-Trajectory Reasoning: Can LLMs Collaborate on Reasoning Trajectory?
- ICLR 2026On the Reasoning Abilities of Masked Diffusion Language Models
- ICLR 2026OpenThoughts: Data Recipes for Reasoning Models
- ICLR 2026Parallel-R1: Towards Parallel Thinking via Reinforcement Learning
- ICLR 2026Patching Gaps In LLM Reasoning With Interventional Training
- ICLR 2026PEAR: Phase Entropy Aware Reward for Efficient Reasoning
- ICLR 2026Plan and Budget: Effective and Efficient Test-Time Scaling on Reasoning Large Language Models
- ICLR 2026Premise Selection for a Lean Hammer
- ICLR 2026PRISM-Physics: Causal DAG-Based Process Evaluation for Physics Reasoning
- ICLR 2026ProofBridge: Auto-Formalization of Natural Language Proofs in Lean via Joint Embeddings
- ICLR 2026ProofOptimizer: Training Language Models to Simplify Proofs without Human Demonstrations
- ICLR 2026PROS: Towards Compute-Efficient RLVR via Rollout Prefix Reuse
- ICLR 2026Pruning Long Chain-of-Thought of Large Reasoning Models via Small-Scale Preference Optimization
- ICLR 2026Pushing on Multilingual Reasoning Models with Language-Mixed Chain-of-Thought
- ICLR 2026QuestA: Expanding Reasoning Capacity in LLMs via Question Augmentation
- ICLR 2026R-Horizon: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?
- ICLR 2026R-Zero: Self-Evolving Reasoning LLM from Zero Data
- ICLR 2026RADAR: Reasoning–Ability and Difficulty-Aware Routing in Language Models
- ICLR 2026RAIN-Merging: A Gradient-Free Method to Enhance Instruction Following in Large Reasoning Models with Preserved Thinking Format
- ICLR 2026Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards
- ICLR 2026Real-Time Reasoning Agents in Evolving Environments
- ICLR 2026Reasoning in Space via Grounding in the World
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"}
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。