推理模型论文趋势 2023–2026
链式推理、测试时计算与显式推理能力
统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 3,512 篇被归入该主题的论文。
主要结论
- 推理模型论文从 2023 年的 470 篇增长到 2026 年的 1,799 篇，三年间扩大到约 3.8 倍。
- NeurIPS 是发推理模型论文最多的会议，2023–2026 年共 1,253 篇，占四个会议总量的 36%。
- 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 9 篇增至 2026 年的 437 篇。
- 在「前沿系统」的 4 个主题里，它的 2026 年论文数排第 3；同组增长最快的AI 智能体是 12 倍，推理模型是 3.8 倍。
各会议论文数
按年份和会议统计
| 年份 | NeurIPS | ICLR | ICML | CVPR | 合计 |
|---|---|---|---|---|---|
| 2023 | 271 | 57 | 109 | 33 | 470 |
| 2024 | 121 | 84 | 63 | 43 | 311 |
| 2025 | 500 | 203 | 159 | 70 | 932 |
| 2026 | 361 | 652 | 582 | 204 | 1,799 |
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 Mechanistic Analysis of Looped Reasoning Language Models
- ICML 2026A Formal Comparison Between Chain of Thought and Latent Thought
- ICLR 2026A Simple "Motivation" Can Enhance Reinforcement Finetuning of Large Reasoning Models
- CVPR 2026Asking like Socrates Socrates helps VLMs understand remote sensing images
- NeurIPS 2025A Frustratingly Simple Yet Highly Effective Attack Baseline: Over 90% Success Rate Against the Strong Black-box Models of GPT-4.5/4o/o1
- ICML 2025A Reasoning-Based Approach to Cryptic Crossword Clue Solving
- ICLR 2025A Formal Framework for Understanding Length Generalization in Transformers
- CVPR 2025Argus: Vision-Centric Reasoning with Grounded Chain-of-Thought
- NeurIPS 2024Are Large Language Models Good Statisticians?
- ICML 2024Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
- ICLR 2024Amortizing intractable inference in large language models
- CVPR 2024Generate Subgoal Images before Act: Unlocking the Chain-of-Thought Reasoning in Diffusion Model for Robot Manipulation with Multimodal Prompts
- NeurIPS 2023A benchmark of categorical encoders for binary classification
- ICML 2023Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models?
- ICLR 2023Automatic Chain of Thought Prompting in Large Language Models
- CVPR 2023ActMAD: Activation Matching To Align Distributions for Test-Time-Training
- NeurIPS 2026A Solvable Model of Chain-of-Thought in In-Context Learning
- ICML 2026A Narrowing Geometry in Contaminated Reasoning
- ICLR 2026A State-Transition Framework for Efficient LLM Reasoning
- CVPR 2026Beyond Weak Supervision MLLMs-Guided Graded Knowledge Distillation for Unsupervised Camouflaged
全部推理模型论文
共 3,512 篇，第 44 / 71 页
- ICML 2025Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worse
- ICML 2025Mirror, Mirror of the Flow: How Does Regularization Shape Implicit Bias?
- ICML 2025MME-CoT: Benchmarking Chain-of-Thought in Large Multimodal Models for Reasoning Quality, Robustness, and Efficiency
- ICML 2025Neural Genetic Search in Discrete Spaces
- ICML 2025Noise Conditional Variational Score Distillation
- ICML 2025On the Power of Context-Enhanced Learning in LLMs
- ICML 2025One Example Shown, Many Concepts Known! Counterexample-Driven Conceptual Reasoning in Mathematical LLMs
- ICML 2025Optimizing Test-Time Compute via Meta Reinforcement Finetuning
- ICML 2025PDE-Controller: LLMs for Autoformalization and Reasoning of PDEs
- ICML 2025PENCIL: Long Thoughts with Short Memory
- ICML 2025PokéChamp: an Expert-level Minimax Language Agent
- ICML 2025Policy Guided Tree Search for Enhanced LLM Reasoning
- ICML 2025Position: A Theory of Deep Learning Must Include Compositional Sparsity
- ICML 2025Position: Formal Mathematical Reasoning—A New Frontier in AI
- ICML 2025Position: It Is Time We Test Neural Computation In Vitro
- ICML 2025Position: Language model developers should report train-test overlap
- ICML 2025Position: LLMs Need a Bayesian Meta-Reasoning Framework for More Robust and Generalizable Reasoning
- ICML 2025Position: Trustworthy AI Agents Require the Integration of Large Language Models and Formal Methods
- ICML 2025Position: We Need An Algorithmic Understanding of Generative AI
- ICML 2025Premise-Augmented Reasoning Chains Improve Error Identification in Math reasoning with LLMs
- ICML 2025Primal-Dual Neural Algorithmic Reasoning
- ICML 2025Product of Experts with LLMs: Boosting Performance on ARC Is a Matter of Perspective
- ICML 2025PTTA: Purifying Malicious Samples for Test-Time Model Adaptation
- ICML 2025Putnam-AXIOM: A Functional & Static Benchmark for Measuring Higher Level Mathematical Reasoning in LLMs
- ICML 2025R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts
- ICML 2025Ranked Entropy Minimization for Continual Test-Time Adaptation
- ICML 2025RE-IMAGINE: Symbolic Benchmark Synthesis for Reasoning Evaluation
- ICML 2025Reasoning Limitations of Multimodal Large Language Models. A case study of Bongard Problems
- ICML 2025Reasoning Through Execution: Unifying Process and Outcome Rewards for Code Generation
- ICML 2025Reasoning-as-Logic-Units: Scaling Test-Time Reasoning in Large Language Models Through Logic Unit Alignment
- ICML 2025ReFocus: Visual Editing as a Chain of Thought for Structured Image Understanding
- ICML 2025Regress, Don't Guess: A Regression-like Loss on Number Tokens for Language Models
- ICML 2025RepoAudit: An Autonomous LLM-Agent for Repository-Level Code Auditing
- ICML 2025Rethinking Chain-of-Thought from the Perspective of Self-Training
- ICML 2025Rethinking External Slow-Thinking: From Snowball Errors to Probability of Correct Reasoning
- ICML 2025ReVISE: Learning to Refine at Test-Time via Intrinsic Self-Verification
- ICML 2025Revisiting Chain-of-Thought in Code Generation: Do Language Models Need to Learn Reasoning before Coding?
- ICML 2025rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking
- ICML 2025RULEBREAKERS: Challenging LLMs at the Crossroads between Formal Logic and Human-like Reasoning
- ICML 2025Safety Reasoning with Guidelines
- ICML 2025SafetyAnalyst: Interpretable, Transparent, and Steerable Safety Moderation for AI Behavior
- ICML 2025Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verification
- ICML 2025Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search
- ICML 2025Scaling Test-Time Compute Without Verification or RL is Suboptimal
- ICML 2025Self-Consistency Preference Optimization
- ICML 2025ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning
- ICML 2025Soft Reasoning: Navigating Solution Spaces in Large Language Models through Controlled Embedding Exploration
- ICML 2025Spatial Reasoning with Denoising Models
- ICML 2025Stability and Generalization Capability of Subgraph Reasoning Models for Inductive Knowledge Graph Completion
- ICML 2025STAIR: Improving Safety Alignment with Introspective Reasoning
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 查推理模型在 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": "reasoning-models"}
常见问题
- 2023–2026 年四大会议一共有多少篇推理模型论文？
- 共 3,512 篇，按年份分别是 2023 年 470 篇、2024 年 311 篇、2025 年 932 篇、2026 年 1,799 篇。
- NeurIPS 2026 有多少篇推理模型论文？
- NeurIPS 2026 有 361 篇推理模型论文，上一届是 500 篇。
- ICLR 2026 有多少篇推理模型论文？
- ICLR 2026 有 652 篇推理模型论文，上一届是 203 篇。
- ICML 2026 有多少篇推理模型论文？
- ICML 2026 有 582 篇推理模型论文，上一届是 159 篇。
- CVPR 2026 有多少篇推理模型论文？
- CVPR 2026 有 204 篇推理模型论文，上一届是 70 篇。
- 哪个会议的推理模型论文最多？
- NeurIPS，2023–2026 年共 1,253 篇。