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

每年论文数

各会议论文数

按年份和会议统计

推理模型论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
20232715710933470
2024121846343311
202550020315970932
20263616525822041,799

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

全部推理模型论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 913 篇，第 10 / 19 页

  1. ICML 2026Spectral-Progressive Thought Flow for Lightweight Multimodal Reasoning
  2. ICML 2026SPLIT-VLM: Salience-Guided Partitioning towards Local Coverage for Importance-Aware Token Dropping in Vision-Language Models
  3. ICML 2026SRPO: Self-Reflective Policy Optimization for Long-Horizon Reasoning
  4. ICML 2026ST-Veto: Spatio-Temporal Token Veto for Diffusion MLLMs via Taylor Prediction and Visual Grounding
  5. ICML 2026Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers
  6. ICML 2026Stabilizing Recurrent Dynamics for Test-Time Scalable Latent Reasoning in Looped Language Models
  7. ICML 2026Stable Asynchrony: Variance-Controlled Off-Policy RL for LLMs
  8. ICML 2026Star Elastic: Many-in-One Reasoning LLMs with Efficient Budget Control
  9. ICML 2026State Space Model with Continuous Limit of HiPPO Matrix: Eigenvalue Analysis and Explicit Solution Formula
  10. ICML 2026Statistical Early Stopping for Reasoning Models
  11. ICML 2026Step-Level Sparse Autoencoder for Reasoning Process Interpretation
  12. ICML 2026Step-Resolved Data Attribution for Looped Transformers
  13. ICML 2026StepCodeReasoner: Aligning Code Reasoning with Stepwise Execution Traces via Reinforcement Learning
  14. ICML 2026Stop When Further Reasoning Won’t Help: Attention-State Adaptive Generation in Reasoning Models
  15. ICML 2026Strat-Reasoner: Reinforcing Strategic Reasoning of LLMs in Multi-Agent Games
  16. ICML 2026Strategy Executability in Mathematical Reasoning: Leveraging Human–Model Differences for Effective Guidance
  17. ICML 2026Structure Enables Effective Self-Localization of Errors in LLMs
  18. ICML 2026SuCo: Sufficiency-guided Continuous Adaptive Reasoning
  19. ICML 2026Symbol-Equivariant Recurrent Reasoning Models
  20. ICML 2026Syntax vs. Semantics: How Transformers Learn Deep Dependencies
  21. ICML 2026T-POP: Test-Time Personalization with Online Preference Feedback
  22. ICML 2026Tailoring the Training: Difficulty-Aware Learning Strategy Allocation for Large Language Models
  23. ICML 2026Task-and-Model-Aware Fractal-Consistency for Efficient LLM Reasoning
  24. ICML 2026Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability
  25. ICML 2026Test-Time Detoxification without Training or Learning Anything
  26. ICML 2026Test-time Generalization for Physics through Neural Operator Splitting
  27. ICML 2026Test-Time Learning of Causal Structure from Interventional Data
  28. ICML 2026Test-Time Training with KV Binding Is Secretly Linear Attention
  29. ICML 2026Testing For Distribution Shifts with Conditional Conformal Test Martingales
  30. ICML 2026TFRBench: A Reasoning Benchmark for Evaluating Forecasting Systems
  31. ICML 2026TG-RAG: A Retrieval-Augmented Framework for Reasoning Guidance in Specialized Domains
  32. ICML 2026The Abstraction Gap in Vision-Language Causal Reasoning
  33. ICML 2026The Art of Interrogation: Consistency Amplifies Factuality in Spatial Reasoning
  34. ICML 2026The Decrypto Benchmark for Multi-Agent Reasoning and Theory of Mind
  35. ICML 2026The Deterministic Horizon: When Extended Reasoning Fails and Tool Delegation Becomes Necessary
  36. ICML 2026The Easy, the Hard, and the Learnable: Confidence and Difficulty-Adaptive Policy Optimization for LLM Reasoning
  37. ICML 2026The Expressive Power of Low Precision Softmax Transformers with (Summarized) Chain-of-Thought
  38. ICML 2026The Flexibility Trap: Rethinking the Value of Arbitrary Order in Diffusion Language Models
  39. ICML 2026The Geometric Reasoner: Manifold-Informed Latent Foresight Search for Long-Context Reasoning
  40. ICML 2026The Geometry of Reasoning: Self-Evaluation via Layerwise Trajectory Evolution
  41. ICML 2026The Power of Power Law: Asymmetry Enables Compositional Reasoning
  42. ICML 2026The Quality-Utility Paradox: Why High-Reward Data Impairs Small Model Mathematical Reasoning
  43. ICML 2026The Shadow Price of Reasoning: Economic Perspective on Optimal Budget Allocation for LLMs
  44. ICML 2026The Shape of Addition: Geometric Structures of Arithmetic in Large Language Models
  45. ICML 2026The Signal is in the Steps: Local Scoring for Reasoning Data Selection
  46. ICML 2026The Silent Thought: Modeling Internal Cognition in Full-Duplex Spoken Dialogue Models via Latent Reasoning
  47. ICML 2026The Tell-Tale Norm: $\ell_2$ Magnitude as a Signal for Reasoning Dynamics in Large Language Models
  48. ICML 2026The Two-Hump Problem: Bridging the Difficulty Gap in Mathematical Reinforcement Learning
  49. ICML 2026The Value of Variance: Mitigating Debate Collapse in Multi-Agent Systems via Uncertainty-Driven Policy Optimization
  50. ICML 2026Theoretical Perspectives on Data Quality and Synergistic Effects in Pre- and Post-Training Reasoning Models

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"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

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。

「前沿系统」下的其他主题