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 3,512 篇，第 51 / 71 页

  1. NeurIPS 2025Neurosymbolic Diffusion Models
  2. NeurIPS 2025Noise Hypernetworks: Amortizing Test-Time Compute in Diffusion Models
  3. NeurIPS 2025NoisyGRPO: Incentivizing Multimodal CoT Reasoning via Noise Injection and Bayesian Estimation
  4. NeurIPS 2025NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation
  5. NeurIPS 2025nvBench 2.0: Resolving Ambiguity in Text-to-Visualization through Stepwise Reasoning
  6. NeurIPS 2025OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning
  7. NeurIPS 2025OMEGA: Can LLMs Reason Outside the Box in Math? Evaluating Exploratory, Compositional, and Transformative Generalization
  8. NeurIPS 2025On Learning Verifiers and Implications to Chain-of-Thought Reasoning
  9. NeurIPS 2025On Reasoning Strength Planning in Large Reasoning Models
  10. NeurIPS 2025On the Bias of Next-Token Predictors Toward Systematically Inefficient Reasoning: A Shortest-Path Case Study
  11. NeurIPS 2025On the Effect of Negative Gradient in Group Relative Deep Reinforcement Optimization
  12. NeurIPS 2025Once Upon an Input: Reasoning via Per-Instance Program Synthesis
  13. NeurIPS 2025One Token Embedding Is Enough to Deadlock Your Large Reasoning Model
  14. NeurIPS 2025Open-Reasoner-Zero: An Open Source Approach to Scaling Up Reinforcement Learning on the Base Model
  15. NeurIPS 2025OpenCUA: Open Foundations for Computer-Use Agents
  16. NeurIPS 2025OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles
  17. NeurIPS 2025Optimizing Anytime Reasoning via Budget Relative Policy Optimization
  18. NeurIPS 2025Optimizing Chain-of-Thought Reasoners via Gradient Variance Minimization in Rejection Sampling and RL
  19. NeurIPS 2025OST-Bench: Evaluating the Capabilities of MLLMs in Online Spatio-temporal Scene Understanding
  20. NeurIPS 2025PAID: Pairwise Angular-Invariant Decomposition for Continual Test-Time Adaptation
  21. NeurIPS 2025Parameter Dynamics of Online Machine Learning and Test-time Adaptation
  22. NeurIPS 2025Partition-Then-Adapt: Combating Prediction Bias for Reliable Multi-Modal Test-Time Adaptation
  23. NeurIPS 2025Pause Tokens Strictly Increase the Expressivity of Constant-Depth Transformers
  24. NeurIPS 2025PeRL: Permutation-Enhanced Reinforcement Learning for Interleaved Vision-Language Reasoning
  25. NeurIPS 2025Personalized Decision Modeling: Utility Optimization or Textualized-Symbolic Reasoning
  26. NeurIPS 2025PHYBench: Holistic Evaluation of Physical Perception and Reasoning in Large Language Models
  27. NeurIPS 2025PhyBlock: A Progressive Benchmark for Physical Understanding and Planning via 3D Block Assembly
  28. NeurIPS 2025PhysVLM-AVR: Active Visual Reasoning for Multimodal Large Language Models in Physical Environments
  29. NeurIPS 2025Pixel Reasoner: Incentivizing Pixel Space Reasoning via Curiosity-Driven Reinforcement Learning
  30. NeurIPS 2025Planning without Search: Refining Frontier LLMs with Offline Goal-Conditioned RL
  31. NeurIPS 2025PlanU: Large Language Model Reasoning through Planning under Uncertainty
  32. NeurIPS 2025Point-RFT: Improving Multimodal Reasoning with Visually Grounded Reinforcement Finetuning
  33. NeurIPS 2025PolyMath: Evaluating Mathematical Reasoning in Multilingual Contexts
  34. NeurIPS 2025Praxis-VLM: Vision-Grounded Decision Making via Text-Driven Reinforcement Learning
  35. NeurIPS 2025Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning
  36. NeurIPS 2025Privacy Reasoning in Ambiguous Contexts
  37. NeurIPS 2025Probabilistic Reasoning with LLMs for Privacy Risk Estimation
  38. NeurIPS 2025Prompting as Scientific Inquiry
  39. NeurIPS 2025ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models
  40. NeurIPS 2025ProtoPairNet: Interpretable Regression through Prototypical Pair Reasoning
  41. NeurIPS 2025Provable Scaling Laws for the Test-Time Compute of Large Language Models
  42. NeurIPS 2025PSI: A Benchmark for Human Interpretation and Response in Traffic Interactions
  43. NeurIPS 2025PurpCode: Reasoning for Safer Code Generation
  44. NeurIPS 2025QFFT, Question-Free Fine-Tuning for Adaptive Reasoning
  45. NeurIPS 2025Quantifying Elicitation of Latent Capabilities in Language Models
  46. NeurIPS 2025QuestBench: Can LLMs ask the right question to acquire information in reasoning tasks?
  47. NeurIPS 2025R-KV: Redundancy-aware KV Cache Compression for Reasoning Models
  48. NeurIPS 2025R1-ShareVL: Incentivizing Reasoning Capabilities of Multimodal Large Language Models via Share-GRPO
  49. NeurIPS 2025R2R: Efficiently Navigating Divergent Reasoning Paths with Small-Large Model Token Routing
  50. NeurIPS 2025RADAR: Benchmarking Language Models on Imperfect Tabular Data

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