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 15 / 71 页

  1. ICLR 2026Spatial Reasoning with Vision-Language Models in Ego-Centric Multi-View Scenes
  2. ICLR 2026Spatial-DISE: A Unified Benchmark for Evaluating Spatial Reasoning in Vision-Language Models
  3. ICLR 2026SpatiaLab: Can Vision–Language Models Perform Spatial Reasoning in the Wild?
  4. ICLR 2026SpatialLadder: Progressive Training for Spatial Reasoning in Vision-Language Models
  5. ICLR 2026SpatialViz-Bench: A Cognitively-Grounded Benchmark for Diagnosing Spatial Visualization in MLLMs
  6. ICLR 2026Specialization after Generalization: Towards Understanding Test-Time Training in Foundation Models
  7. ICLR 2026Speech World Model: Causal State–Action Planning with Explicit Reasoning for Speech
  8. ICLR 2026SpeechJudge: Towards Human-Level Judgment for Speech Naturalness
  9. ICLR 2026SPELL: Self-Play Reinforcement Learning for Evolving Long-Context Language Models
  10. ICLR 2026SPIKE-RL: Video-LLMs meet Bayesian Surprise
  11. ICLR 2026SpinBench: Perspective and Rotation as a Lens on Spatial Reasoning in VLMs
  12. ICLR 2026SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning
  13. ICLR 2026SportR: A Benchmark for Multimodal Large Language Model Reasoning in Sports
  14. ICLR 2026Spotlight on Token Perception for Multimodal Reinforcement Learning
  15. ICLR 2026Squeeze the Soaked Sponge: Efficient Off-policy RFT for Large Language Model
  16. ICLR 2026SRFT: A Single-Stage Method with Supervised and Reinforcement Fine-Tuning for Reasoning
  17. ICLR 2026Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping
  18. ICLR 2026STAR-Bench: Probing Deep Spatio-Temporal Reasoning as Audio 4D Intelligence
  19. ICLR 2026STITCH: Simultaneous Thinking and Talking with Chunked Reasoning for Spoken Language Models
  20. ICLR 2026Stop Unnecessary Reflection: Training LRMs for Efficient Reasoning with Adaptive Reflection and Length Coordinated Penalty
  21. ICLR 2026Strategic Obfuscation of Deceptive Reasoning in Language Models
  22. ICLR 2026Strategic Scaling of Test-Time Compute: A Bandit Learning Approach
  23. ICLR 2026StreamingThinker: Large Language Models Can Think While Reading
  24. ICLR 2026String Seed of Thought: Prompting LLMs for Distribution-Faithful and Diverse Generation
  25. ICLR 2026Structure Learning from Time-Series Data with Lag-Agnostic Structural Prior
  26. ICLR 2026Structured Reasoning for LLMs: A Unified Framework for Efficiency and Explainability
  27. ICLR 2026Supervised Reinforcement Learning: From Expert Trajectories to Step-wise Reasoning
  28. ICLR 2026SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs
  29. ICLR 2026Synergizing Understanding and Generation with Interleaved Analyzing-Drafting Thinking
  30. ICLR 2026Synthesizing Multimodal Verifiable Game Data to Boost VLMs' General Reasoning
  31. ICLR 2026T1: Tool-integrated Verification for Test-time Compute Scaling in Small Language Models
  32. ICLR 2026TableMaster: A Recipe to Advance Table Understanding with Language Models
  33. ICLR 2026TAMMs: Change Understanding and Forecasting in Satellite Image Time Series with Temporal-Aware Multimodal Models
  34. ICLR 2026TaTToo: Tool-Grounded Thinking PRM for Test-Time Scaling in Tabular Reasoning
  35. ICLR 2026Teach to Reason Safely: Policy-Guided Safety Tuning for MLRMs
  36. ICLR 2026Test-Time Adaptation without Source Data for Out-of-Domain Bioactivity Prediction
  37. ICLR 2026Test-time Domain Generalization for Image Super-resolution
  38. ICLR 2026Test-Time Efficient Pretrained Model Portfolios for Time Series Forecasting
  39. ICLR 2026Test-Time Matching: Unlocking Compositional Reasoning in Multimodal Models
  40. ICLR 2026TEST-TIME SCALING IN DIFFUSION LLMS VIA HIDDEN SEMI-AUTOREGRESSIVE EXPERTS
  41. ICLR 2026Test-Time Scaling with Reflective Generative Model
  42. ICLR 2026Test-time Verification via Optimal Transport: Coverage, ROC, & Sub-optimality
  43. ICLR 2026The Art of Scaling Reinforcement Learning Compute for LLMs
  44. ICLR 2026The Choice of Divergence: A Neglected Key to Mitigating Diversity Collapse in Reinforcement Learning with Verifiable Reward
  45. ICLR 2026The CoT Encyclopedia: Analyzing, Predicting, and Controlling how a Reasoning Model will Think
  46. ICLR 2026The First Impression Problem: Internal Bias Triggers Overthinking in Reasoning Models
  47. ICLR 2026The Geometry of Reasoning: Flowing Logics in Representation Space
  48. ICLR 2026The Illusion of Diminishing Returns: Measuring Long Horizon Execution in LLMs
  49. ICLR 2026The Imitation Game: Turing Machine Imitator is Length Generalizable Reasoner
  50. ICLR 2026The Limits of Inference Scaling Through Resampling

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