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 4 / 19 页

  1. ICML 2026Effective Reasoning Chains Reduce Intrinsic Dimensionality
  2. ICML 2026Efficient Multi-Agent Reasoning via Confidence-Guided Adaptive Debate
  3. ICML 2026Efficient Parallel Samplers for Recurrent-Depth Models and Their Connection to Diffusion Language Models
  4. ICML 2026Efficient Reasoning with Hidden Thinking
  5. ICML 2026Efficient Test-time Inference for Generative Planning Models with OCL Search
  6. ICML 2026Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models
  7. ICML 2026Ego3S: Select, Strengthen, and Synchronize for Efficient Egocentric Reasoning
  8. ICML 2026Elastic Attention: Test-time Adaptive Sparsity Ratios for Efficient Transformers
  9. ICML 2026ElicitR: Unlocking Latent Reasoning in Dense Retrievers via Generative Regularization
  10. ICML 2026Emergent Analogical Reasoning in Transformers
  11. ICML 2026Entropy-Aware On-Policy Distillation of Language Models
  12. ICML 2026Equilibrium Reasoners: Learning Attractors Enables Scalable Reasoning
  13. ICML 2026ERGeoBench: A Comprehensive Benchmark for Embodied Reasoning and Geo-localization in Multimodal Large Language Models
  14. ICML 2026Escaping the Mode: Multi-Answer Reinforcement Learning in LMs
  15. ICML 2026Escaping the Verifier: Learning to Reason via Demonstrations
  16. ICML 2026Evaluating AI Grading on Real-World Handwritten College Mathematics: A Large-Scale Study Toward a Benchmark
  17. ICML 2026Evaluating Contextual Illegality: AI Compliance in Corporate Law Scenarios
  18. ICML 2026Evaluating LLM Uncertainty in Long-Form Generation Using Deterministic Ground Truth
  19. ICML 2026Evaluating Relational Reasoning in LLMs with REL
  20. ICML 2026Evaluating Robustness of Reasoning Models on Parameterized Logical Problems
  21. ICML 2026Evolution Strategies at the Hyperscale
  22. ICML 2026Evolving Quantitative Reasoning through Self-Play in Digital Twin Markets
  23. ICML 2026Expanding the Capabilities of Reinforcement Learning via Text Feedback
  24. ICML 2026Extracting alignment data in open models
  25. ICML 2026ExVerus: Verus Proof Repair via Counterexample Reasoning
  26. ICML 2026Faithful Relational Reasoning with Region-based Embeddings: Expressivity of Convex Coordinate-wise Models
  27. ICML 2026Fantastic Reasoning Behaviors and Where to Find Them: Unsupervised Discovery of the Reasoning Process
  28. ICML 2026Feature Resemblance: Towards a Theoretical Understanding of Analogical Reasoning in Transformers
  29. ICML 2026FedPAT: Federated Test-Time Adaptation via Prototype Affinity Topology
  30. ICML 2026Finding the Minimal Parameter Budget for Implicit Reasoning: A Data Complexity Driven Scaling Law for Language Models
  31. ICML 2026FIRE: Learning to Navigate and Act on Real-World Files via Stateful Reinforcement Learning
  32. ICML 2026FloorplanQA: A Benchmark for Spatial Reasoning in LLMs using Structured Representations
  33. ICML 2026ForesightKV: Optimizing KV Cache Eviction for Reasoning Models by Learning Long-Term Contribution
  34. ICML 2026Forward-Chaining Temporal Point Process
  35. ICML 2026FRISM: Fine-Grained Reasoning Injection via Subspace-Level Model Merging for Vision–Language Models
  36. ICML 2026From Growing to Looping: A Unified View of Iterative Computation in LLMs
  37. ICML 2026From Historical Patches to Repair Plans: Outcome-Conditioned Reasoning for Repository-Level Program Repair
  38. ICML 2026From Perception to Planning: Evolving Ego-Centric Task-Oriented Spatiotemporal Reasoning via Curriculum Learning
  39. ICML 2026From Prompts to Tokens: Internalizing Causal Supervision in Vision-Language Model for Multi-Image Causal Reasoning
  40. ICML 2026From Reasoning Traces to Reusable Modules: Understanding Compositional Generalization in Language Model Reasoning
  41. ICML 2026From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models
  42. ICML 2026Front-Loaded Robust Conformal Prediction: Heavy Calibration, Minimal Test-Time Cost
  43. ICML 2026Frontier Models Can Take Actions at Low Probabilities
  44. ICML 2026FrontierCS: Evolving Challenges for Evolving Intelligence
  45. ICML 2026Future-Gain Guided Test-Time Learning for Large Language Models
  46. ICML 2026G$^2$RPO: Geometric GRPO; Escaping LLM's Reasoning Rut to Break Accuracy--Entropy Trade-off
  47. ICML 2026Gateways to Tractability for Satisfiability in Pearl’s Causal Hierarchy
  48. ICML 2026GaussTrace: Provenance Analysis of 3D Gaussian Splatting Models with Evidence-based LLM Reasoning
  49. ICML 2026GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
  50. ICML 2026General Synthetic-Powered Inference

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