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 1,799 篇，第 33 / 36 页

  1. NeurIPS 2026Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs
  2. NeurIPS 2026Less Language, More Latents: Annotation-Efficient VLAs for Driving
  3. NeurIPS 2026LIME: Link-based User-item Interaction Modeling with Decoupled XOR Attention for Efficient Test Time Scaling
  4. NeurIPS 2026Linear-time rule mining under formal guarantees
  5. NeurIPS 2026LLMs Show No Signs Of Individuated Metacognition
  6. NeurIPS 2026LoopRPT: Reinforcement Pre-Training for Looped Language Models
  7. NeurIPS 2026Low-Rank Hierarchical Merging for Efficient Long-to-Short Reasoning
  8. NeurIPS 2026Loyalty Capture: Reporting Relationships and Structural Sycophancy in Frontier AI Models
  9. NeurIPS 2026LPG: Balancing Efficiency and Policy Reasoning in Latent Policy Guardrails
  10. NeurIPS 2026LUMOS: Tracing Parametric Knowledge from Training Data to Behavioral Outputs in LLMs
  11. NeurIPS 2026Mamba Can Learn Low-Dimensional Targets In-Context via Test-Time Feature Learning
  12. NeurIPS 2026MARBLE: an Agent Benchmark for Spatial Reasoning and Visual Abstraction
  13. NeurIPS 2026Matching-while-Decoding: Enhancing Template-Free Retrosynthesis via Explicit Structural Alignment
  14. NeurIPS 2026Measuring Weak-to-Strong Legibility of Reasoning Models
  15. NeurIPS 2026Med-Agentic: Distilling Agentic Medical Reasoning with Internalized Meta-Capabilities
  16. NeurIPS 2026MedPsy: State‑of‑the‑Art Small Medical Language Models for Efficient Edge Deployment
  17. NeurIPS 2026Membrane Sensitivity and Deployment Fragility of Learnable Time Constants in Spiking Neural Networks
  18. NeurIPS 2026Mental Imagery That Matters: Curing Latent Collapse in Multimodal Reasoning
  19. NeurIPS 2026MILES: Modular Instruction Memory with Learnable Selection for Self-Improving LLM Reasoning
  20. NeurIPS 2026MindLoom: Composing Thought Modes for Frontier-Level Reasoning Data Synthesis
  21. NeurIPS 2026Mining Logic under Uncertainty: Probabilistic Soft Logic with Energy-Based Inference for Chain-of-Thought Verification
  22. NeurIPS 2026MinPath: Learning Efficient LLM Reasoning via Minimal Dependency Paths
  23. NeurIPS 2026MIRA: Mutual Information guided calibration for Reliable Test-Time Adaptation
  24. NeurIPS 2026Mitigating Factual Hallucination in Large Reasoning Models via Mixed-Mode Advantage Regularization
  25. NeurIPS 2026Mitigating Overthinking in Large Reasoning Language Models via Reasoning Path Deviation Monitoring
  26. NeurIPS 2026Mixture of Layers: Dynamic Layer Routing for Visual Reasoning
  27. NeurIPS 2026MLLMEraser: Achieving Test-Time Unlearning in Multimodal Large Language Models through Activation Steering
  28. NeurIPS 2026MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI
  29. NeurIPS 2026MMFineReason: Closing the Multimodal Reasoning Gap via Open Data-Centric Methods
  30. NeurIPS 2026Monitoring the Internal Monologue: Probe Trajectories Reveal Reasoning Dynamics
  31. NeurIPS 2026Monotone Inclusion Approach to Weakly Monotone Discrete-Time Finite-Horizon Mean-Field Games
  32. NeurIPS 2026More Than Meets the Eye? Uncovering the Reasoning-Planning Disconnect in Training Vision-Language Driving Models
  33. NeurIPS 2026MotionHalluc: Diagnosing Kinematic Hallucinations in Fine-Grained Motion Reasoning
  34. NeurIPS 2026MUST: Stage-Adaptive Stability Control for Test-Time Scaling in Multimodal Reasoning
  35. NeurIPS 2026MUX: Continuous Reasoning via Multiplexed Tokens
  36. NeurIPS 2026Natural Synthesis: Outperforming Reactive Synthesis Tools with Large Reasoning Models
  37. NeurIPS 2026Nemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models
  38. NeurIPS 2026NEST: Nascent Encoded Steganographic Thoughts
  39. NeurIPS 2026Neural Garbage Collection: Learning to Forget while Learning to Reason
  40. NeurIPS 2026Neural Spectral Capacity: An Architectural Quantity from Network Specification Alone
  41. NeurIPS 2026Neurosymbolic Learning for Inference-Time Argumentation
  42. NeurIPS 2026Never Stop Learning: Test-Time Reinforcement Learning for Vision-Language-Action Models
  43. NeurIPS 2026Nexus: Same Pretraining Loss, Better Downstream Generalization via Common Minima
  44. NeurIPS 2026NPCBench: A Clinical Apprenticeship Benchmark for Guideline-Constrained Care-Pathway Reasoning in Nasopharyngeal Carcinoma
  45. NeurIPS 2026OmniSimulator: Aligning Small Language Models for Authentic Heterogeneous Behavior Modeling
  46. NeurIPS 2026OmniTraffic: A Controllable Generation Pipeline and Benchmark for Spatio-Temporal Traffic Reasoning
  47. NeurIPS 2026On the Primacy Bias in RLVR Training
  48. NeurIPS 2026On the Token Value Inequality in Efficient Reasoning
  49. NeurIPS 2026One-Layer Transformers Provably Learn In-Context K-Nearest Neighbor Prediction with Chain-of-Thought
  50. NeurIPS 2026OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework

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