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 35 / 36 页

  1. NeurIPS 2026Safe in Its Own Words: Self-Guided Safety Alignment for Multimodal Reasoning Models
  2. NeurIPS 2026SAGE: Mitigating Long-Horizon Reasoning Biases via Topological Guidance
  3. NeurIPS 2026Saliency-Aware Multi-Route Thinking: Grounding and Reasoning on Vision-Language Agents
  4. NeurIPS 2026Same Signal, Opposite Meaning: Direction-Informed Adaptive Learning for LLM Agents
  5. NeurIPS 2026SAVeR$^2$: Reasoning-based Safety Alignment for Large Reasoning Models via Verifiable Rewards
  6. NeurIPS 2026Scalable Token-Level Hallucination Detection in Large Language Models
  7. NeurIPS 2026Scaling Causal Reasoning with Increasingly Complex Causal Simulators
  8. NeurIPS 2026SciR: A Controllable Benchmark for Scientific Reasoning in LLMs
  9. NeurIPS 2026Scratchpad Patching: Decoupling Compute from Patch Size in Byte-Level Language Models
  10. NeurIPS 2026See to Believe: Segmentation by Reasoning with Visual Evidence
  11. NeurIPS 2026Seeing or Rationalizing? Scene-Evidence-Guided Chain-of-Thought for Faithful 3D Multimodal Reasoning
  12. NeurIPS 2026Seeing Through the Chain: Understanding and Mitigating Hallucinations in Multimodal Large Reasoning Models
  13. NeurIPS 2026Seeing Together: Multi-Robot Cooperative Egocentric Spatial Reasoning with Multimodal Large Language Models
  14. NeurIPS 2026Seirênes: Adversarial Self-Play with Evolving Distractions for LLM Reasoning
  15. NeurIPS 2026Selected-Tail Reliability in Verifier-Guided Best-of-$N$ Inference
  16. NeurIPS 2026Self-Correction as Transition Geometry: Internalizing Reasoning via Lifted State Policy Optimization
  17. NeurIPS 2026Self-Rewarded Multimodal Coherent Reasoning Across Diverse Visual Domains
  18. NeurIPS 2026Self-Trained Verification for Training- and Test-Time Self-Improvement
  19. NeurIPS 2026SIGMA: A Sigmoid-Gated Sampler for Test-Time Scaling in Diffusion Language Models
  20. NeurIPS 2026Simple Test-Time Refinement for Plot-to-Code Generation via Visual-Code Diagnostics
  21. NeurIPS 2026SLVR: Structured Latent Visual Reasoning via Human-like Reasoning Flows
  22. NeurIPS 2026Smart Picks in the Dark: Towards Efficient RLVR for Reasoning via Tracing Metacognitive Pivots
  23. NeurIPS 2026SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning
  24. NeurIPS 2026Sparse Expansion Utility: Identifying and Routing to Pivotal Steps in LLM Reasoning Chains
  25. NeurIPS 2026Spatial-IQ: Deconstructing Spatial Intelligence via Hierarchical Capability Tests
  26. NeurIPS 2026SPECS: Faster Test-Time Scaling through Speculative Drafts and Dynamic Switching
  27. NeurIPS 2026Speculative Decoding Is Per-Token, Not Path
  28. NeurIPS 2026Spend Only What You Need: Defect-Aware Residual Coverage for Efficient Multi-Agent Reasoning
  29. NeurIPS 2026SSR3D-LLM: Structured Spatial Reasoning via Latent Steps for Fine-Grained Grounding in Unified 3D-LLMs
  30. NeurIPS 2026Stability in Multi-Step Reasoning via Jacobian-based Error Accumulation Analysis
  31. NeurIPS 2026State of Thought Enables Endogenous Reasoning
  32. NeurIPS 2026SteadyThought: Mitigating LLM Under-Thinking via Thought-Level Preference Optimization
  33. NeurIPS 2026STEER: Route-Aware Adaptive Reasoning for Autonomous Driving
  34. NeurIPS 2026Stepwise Penalization for Length-Efficient Chain-of-Thought Reasoning
  35. NeurIPS 2026Stop or Restart? Principled Inference Control for Large Reasoning Models via the Pandora's Box
  36. NeurIPS 2026StreamGaze: Gaze-Guided Temporal Reasoning and Proactive Understanding in Streaming Videos
  37. NeurIPS 2026Stress Testing Chain-of-Thought Monitoring Against Covert Misalignment
  38. NeurIPS 2026Structural Rationale Distillation via Reasoning Space Compression
  39. NeurIPS 2026Structured Sparse Memory for Recurrent Reasoning
  40. NeurIPS 2026SWE-Chain: Benchmarking Coding Agents on Chained Release-Level Package Upgrades
  41. NeurIPS 2026Taming CoT Obfuscation in VLMs: From Mechanistic Evidence to Activation-Level Enforcement
  42. NeurIPS 2026TaskGround: Structured Executable Task Inference for Full-Scene Household Reasoning
  43. NeurIPS 2026Tatemae: Detecting Alignment Faking via Tool Selection in LLMs
  44. NeurIPS 2026Teach-to-Reason: Competition-Guided Reasoning with a Self-Improving Teacher
  45. NeurIPS 2026Teaching Large Language Models When Not to Know: Learning Temporal Critique for Ex-Ante Reasoning
  46. NeurIPS 2026Teaching VLMs What to Say, Not How to Reason: Rethinking Counterfactual Reasoning in Autonomous Driving
  47. NeurIPS 2026Temporal Backtracking Search for Test-time Generative Video Reasoning
  48. NeurIPS 2026TERMINATOR: Learning Optimal Exit Points for Early Stopping in Chain-of-Thought Reasoning
  49. NeurIPS 2026Test Time Search Requires Training for Diversity
  50. NeurIPS 2026Test time training enhances in-context learning of nonlinear functions

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