推理模型论文趋势 2023–2026

主要结论

• 推理模型论文从 2023 年的 470 篇增长到 2026 年的 1,799 篇，三年间扩大到约 3.8 倍。
• NeurIPS 是发推理模型论文最多的会议，2023–2026 年共 1,253 篇，占四个会议总量的 36%。
• 交叉方向里涨得最快的是强化学习：同时属于两个主题的论文从 2023 年的 9 篇增至 2026 年的 437 篇。
• 在「前沿系统」的 4 个主题里，它的 2026 年论文数排第 3；同组增长最快的AI 智能体是 12 倍，推理模型是 3.8 倍。

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

• NeurIPS 2026A Mechanistic Analysis of Looped Reasoning Language Models
• ICML 2026A Formal Comparison Between Chain of Thought and Latent Thought
• ICLR 2026A Simple "Motivation" Can Enhance Reinforcement Finetuning of Large Reasoning Models
• CVPR 2026Asking like Socrates Socrates helps VLMs understand remote sensing images
• NeurIPS 2025A Frustratingly Simple Yet Highly Effective Attack Baseline: Over 90% Success Rate Against the Strong Black-box Models of GPT-4.5/4o/o1
• ICML 2025A Reasoning-Based Approach to Cryptic Crossword Clue Solving
• ICLR 2025A Formal Framework for Understanding Length Generalization in Transformers
• CVPR 2025Argus: Vision-Centric Reasoning with Grounded Chain-of-Thought
• NeurIPS 2024Are Large Language Models Good Statisticians?
• ICML 2024Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
• ICLR 2024Amortizing intractable inference in large language models
• CVPR 2024Generate Subgoal Images before Act: Unlocking the Chain-of-Thought Reasoning in Diffusion Model for Robot Manipulation with Multimodal Prompts
• NeurIPS 2023A benchmark of categorical encoders for binary classification
• ICML 2023Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models?
• ICLR 2023Automatic Chain of Thought Prompting in Large Language Models
• CVPR 2023ActMAD: Activation Matching To Align Distributions for Test-Time-Training
• NeurIPS 2026A Solvable Model of Chain-of-Thought in In-Context Learning
• ICML 2026A Narrowing Geometry in Contaminated Reasoning
• ICLR 2026A State-Transition Framework for Efficient LLM Reasoning
• CVPR 2026Beyond Weak Supervision MLLMs-Guided Graded Knowledge Distillation for Unsupervised Camouflaged

全部推理模型论文

共 204 篇，第 2 / 5 页

51. CVPR 2026Efficiently Reconstructing Dynamic Scenes One D4RT at a Time
52. CVPR 2026EgoMind Activating Spatial Cognition through Linguistic Reasoning in MLLMs
53. CVPR 2026EgoProx Evaluating MLLMs on Egocentric 3D Proximity Reasoning Across a
54. CVPR 2026EmoThinker Advancing Visual-Acoustic Emotion Analysis via Structural Token Selection and
55. CVPR 2026Energy Waveify and Redistribution for Test-Time Adaptation A Control System
56. CVPR 2026Enhancing Video Vision Language Model with Hippocampal Sensing
57. CVPR 2026Evolutionary Multimodal Reasoning via Hierarchical Semantic Representation for Intent Recognition
58. CVPR 2026FantasyVLN Unified Multimodal Chain-of-Thought Reasoning for Vision-and-Language Navigation
59. CVPR 2026Fast Reasoning Segmentation for Images and Videos
60. CVPR 2026Fast-ThinkAct Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning
61. CVPR 2026FireScope Wildfire Risk Raster Prediction With a Chain-of-Thought Oracle
62. CVPR 2026FOZO Forward-Only Zeroth-Order Prompt Optimization for Test-Time Adaptation
63. CVPR 2026From Contrast to Consistency Rethinking Event-based Continuous-Time Optical Flow Estimation
64. CVPR 2026From Intuition to Investigation A Tool-Augmented Reasoning MLLM Framework for
65. CVPR 2026From Manuals to Actions A Unified VLA Model for Chain-of-Thought
66. CVPR 2026From Scale to Speed Adaptive Test-Time Scaling for Image Editing
67. CVPR 2026Fuel Gauge Estimating Chain-of-Thought Length Ahead of Time in Large
68. CVPR 2026FunFact Building Probabilistic Functional 3D Scene Graphs via Factor-Graph Reasoning
69. CVPR 2026G-MIXER Geodesic Mixup-based Implicit Semantic Expansion and Explicit Semantic Re-ranking
70. CVPR 2026Gated KalmaNet A Fading Memory Layer through Test-time Ridge Regression
71. CVPR 2026GeoAgent Learning to Geolocate Everywhere with Reinforced Geographic Characteristics
72. CVPR 2026GeoCoT Towards Reliable Remote Sensing Reasoning with Manifold Perspective
73. CVPR 2026Geoint-R1 Formalizing Multimodal Geometric Reasoning with Dynamic Auxiliary Constructions
74. CVPR 2026Global Underwater Geolocation from Time-Lapse Polarization Imagery
75. CVPR 2026Grounded 3D-Aware Spatial Vision-Language Modeling
76. CVPR 2026Grounded Chain-of-Thought for Multimodal Large Language Models
77. CVPR 2026GuardTrace-VL Detecting Unsafe Multimodel Reasoning via Iterative Safety Supervision
78. CVPR 2026Harnessing Chain-of-Thought Reasoning in Multimodal Large Language Models for Face
79. CVPR 2026Hear you are Teaching LLMs Spatial Reasoning with Vision and
80. CVPR 2026Hilbert-Geo Solving Solid Geometric Problems by Neural-Symbolic Reasoning
81. CVPR 2026History to Future Evolving Agent with Experience and Thought for
82. CVPR 2026HoneyBee Data Recipes for Vision-Language Reasoners
83. CVPR 2026HSI-GPT2 A Dual-Granularity Large Motion Reasoning Model with Diffusion Refinement
84. CVPR 2026Human-like Abstract Visual Reasoning via Understanding and Solving Reasoning Loop
85. CVPR 2026HUMORCHAIN Theory-Guided Multi-Stage Reasoning for Interpretable Multimodal Humor Generation
86. CVPR 2026HybridDriveVLA Vision-Language-Action Model with Visual CoT reasoning and ToT Evaluation
87. CVPR 2026Hyperbolic Prototype Learning with Uncertainty-Aware Consistency for Continual Test-Time Segmentation
88. CVPR 2026Hypergraph-State Collaborative Reasoning for Multi-Object Tracking
89. CVPR 2026Incentivizing Versatile Video Reasoning in MLLMs via Data-Efficient Reinforcement Learning
90. CVPR 2026Inference-time Physics Alignment of Video Generative Models with Latent World
91. CVPR 2026Inter-Photon-Limited Videography
92. CVPR 2026iSHIFT Lightweight Slow-Fast GUI Agent with Adaptive Perception
93. CVPR 2026JarvisEvo Towards a Self-Evolving Photo Editing Agent with Synergistic Editor-Evaluator
94. CVPR 2026KV-Tracker Real-Time Pose Tracking with Transformers
95. CVPR 2026LASAR Towards Spatio-temporal Reasoning with Latent Cognitive Map
96. CVPR 2026Latent Chain-of-Thought World Modeling for End-to-End Autonomous Driving
97. CVPR 2026Latent Implicit Visual Reasoning
98. CVPR 2026LayoutAD Exploring Semantic-Geometric Misalignment Reasoning for Scene Layout Anomaly Detection
99. CVPR 2026Let VLMs Grade Their Own Thoughts A Self-Quantification Approach to
100. CVPR 2026Leveraging Verifier-Based Reinforcement Learning in Image Editing

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。