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 470 篇，第 2 / 10 页

51. ICLR 2023Large Language Models are Human-Level Prompt Engineers
52. ICLR 2023Learning Math Reasoning from Self-Sampled Correct and Partially-Correct Solutions
53. ICLR 2023Learning to Extrapolate: A Transductive Approach
54. ICLR 2023Learning to Solve Constraint Satisfaction Problems with Recurrent Transformer
55. ICLR 2023Learning where and when to reason in neuro-symbolic inference
56. ICLR 2023Least-to-Most Prompting Enables Complex Reasoning in Large Language Models
57. ICLR 2023LogicDP: Creating Labels for Graph Data via Inductive Logic Programming
58. ICLR 2023Lower Bounds on the Depth of Integral ReLU Neural Networks via Lattice Polytopes
59. ICLR 2023M-L2O: Towards Generalizable Learning-to-Optimize by Test-Time Fast Self-Adaptation
60. ICLR 2023MECTA: Memory-Economic Continual Test-Time Model Adaptation
61. ICLR 2023Mind's Eye: Grounded Language Model Reasoning through Simulation
62. ICLR 2023Multimodal Analogical Reasoning over Knowledge Graphs
63. ICLR 2023Neural Compositional Rule Learning for Knowledge Graph Reasoning
64. ICLR 2023NTK-SAP: Improving neural network pruning by aligning training dynamics
65. ICLR 2023On Achieving Optimal Adversarial Test Error
66. ICLR 2023PINTO: Faithful Language Reasoning Using Prompt-Generated Rationales
67. ICLR 2023ReAct: Synergizing Reasoning and Acting in Language Models
68. ICLR 2023Relational Attention: Generalizing Transformers for Graph-Structured Tasks
69. ICLR 2023Rethinking skip connection model as a learnable Markov chain
70. ICLR 2023ROSCOE: A Suite of Metrics for Scoring Step-by-Step Reasoning
71. ICLR 2023Selection-Inference: Exploiting Large Language Models for Interpretable Logical Reasoning
72. ICLR 2023Self-Consistency Improves Chain of Thought Reasoning in Language Models
73. ICLR 2023Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language
74. ICLR 2023SQA3D: Situated Question Answering in 3D Scenes
75. ICLR 2023STREET: A MULTI-TASK STRUCTURED REASONING AND EXPLANATION BENCHMARK
76. ICLR 2023TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second
77. ICLR 2023Temporal Coherent Test Time Optimization for Robust Video Classification
78. ICLR 2023Test-Time Adaptation via Self-Training with Nearest Neighbor Information
79. ICLR 2023Test-Time Robust Personalization for Federated Learning
80. ICLR 2023Towards Inferential Reproducibility of Machine Learning Research
81. ICLR 2023Towards Stable Test-time Adaptation in Dynamic Wild World
82. ICLR 2023Towards Understanding GD with Hard and Conjugate Pseudo-labels for Test-Time Adaptation
83. ICLR 2023TrojText: Test-time Invisible Textual Trojan Insertion
84. ICLR 2023TTN: A Domain-Shift Aware Batch Normalization in Test-Time Adaptation
85. ICLR 2023UL2: Unifying Language Learning Paradigms
86. ICLR 2023UniKGQA: Unified Retrieval and Reasoning for Solving Multi-hop Question Answering Over Knowledge Graph
87. ICLR 2023Visual Recognition with Deep Nearest Centroids
88. ICLR 2023Visually-Augmented Language Modeling
89. ICLR 2023Weakly Supervised Explainable Phrasal Reasoning with Neural Fuzzy Logic
90. ICLR 2023Weakly Supervised Knowledge Transfer with Probabilistic Logical Reasoning for Object Detection
91. ICML 2023A Statistical Perspective on Retrieval-Based Models
92. ICML 2023A Study of Global and Episodic Bonuses for Exploration in Contextual MDPs
93. ICML 2023A Unified Optimization Framework of ANN-SNN Conversion: Towards Optimal Mapping from Activation Values to Firing Rates
94. ICML 2023AdaNPC: Exploring Non-Parametric Classifier for Test-Time Adaptation
95. ICML 2023Adaptive Computation with Elastic Input Sequence
96. ICML 2023An Effective Meaningful Way to Evaluate Survival Models
97. ICML 2023Bag of Tricks for Training Data Extraction from Language Models
98. ICML 2023Bandits with Knapsacks: Advice on Time-Varying Demands
99. ICML 2023Bayesian Reparameterization of Reward-Conditioned Reinforcement Learning with Energy-based Models
100. ICML 2023Benign Overfitting in Deep Neural Networks under Lazy Training

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。