混合专家论文趋势 2023–2026

主要结论

• 混合专家论文从 2023 年的 236 篇增长到 2026 年的 326 篇，三年间扩大到约 1.4 倍。
• NeurIPS 是发混合专家论文最多的会议，2023–2026 年共 354 篇，占四个会议总量的 39%。
• 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 4 篇增至 2026 年的 79 篇。
• 在「优化与缩放」的 5 个主题里，它的 2026 年论文数排第 4；同组增长最快的缩放定律是 4.1 倍，混合专家是 1.4 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 Biconvex Formulation for Stable Transport of Mixture Models with a Unique Solution
• ICML 2026$\phi$-Balancing for Mixture-of-Experts Training
• ICLR 2026A Federated Generalized Expectation-Maximization Algorithm for Mixture Models with an Unknown Number of Components
• CVPR 2026Divide Conquer and Aggregate Asymmetric Experts for Class-Imbalanced Semi-Supervised Medical
• NeurIPS 2025Advancing Expert Specialization for Better MoE
• ICML 2025$\texttt{I$^2$MoE}$: Interpretable Multimodal Interaction-aware Mixture-of-Experts
• ICLR 2025A Sanity Check for AI-generated Image Detection
• CVPR 2025Complexity Experts are Task-Discriminative Learners for Any Image Restoration
• NeurIPS 2024$\texttt{Model-GLUE}$: Democratized LLM Scaling for A Large Model Zoo in the Wild
• ICML 2024$\texttt{MoE-RBench}$: Towards Building Reliable Language Models with Sparse Mixture-of-Experts
• ICLR 2024Fusing Models with Complementary Expertise
• CVPR 2024MoDE: CLIP Data Experts via Clustering
• NeurIPS 2023[Re] Numerical influence of ReLU'(0) on backpropagation
• ICML 2023A Distribution Optimization Framework for Confidence Bounds of Risk Measures
• ICLR 2023Knowledge-in-Context: Towards Knowledgeable Semi-Parametric Language Models
• CVPR 2023Balanced Product of Calibrated Experts for Long-Tailed Recognition
• NeurIPS 2026AdERA: Adaptive Exponent Reuse for Lossless Allgather in Sharded MoE Training
• ICML 2026Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse
• ICLR 2026Abstracting Robot Manipulation Skills via Mixture-of-Experts Diffusion Policies
• CVPR 2026ERMoE Eigen-Reparameterized Mixture-of-Experts for Stable Routing and Interpretable Specialization

全部混合专家论文

共 138 篇，第 2 / 3 页

51. NeurIPS 2023Global Optimality and Finite Sample Analysis of Softmax Off-Policy Actor Critic under State Distribution Mismatch
52. NeurIPS 2023Graph Mixture of Experts: Learning on Large-Scale Graphs with Explicit Diversity Modeling
53. NeurIPS 2023Group Robust Classification Without Any Group Information
54. NeurIPS 2023Hardness of Low Rank Approximation of Entrywise Transformed Matrix Products
55. NeurIPS 2023Imbalanced Mixed Linear Regression
56. NeurIPS 2023Implicit Bias of Gradient Descent for Logistic Regression at the Edge of Stability
57. NeurIPS 2023Improved Convergence in High Probability of Clipped Gradient Methods with Heavy Tailed Noise
58. NeurIPS 2023Inconsistency, Instability, and Generalization Gap of Deep Neural Network Training
59. NeurIPS 2023Initialization Matters: Privacy-Utility Analysis of Overparameterized Neural Networks
60. NeurIPS 2023Intriguing Properties of Quantization at Scale
61. NeurIPS 2023Kernel Stein Discrepancy thinning: a theoretical perspective of pathologies and a practical fix with regularization
62. NeurIPS 2023Large Language Model as Attributed Training Data Generator: A Tale of Diversity and Bias
63. NeurIPS 2023Local Convergence of Gradient Methods for Min-Max Games: Partial Curvature Generically Suffices
64. NeurIPS 2023Massively Multilingual Corpus of Sentiment Datasets and Multi-faceted Sentiment Classification Benchmark
65. NeurIPS 2023Minimax Optimal Rate for Parameter Estimation in Multivariate Deviated Models
66. NeurIPS 2023Mitigating the Popularity Bias of Graph Collaborative Filtering: A Dimensional Collapse Perspective
67. NeurIPS 2023Mixed-Initiative Multiagent Apprenticeship Learning for Human Training of Robot Teams
68. NeurIPS 2023Multi-Modal Inverse Constrained Reinforcement Learning from a Mixture of Demonstrations
69. NeurIPS 2023MVDoppler: Unleashing the Power of Multi-View Doppler for MicroMotion-based Gait Classification
70. NeurIPS 2023Near Optimal Reconstruction of Spherical Harmonic Expansions
71. NeurIPS 2023No-Regret Learning with Unbounded Losses: The Case of Logarithmic Pooling
72. NeurIPS 2023No-Regret Online Prediction with Strategic Experts
73. NeurIPS 2023Non-Asymptotic Analysis of a UCB-based Top Two Algorithm
74. NeurIPS 2023Objaverse-XL: A Universe of 10M+ 3D Objects
75. NeurIPS 2023Offline Reinforcement Learning for Mixture-of-Expert Dialogue Management
76. NeurIPS 2023On Convergence of Polynomial Approximations to the Gaussian Mixture Entropy
77. NeurIPS 2023On the Complexity of Differentially Private Best-Arm Identification with Fixed Confidence
78. NeurIPS 2023On the Connection between Pre-training Data Diversity and Fine-tuning Robustness
79. NeurIPS 2023On the Convergence of CART under Sufficient Impurity Decrease Condition
80. NeurIPS 2023On the Convergence of Encoder-only Shallow Transformers
81. NeurIPS 2023On the Exploration of Local Significant Differences For Two-Sample Test
82. NeurIPS 2023On the Generalization Error of Stochastic Mirror Descent for Quadratically-Bounded Losses: an Improved Analysis
83. NeurIPS 2023On the Identifiability and Interpretability of Gaussian Process Models
84. NeurIPS 2023On the Implicit Bias of Linear Equivariant Steerable Networks
85. NeurIPS 2023On the Powerfulness of Textual Outlier Exposure for Visual OoD Detection
86. NeurIPS 2023On the Size and Approximation Error of Distilled Datasets
87. NeurIPS 2023On the Variance, Admissibility, and Stability of Empirical Risk Minimization
88. NeurIPS 2023One-step differentiation of iterative algorithms
89. NeurIPS 2023Opening the Vocabulary of Egocentric Actions
90. NeurIPS 2023Optimal Parameter and Neuron Pruning for Out-of-Distribution Detection
91. NeurIPS 2023Optimization of Inter-group criteria for clustering with minimum size constraints
92. NeurIPS 2023Persuading Farsighted Receivers in MDPs: the Power of Honesty
93. NeurIPS 2023PoET: A generative model of protein families as sequences-of-sequences
94. NeurIPS 2023Predict-then-Calibrate: A New Perspective of Robust Contextual LP
95. NeurIPS 2023Private estimation algorithms for stochastic block models and mixture models
96. NeurIPS 2023Provable benefits of annealing for estimating normalizing constants: Importance Sampling, Noise-Contrastive Estimation, and beyond
97. NeurIPS 2023Provably Fast Finite Particle Variants of SVGD via Virtual Particle Stochastic Approximation
98. NeurIPS 2023RAPHAEL: Text-to-Image Generation via Large Mixture of Diffusion Paths
99. NeurIPS 2023Reconciling Competing Sampling Strategies of Network Embedding
100. NeurIPS 2023Regularization properties of adversarially-trained linear regression

常见问题

2023–2026 年四大会议一共有多少篇混合专家论文？

共 904 篇，按年份分别是 2023 年 236 篇、2024 年 119 篇、2025 年 223 篇、2026 年 326 篇。

NeurIPS 2026 有多少篇混合专家论文？

NeurIPS 2026 有 75 篇混合专家论文，上一届是 86 篇。

ICLR 2026 有多少篇混合专家论文？

ICLR 2026 有 73 篇混合专家论文，上一届是 59 篇。

ICML 2026 有多少篇混合专家论文？

ICML 2026 有 132 篇混合专家论文，上一届是 57 篇。

CVPR 2026 有多少篇混合专家论文？

CVPR 2026 有 46 篇混合专家论文，上一届是 21 篇。

哪个会议的混合专家论文最多？

NeurIPS，2023–2026 年共 354 篇。