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 119 篇，第 2 / 3 页

51. ICML 2024Mixtures of Experts Unlock Parameter Scaling for Deep RL
52. ICML 2024MVMoE: Multi-Task Vehicle Routing Solver with Mixture-of-Experts
53. ICML 2024On Interpolating Experts and Multi-Armed Bandits
54. ICML 2024On Least Square Estimation in Softmax Gating Mixture of Experts
55. ICML 2024One Prompt is not Enough: Automated Construction of a Mixture-of-Expert Prompts
56. ICML 2024OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models
57. ICML 2024Receptive Fields As Experts in Convolutional Neural Architectures
58. ICML 2024Scaling Beyond the GPU Memory Limit for Large Mixture-of-Experts Model Training
59. ICML 2024Scaling Laws for Fine-Grained Mixture of Experts
60. ICML 2024See More Details: Efficient Image Super-Resolution by Experts Mining
61. ICML 2024Theoretical Guarantees for Variational Inference with Fixed-Variance Mixture of Gaussians
62. ICML 2024Theoretical insights for diffusion guidance: A case study for Gaussian mixture models
63. ICML 2024Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts
64. ICML 2024Whispering Experts: Neural Interventions for Toxicity Mitigation in Language Models
65. NeurIPS 2024$\textit{Read-ME}$: Refactorizing LLMs as Router-Decoupled Mixture of Experts with System Co-Design
66. NeurIPS 2024$\texttt{Model-GLUE}$: Democratized LLM Scaling for A Large Model Zoo in the Wild
67. NeurIPS 2024Achieving Optimal Clustering in Gaussian Mixture Models with Anisotropic Covariance Structures
68. NeurIPS 2024BAM! Just Like That: Simple and Efficient Parameter Upcycling for Mixture of Experts
69. NeurIPS 2024Breaking Long-Tailed Learning Bottlenecks: A Controllable Paradigm with Hypernetwork-Generated Diverse Experts
70. NeurIPS 2024Continual learning with the neural tangent ensemble
71. NeurIPS 2024CuMo: Scaling Multimodal LLM with Co-Upcycled Mixture-of-Experts
72. NeurIPS 2024Customizing Language Models with Instance-wise LoRA for Sequential Recommendation
73. NeurIPS 2024Data Mixture Inference Attack: BPE Tokenizers Reveal Training Data Compositions
74. NeurIPS 2024DenseFusion-1M: Merging Vision Experts for Comprehensive Multimodal Perception
75. NeurIPS 2024Diffusion Actor-Critic with Entropy Regulator
76. NeurIPS 2024Evidential Mixture Machines: Deciphering Multi-Label Correlations for Active Learning Sensitivity
77. NeurIPS 2024Exploiting Activation Sparsity with Dense to Dynamic-k Mixture-of-Experts Conversion
78. NeurIPS 2024Federated Online Prediction from Experts with Differential Privacy: Separations and Regret Speed-ups
79. NeurIPS 2024Flex-MoE: Modeling Arbitrary Modality Combination via the Flexible Mixture-of-Experts
80. NeurIPS 2024FuseMoE: Mixture-of-Experts Transformers for Fleximodal Fusion
81. NeurIPS 2024GraphMETRO: Mitigating Complex Graph Distribution Shifts via Mixture of Aligned Experts
82. NeurIPS 2024How does Architecture Influence the Base Capabilities of Pre-trained Language Models? A Case Study Based on FFN-Wider and MoE Transformers
83. NeurIPS 2024LFME: A Simple Framework for Learning from Multiple Experts in Domain Generalization
84. NeurIPS 2024LSH-MoE: Communication-efficient MoE Training via Locality-Sensitive Hashing
85. NeurIPS 2024Mixture of Demonstrations for In-Context Learning
86. NeurIPS 2024Mixture of Experts Meets Prompt-Based Continual Learning
87. NeurIPS 2024Mixture of In-Context Experts Enhance LLMs' Long Context Awareness
88. NeurIPS 2024Mixture of Link Predictors on Graphs
89. NeurIPS 2024Mixture of Nested Experts: Adaptive Processing of Visual Tokens
90. NeurIPS 2024Mixture of neural fields for heterogeneous reconstruction in cryo-EM
91. NeurIPS 2024Mixture of Scales: Memory-Efficient Token-Adaptive Binarization for Large Language Models
92. NeurIPS 2024Mixture of Tokens: Continuous MoE through Cross-Example Aggregation
93. NeurIPS 2024MoE Jetpack: From Dense Checkpoints to Adaptive Mixture of Experts for Vision Tasks
94. NeurIPS 2024MoEUT: Mixture-of-Experts Universal Transformers
95. NeurIPS 2024MoLE: Enhancing Human-centric Text-to-image Diffusion via Mixture of Low-rank Experts
96. NeurIPS 2024MoME: Mixture of Multimodal Experts for Generalist Multimodal Large Language Models
97. NeurIPS 2024MomentumSMoE: Integrating Momentum into Sparse Mixture of Experts
98. NeurIPS 2024MoVA: Adapting Mixture of Vision Experts to Multimodal Context
99. NeurIPS 2024Multi-Head Mixture-of-Experts
100. NeurIPS 2024Multilinear Mixture of Experts: Scalable Expert Specialization through Factorization

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。