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 354 篇，第 2 / 8 页

51. NeurIPS 2026Move on Muon : A Hamiltonian probability gradient flow perspective of Muon optimizer
52. NeurIPS 2026N-vium: Mixture-of-Exits Transformer for Accelerated Exact Generation
53. NeurIPS 2026Non-Expanding Gated Continued Fraction Architecture for Feedforward Layers in Language Models
54. NeurIPS 2026Not All Routing Drift Is Harmful: Trust-Region Projection for Class-Incremental Learning
55. NeurIPS 2026One-Bit Clustering for Two Component Sub-Gaussian Mixture Models
56. NeurIPS 2026PGMS: Pyramidal Gaussian Mixture Splatting for 3DGS Compression
57. NeurIPS 2026Physics-Informed Video Generation via Mixture-of-Experts Latent Alignment
58. NeurIPS 2026PithTrain: A Compact and Agent-Native MoE Training System
59. NeurIPS 2026POISE: Instance-Specific Prompt Tuning under Latent Mixture Target Distributions
60. NeurIPS 2026Predicting Only from Selected Evidence: A Tempered Product-of-Experts Bottleneck for Auditable EEG Diagnosis
61. NeurIPS 2026PriSM: Prior-guided Shared-basis Mixture Personalization for LLMs under Sparse User Histories
62. NeurIPS 2026Private Online Prediction from Experts with Small Losses
63. NeurIPS 2026Raw-Routed Mixture of Adapters: A Causal Intervention for Routing Collapse in Time Series Foundation Models
64. NeurIPS 2026Replicas-as-Variables: A Planner for Throughput-Optimal Routing in LLM Deployment
65. NeurIPS 2026Residual Paving: Diagnosing the Routing Bottleneck in Selective Refusal Editing
66. NeurIPS 2026Route-Consistent Adaptation for Stable Quantization of Mixture-of-Experts Models with Theoretical Guarantees
67. NeurIPS 2026S²MoE: Shared-Subspace Mixture of Sparse Experts
68. NeurIPS 2026SAME: Stability-Aware Embedding Extraction in Mixture-of-Experts Language Models
69. NeurIPS 2026SimReg: Achieving Higher Performance in the Pretraining via Embedding Similarity Regularization
70. NeurIPS 2026Spectral Geometry of Attention: From Information Routing to Uncertainty
71. NeurIPS 2026Structure, Subspace and System: Push the Real Limit of Extremely Low-Bit Quantization for MoE-LLMs
72. NeurIPS 2026Swimba: Switch Mamba Model Scales State Space Models
73. NeurIPS 2026Training-Free Dynamic Upcycling of Expert Language Models
74. NeurIPS 2026vExpert: Virtualizing Expert Storage for Adaptive Load Balancing in Distributed MoE Inference
75. NeurIPS 2026Why Routers Freeze: Infinite Width Learning Dynamics for Mixture of Experts
76. NeurIPS 2025A Unified Framework for Variable Selection in Model-Based Clustering with Missing Not at Random
77. NeurIPS 2025Adaptive LoRA Experts Allocation and Selection for Federated Fine-Tuning
78. NeurIPS 2025Advancing Expert Specialization for Better MoE
79. NeurIPS 2025ALTER: All-in-One Layer Pruning and Temporal Expert Routing for Efficient Diffusion Generation
80. NeurIPS 2025Blindfolded Experts Generalize Better: Insights from Robotic Manipulation and Videogames
81. NeurIPS 2025Brain-Like Processing Pathways Form in Models With Heterogeneous Experts
82. NeurIPS 2025BrainMoE: Cognition Joint Embedding via Mixture-of-Expert Towards Robust Brain Foundation Model
83. NeurIPS 2025Consistent Sampling and Simulation: Molecular Dynamics with Energy-Based Diffusion Models
84. NeurIPS 2025Continual Gaussian Mixture Distribution Modeling for Class Incremental Semantic Segmentation
85. NeurIPS 2025CryptoMoE: Privacy-Preserving and Scalable Mixture of Experts Inference via Balanced Expert Routing
86. NeurIPS 2025Data Mixture Optimization: A Multi-fidelity Multi-scale Bayesian Framework
87. NeurIPS 2025Deferring Concept Bottleneck Models: Learning to Defer Interventions to Inaccurate Experts
88. NeurIPS 2025Dense Backpropagation Improves Training for Sparse Mixture-of-Experts
89. NeurIPS 2025DiEP: Adaptive Mixture-of-Experts Compression through Differentiable Expert Pruning
90. NeurIPS 2025Discovering Important Experts for Mixture-of-Experts Models Pruning Through a Theoretical Perspective
91. NeurIPS 2025Diverse Influence Component Analysis: A Geometric Approach to Nonlinear Mixture Identifiability
92. NeurIPS 2025Diversity as a Reward: Fine-Tuning LLMs on a Mixture of Domain-Undetermined Data
93. NeurIPS 2025Domain-Specific Pruning of Large Mixture-of-Experts Models with Few-shot Demonstrations
94. NeurIPS 2025Enhanced Expert Merging for Mixture-of-Experts in Graph Foundation Models
95. NeurIPS 2025Evaluating multiple models using labeled and unlabeled data
96. NeurIPS 2025FAME: Adaptive Functional Attention with Expert Routing for Function-on-Function Regression
97. NeurIPS 2025Fisher meets Feynman: score-based variational inference with a product of experts
98. NeurIPS 2025FlashMoE: Fast Distributed MoE in a Single Kernel
99. NeurIPS 2025FlowMoE: A Scalable Pipeline Scheduling Framework for Distributed Mixture-of-Experts Training
100. NeurIPS 2025FlyLoRA: Boosting Task Decoupling and Parameter Efficiency via Implicit Rank-Wise Mixture-of-Experts

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。