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 296 篇，第 3 / 6 页

101. ICML 2026Single-Head Attention in High Dimensions: A Theory of Generalization, Weights Spectra, and Scaling Laws
102. ICML 2026Skill-Based Mixture-of-Experts: Adaptive Routing for Heterogeneous Reasoning via Inferred Skills
103. ICML 2026SoftMoE: Soft Differentiable Routing for Mixture-of-Experts in LLMs
104. ICML 2026SPARKLING: Balancing Signal Preservation and Symmetry Breaking for Width-Progressive Learning
105. ICML 2026Sparse Models, Sparse Safety: Unsafe Routes in Mixture-of-Experts LLMs
106. ICML 2026SpecMD: A Comprehensive Study On Speculative Expert Prefetching
107. ICML 2026SPHERE: Mitigating the Loss of Spectral Plasticity in Mixture-of-Experts for Deep Reinforcement Learning
108. ICML 2026Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers
109. ICML 2026STAR: Rethinking MoE Routing as Structure-Aware Subspace Learning
110. ICML 2026Synergistic Intra- and Cross-Layer Regularization Losses for MoE Expert Specialization
111. ICML 2026Task-Aware Mechanism: Hybrid MoE Vision Tower Towards Holistic Video Understanding
112. ICML 2026TEAM: Temporal–Spatial Consistency Guided Expert Activation for MoE Diffusion Language Model Acceleration
113. ICML 2026The Catastrophic Failure of *the* k-Means Algorithm in High Dimensions, and How Hartigan's Algorithm Avoids It
114. ICML 2026The Cost of Commitment in Option-Based Hierarchical RL
115. ICML 2026The Expert Strikes Back: Interpreting Mixture-of-Experts Language Models at Expert Level
116. ICML 2026TileQ: Efficient Low-Rank Quantization of Mixture-of-Experts with 2D Tiling
117. ICML 2026TIME: Tensor-Factorized Mixture-of-Experts with Intrinsic Routing for Lifelong Multimodal Knowledge Editing
118. ICML 2026TiME: Test-Time Mixture-of-Experts Routing via Asymmetric CO-Optimal Transport for Continual Test-Time Adaptation
119. ICML 2026TiMi: Empower Time Series Transformers with Multimodal Mixture of Experts
120. ICML 2026Toward Calibrated Mixture-of-Experts Under Distribution Shift
121. ICML 2026Toward Structural Multimodal Representations: Specialization, Selection, and Sparsification via Mixture-of-Experts
122. ICML 2026Towards a Unified Generative Model for Scarce Time Series with Domain Experts
123. ICML 2026Towards One-for-All Anomaly Detection for Tabular Data
124. ICML 2026TRACE: Trajectory Recovery for Continuous Mechanism Evolution in Causal Representation Learning
125. ICML 2026UB-SMoE: Universally Balanced Sparse Mixture-of-Experts for Resource-adaptive Federated Fine-tuning of Foundation Models
126. ICML 2026Unified Multimodal Visual Tracking with Dual Mixture-of-Experts
127. ICML 2026Variational Routing: A Scalable Bayesian Framework for Calibrated Mixture-of-Experts Transformers
128. ICML 2026WestWorld: A Knowledge-Encoded Scalable Trajectory World Model for Diverse Robotic Systems
129. ICML 2026When Model Merging Breaks Routing: Training-Free Calibration for MoE
130. ICML 2026When More Experts Hurt: Underfitting in Multi-Expert Learning to Defer
131. ICML 2026X-MoGe: A Cross-Modal Adaptation Framework with Mixture-of-Experts and Geometry Guidance for Heterogeneous Collaborative Perception
132. ICML 2026ZipMoE: Efficient On-Device MoE Serving via Lossless Compression and Cache-Affinity Scheduling
133. ICML 2025$\texttt{I$^2$MoE}$: Interpretable Multimodal Interaction-aware Mixture-of-Experts
134. ICML 2025A Generic Family of Graphical Models: Diversity, Efficiency, and Heterogeneity
135. ICML 2025A Stronger Mixture of Low-Rank Experts for Fine-Tuning Foundation Models
136. ICML 2025Addressing Imbalanced Domain-Incremental Learning through Dual-Balance Collaborative Experts
137. ICML 2025Autonomy-of-Experts Models
138. ICML 2025Bayesian Inference for Correlated Human Experts and Classifiers
139. ICML 2025Chameleon: A Flexible Data-mixing Framework for Language Model Pretraining and Finetuning
140. ICML 2025Cooperation of Experts: Fusing Heterogeneous Information with Large Margin
141. ICML 2025DEFAME: Dynamic Evidence-based FAct-checking with Multimodal Experts
142. ICML 2025Delta Decompression for MoE-based LLMs Compression
143. ICML 2025Diff-MoE: Diffusion Transformer with Time-Aware and Space-Adaptive Experts
144. ICML 2025Domain2Vec: Vectorizing Datasets to Find the Optimal Data Mixture without Training
145. ICML 2025Dynamic Mixture of Curriculum LoRA Experts for Continual Multimodal Instruction Tuning
146. ICML 2025Expert Race: A Flexible Routing Strategy for Scaling Diffusion Transformer with Mixture of Experts
147. ICML 2025FlexiReID: Adaptive Mixture of Expert for Multi-Modal Person Re-Identification
148. ICML 2025FloE: On-the-Fly MoE Inference on Memory-constrained GPU
149. ICML 2025Gaussian Mixture Flow Matching Models
150. ICML 2025Graph Neural Network Generalization With Gaussian Mixture Model Based Augmentation

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。