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 1 / 8 页

1. NeurIPS 2026A Biconvex Formulation for Stable Transport of Mixture Models with a Unique Solution
2. NeurIPS 2026A Statistical Theory of Gated Attention through the Lens of Hierarchical Mixture of Experts
3. NeurIPS 2026AdERA: Adaptive Exponent Reuse for Lossless Allgather in Sharded MoE Training
4. NeurIPS 2026AlphaQ: Calibration-Free Bit Allocation for Mixture-of-Experts Quantization
5. NeurIPS 2026AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting
6. NeurIPS 2026Anatomy-Activated Mixture-of-Experts for 3D Medical Vision-Language Pre-training
7. NeurIPS 2026Beyond LoRA vs. Full Fine-Tuning: Gradient-Guided Optimizer Routing for LLM Adaptation
8. NeurIPS 2026Biomedical Acquisition-induced Style Shifts as Mixture Shifts: Style-aware Mixture-of-Experts Multimodal Prompt Learning
9. NeurIPS 2026Boundary-Induced Forgetting in Large Language Model Tool Chains: Measuring Field-Routing Failures
10. NeurIPS 2026CECAR: Cache & Expert Co-Aware Routing Accelerates On-Device Inference of MoE LLMs
11. NeurIPS 2026Class-Incremental Learning via LoRA-based Elastic Ensemble of Experts
12. NeurIPS 2026DISTMOE: Rehearsal-free Routing in Mixture-of-Experts for Distributed Instruction Tuning
13. NeurIPS 2026Distributionally Robust Mixture-of-Experts Training
14. NeurIPS 2026DivMoE: Fine-Grained MoE Upcycling via Cross-Domain Expert Composition
15. NeurIPS 2026Drift-Aware Multimodal User Representation Learning via Multi-Scale Temporal Modeling and Sparse Mixture-of-Experts
16. NeurIPS 2026Efficient Pre-Training with Token Superposition
17. NeurIPS 2026EMO: Pretraining Mixture of Experts for Emergent Modularity
18. NeurIPS 2026Evidence Guided Dual Expert Memory with Joint Routing for VLLMs Online Correction
19. NeurIPS 2026FineRMoE: Dimension Expansion for Finer-Grained Expert with an Upcycling Approach
20. NeurIPS 2026FLUX: Geometry-Aware Longitudinal Flow Matching with Mixture of Experts
21. NeurIPS 2026From Experts to Sub-experts: Fine-grained Parameter-Efficient Fine-Tuning for MoE LLMs
22. NeurIPS 2026From Facts to Personas: Interpretable Role Unlearning in LLMs via Mixture-of-Experts
23. NeurIPS 2026Gaussian Mixture Models in Hilbert Spaces via Kernel Methods
24. NeurIPS 2026Geometry is an Operator: Lie-Algebraic Space Routing for View-Robust 3D MLLMs
25. NeurIPS 2026Holistic Scaling Laws for Optimal Mixture-of-Experts Architecture Optimization
26. NeurIPS 2026How to Scale Mixture-of-Experts: From muP to the Maximally Scale-Stable Parameterization
27. NeurIPS 2026HypMoE-ReID: Hyperspherical Mixture-of-Experts for Large Scale Person Re-Identification
28. NeurIPS 2026Improving Continual Video Instance Segmentation via Spatial-Temporal Balanced Mixture-of-Experts Adapters
29. NeurIPS 2026Knowledge Localization in Mixture-of-Experts LLMs Using Cross-Lingual Inconsistency
30. NeurIPS 2026Load Balancing Mixture of Experts with Similarity Preserving Routers
31. NeurIPS 2026Managing Self-Learning Experts under Per-Round Budget Constraints
32. NeurIPS 2026Mask-Conditioned Gradient Masking for Fine-Tuning Mixture-of-Experts Diffusion Language Models
33. NeurIPS 2026MDK-MoE: Multi-view Decomposed Kalman Mixture of Experts Framework for Non-stationary Time Series Forecasting
34. NeurIPS 2026MEME: Lightweight Hierarchical Mixture-of-Experts for Unified Affective Computing
35. NeurIPS 2026Merging RLVR-Trained Experts via Policy-Shift-Guided Spectral Alignment
36. NeurIPS 2026MixScentNet: A Multiscale Graph-based Framework for Predicting Scent Mixture Perception
37. NeurIPS 2026Mixture of Attribute-Aware Attention Experts for Fine-grained E-Commerce Composed Image Retrieval
38. NeurIPS 2026Mixture of Layers: Dynamic Layer Routing for Visual Reasoning
39. NeurIPS 2026Mixture of Probes: Learning with Privileged Modalities in Multimodal LLMs Through Probing
40. NeurIPS 2026Mixture-of-Experts for Online Matrix Completion on a Drifting Union of Subspaces
41. NeurIPS 2026Mixture-of-Hierarchical Experts: Optimized Mamba Architecture for Vision Diffusion
42. NeurIPS 2026Mixture-of-Top-$k$ Attention: Efficient Attention as Scalable Fast Weights
43. NeurIPS 2026MobileMoE: Scaling On-Device Mixture of Experts
44. NeurIPS 2026MoBiQuant: Mixture-of-Bits Quantization for Token-Adaptive Any-Precision LLM
45. NeurIPS 2026MoCA: Mixture-of-Components Attention for Scalable Compositional 3D Generation
46. NeurIPS 2026Modality-Aware Expert Pruning for MoE-Based Multimodal Large Language Models
47. NeurIPS 2026MoE-SpAc: Efficient MoE Inference Based on Speculative Activation Utility in Heterogeneous Edge Scenario
48. NeurIPS 2026MoEZip: Routing-Aware KV Cache Compression for Sparse Mixture-of-Experts LLMs
49. NeurIPS 2026MoME: Mixture-of-Memory Embeddings for Context-Aware Sparse Lookup
50. NeurIPS 2026MoTo: Mixture of Tokenizers Towards Fair Multilingual Language Modeling

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。