算子融合论文趋势 2023–2026

主要结论

• 算子融合论文从 2023 年的 51 篇降到 2026 年的 48 篇，热度在回落。
• NeurIPS 是发算子融合论文最多的会议，2023–2026 年共 73 篇，占四个会议总量的 33%。
• 交叉方向里涨得最快的是传感融合：同时属于两个主题的论文从 2023 年的 2 篇增至 2026 年的 12 篇。
• 在「硬件与编译」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的端侧推理是 4.0 倍，算子融合是 0.9 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 2026Beyond Adjacent Layers: Graph-Guided Layer Fusion for Compressing Large Language Models
• ICML 2026Asymmetric conformal prediction with penalized kernel sum-of-squares
• ICLR 2026Probabilistic Kernel Function for Fast Angle Testing
• CVPR 2026Bridging RGB and Hematoxylin Components An Interleaved Guidance and Fusion
• NeurIPS 2025Beyond Benign Overfitting in Nadaraya-Watson Interpolators
• ICML 2025Equivariant Neural Tangent Kernels
• ICLR 2025DLEFT-MKC: Dynamic Late Fusion Multiple Kernel Clustering with Robust Tensor Learning via Min-Max Optimization
• CVPR 2025Parameterized Blur Kernel Prior Learning for Local Motion Deblurring
• NeurIPS 2024Efficient Convex Algorithms for Universal Kernel Learning
• ICML 2024Characterizing Overfitting in Kernel Ridgeless Regression Through the Eigenspectrum
• ICLR 2024Convolutional Deep Kernel Machines
• CVPR 2024Bi-SSC: Geometric-Semantic Bidirectional Fusion for Camera-based 3D Semantic Scene Completion
• NeurIPS 2023CosNet: A Generalized Spectral Kernel Network
• ICML 2023A Kernel Stein Test of Goodness of Fit for Sequential Models
• ICLR 2023A Kernel Perspective of Skip Connections in Convolutional Networks
• CVPR 2023Compacting Binary Neural Networks by Sparse Kernel Selection
• NeurIPS 2026Dynamical Adapter Fusion: Constructing A Global Adapter for Pre-Trained Model-based Class-Incremental Learning
• ICML 2026Even Faster Kernel Matrix Linear Algebra via Density Estimation
• ICLR 2026LaplacianFormer:Rethinking Linear Attention with Laplacian Kernel
• CVPR 2026CryoKRAQEN Kernel-Regularized Annealing for Quantized Embedding Networks in Cryo-EM Heterogeneous

全部算子融合论文

共 221 篇，第 3 / 5 页

101. NeurIPS 2025Scalable Signature Kernel Computations via Local Neumann Series Expansions
102. NeurIPS 2025SD-KDE: Score-Debiased Kernel Density Estimation
103. NeurIPS 2025Uncertainty Quantification with the Empirical Neural Tangent Kernel
104. NeurIPS 2025When Kernels Multiply, Clusters Unify: Fusing Embeddings with the Kronecker Product
105. CVPR 2024Bi-SSC: Geometric-Semantic Bidirectional Fusion for Camera-based 3D Semantic Scene Completion
106. CVPR 2024Data-Efficient Multimodal Fusion on a Single GPU
107. CVPR 2024Embracing Unimodal Aleatoric Uncertainty for Robust Multimodal Fusion
108. CVPR 2024KPConvX: Modernizing Kernel Point Convolution with Kernel Attention
109. CVPR 2024PeLK: Parameter-efficient Large Kernel ConvNets with Peripheral Convolution
110. CVPR 2024Poly Kernel Inception Network for Remote Sensing Detection
111. CVPR 2024Robust Depth Enhancement via Polarization Prompt Fusion Tuning
112. CVPR 2024SecondPose: SE(3)-Consistent Dual-Stream Feature Fusion for Category-Level Pose Estimation
113. CVPR 2024SFOD: Spiking Fusion Object Detector
114. ICLR 2024An Agnostic View on the Cost of Overfitting in (Kernel) Ridge Regression
115. ICLR 2024Convolutional Deep Kernel Machines
116. ICLR 2024Counterfactual Density Estimation using Kernel Stein Discrepancies
117. ICLR 2024Explaining Kernel Clustering via Decision Trees
118. ICLR 2024Fusing Models with Complementary Expertise
119. ICLR 2024lpNTK: Better Generalisation with Less Data via Sample Interaction During Learning
120. ICLR 2024Posterior Sampling Based on Gradient Flows of the MMD with Negative Distance Kernel
121. ICLR 2024Scalable Neural Network Kernels
122. ICLR 2024Smooth ECE: Principled Reliability Diagrams via Kernel Smoothing
123. ICLR 2024The optimality of kernel classifiers in Sobolev space
124. ICLR 2024Towards Meta-Pruning via Optimal Transport
125. ICLR 2024Transformer Fusion with Optimal Transport
126. ICML 2024Amortized Variational Deep Kernel Learning
127. ICML 2024Bring Your Own (Non-Robust) Algorithm to Solve Robust MDPs by Estimating The Worst Kernel
128. ICML 2024Characterizing Overfitting in Kernel Ridgeless Regression Through the Eigenspectrum
129. ICML 2024Copula-Nested Spectral Kernel Network
130. ICML 2024Deep Fusion: Efficient Network Training via Pre-trained Initializations
131. ICML 2024DeepPolar: Inventing Nonlinear Large-Kernel Polar Codes via Deep Learning
132. ICML 2024Dissecting Multimodality in VideoQA Transformer Models by Impairing Modality Fusion
133. ICML 2024Generalization in Kernel Regression Under Realistic Assumptions
134. ICML 2024High-Dimensional Kernel Methods under Covariate Shift: Data-Dependent Implicit Regularization
135. ICML 2024Is Kernel Prediction More Powerful than Gating in Convolutional Neural Networks?
136. ICML 2024Kernel Debiased Plug-in Estimation: Simultaneous, Automated Debiasing without Influence Functions for Many Target Parameters
137. ICML 2024Kernel Semi-Implicit Variational Inference
138. ICML 2024Kernel-Based Evaluation of Conditional Biological Sequence Models
139. ICML 2024Mean-field Analysis on Two-layer Neural Networks from a Kernel Perspective
140. ICML 2024Neural-Kernel Conditional Mean Embeddings
141. ICML 2024On the Consistency of Kernel Methods with Dependent Observations
142. ICML 2024Optimal Kernel Quantile Learning with Random Features
143. ICML 2024Position: Opportunities Exist for Machine Learning in Magnetic Fusion Energy
144. ICML 2024Predictive Dynamic Fusion
145. ICML 2024Quasi-Monte Carlo Features for Kernel Approximation
146. ICML 2024Revealing the Dark Secrets of Extremely Large Kernel ConvNets on Robustness
147. ICML 2024Reward-Free Kernel-Based Reinforcement Learning
148. ICML 2024Sampling in Unit Time with Kernel Fisher-Rao Flow
149. ICML 2024Scalable Multiple Kernel Clustering: Learning Clustering Structure from Expectation
150. ICML 2024Sequential Kernel Goodness-of-fit Testing

常见问题

2023–2026 年四大会议一共有多少篇算子融合论文？

共 221 篇，按年份分别是 2023 年 51 篇、2024 年 66 篇、2025 年 56 篇、2026 年 48 篇。

NeurIPS 2026 有多少篇算子融合论文？

NeurIPS 2026 有 14 篇算子融合论文，上一届是 25 篇。

ICLR 2026 有多少篇算子融合论文？

ICLR 2026 有 9 篇算子融合论文，上一届是 8 篇。

ICML 2026 有多少篇算子融合论文？

ICML 2026 有 13 篇算子融合论文，上一届是 14 篇。

CVPR 2026 有多少篇算子融合论文？

CVPR 2026 有 12 篇算子融合论文，上一届是 9 篇。

哪个会议的算子融合论文最多？

NeurIPS，2023–2026 年共 73 篇。