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 4 / 5 页

151. NeurIPS 2024A Comprehensive Analysis on the Learning Curve in Kernel Ridge Regression
152. NeurIPS 2024A generalized neural tangent kernel for surrogate gradient learning
153. NeurIPS 2024A Kernel Perspective on Distillation-based Collaborative Learning
154. NeurIPS 2024ACFun: Abstract-Concrete Fusion Facial Stylization
155. NeurIPS 2024Efficient Convex Algorithms for Universal Kernel Learning
156. NeurIPS 2024Error Correction Output Codes for Robust Neural Networks against Weight-errors: A Neural Tangent Kernel Point of View
157. NeurIPS 2024Fair Kernel K-Means: from Single Kernel to Multiple Kernel
158. NeurIPS 2024Faster Neighborhood Attention: Reducing the O(n^2) Cost of Self Attention at the Threadblock Level
159. NeurIPS 2024Get rich quick: exact solutions reveal how unbalanced initializations promote rapid feature learning
160. NeurIPS 2024Inverse M-Kernels for Linear Universal Approximators of Non-Negative Functions
161. NeurIPS 2024Learning to Embed Distributions via Maximum Kernel Entropy
162. NeurIPS 2024Nonstationary Sparse Spectral Permanental Process
163. NeurIPS 2024On the Saturation Effects of Spectral Algorithms in Large Dimensions
164. NeurIPS 2024On the Target-kernel Alignment: a Unified Analysis with Kernel Complexity
165. NeurIPS 2024Reverse Transition Kernel: A Flexible Framework to Accelerate Diffusion Inference
166. NeurIPS 2024Self-Distilled Depth Refinement with Noisy Poisson Fusion
167. NeurIPS 2024Stochastic Kernel Regularisation Improves Generalisation in Deep Kernel Machines
168. NeurIPS 2024Temporal Graph Neural Tangent Kernel with Graphon-Guaranteed
169. NeurIPS 2024Twin-Merging: Dynamic Integration of Modular Expertise in Model Merging
170. NeurIPS 2024Unveiling the Hidden Structure of Self-Attention via Kernel Principal Component Analysis
171. CVPR 2023Compacting Binary Neural Networks by Sparse Kernel Selection
172. CVPR 2023Efficient Multimodal Fusion via Interactive Prompting
173. CVPR 2023Federated Learning With Data-Agnostic Distribution Fusion
174. CVPR 2023K3DN: Disparity-Aware Kernel Estimation for Dual-Pixel Defocus Deblurring
175. CVPR 2023Kernel Aware Resampler
176. CVPR 2023LinK: Linear Kernel for LiDAR-Based 3D Perception
177. CVPR 2023Neural Kernel Surface Reconstruction
178. CVPR 2023Structured Kernel Estimation for Photon-Limited Deconvolution
179. ICLR 2023A Kernel Perspective of Skip Connections in Convolutional Networks
180. ICLR 2023Analyzing Tree Architectures in Ensembles via Neural Tangent Kernel
181. ICLR 2023Characterizing the spectrum of the NTK via a power series expansion
182. ICLR 2023Classically Approximating Variational Quantum Machine Learning with Random Fourier Features
183. ICLR 2023Kernel Neural Optimal Transport
184. ICLR 2023On The Relative Error of Random Fourier Features for Preserving Kernel Distance
185. ICLR 2023On the Saturation Effect of Kernel Ridge Regression
186. ICLR 2023Random Laplacian Features for Learning with Hyperbolic Space
187. ICLR 2023Reliability of CKA as a Similarity Measure in Deep Learning
188. ICLR 2023Scale-invariant Bayesian Neural Networks with Connectivity Tangent Kernel
189. ICLR 2023Self-supervised learning with rotation-invariant kernels
190. ICLR 2023Subquadratic Algorithms for Kernel Matrices via Kernel Density Estimation
191. ICLR 2023Symmetric Pruning in Quantum Neural Networks
192. ICLR 2023Trainability Preserving Neural Pruning
193. ICLR 2023Tuning Frequency Bias in Neural Network Training with Nonuniform Data
194. ICML 2023A Kernel Stein Test of Goodness of Fit for Sequential Models
195. ICML 2023Consistency of Multiple Kernel Clustering
196. ICML 2023Delayed Feedback in Kernel Bandits
197. ICML 2023Estimation Beyond Data Reweighting: Kernel Method of Moments
198. ICML 2023Fast Private Kernel Density Estimation via Locality Sensitive Quantization
199. ICML 2023Gradient Descent in Neural Networks as Sequential Learning in Reproducing Kernel Banach Space
200. ICML 2023Integrating Prior Knowledge in Contrastive Learning with Kernel

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。