mlsanityData to 7 Oct 2026

CondConv: Conditionally Parameterized Convolutions for Efficient Inference

arXiv 1904.04971 ↗First posted 10 Apr 2019cs.CV cs.AI cs.LGcited by 273

Not on the map. Why?

Cited by 28 in Mixture-of-experts, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Convolutional layers are one of the basic building blocks of modern deep neural networks. One fundamental assumption is that convolutional kernels should be shared for all examples in a dataset. We propose conditionally parameterized convolutions (CondConv), which learn specialized convolutional kernels for each example. Replacing normal convolutions with CondConv enables us to increase the size and capacity of a network, while maintaining efficient inference. We demonstrate that scaling networks with CondConv improves the performance and inference cost trade-off of several existing convolutional neural network architectures on both classification and detection tasks. On ImageNet classification, our CondConv approach applied to EfficientNet-B0 achieves state-of-the-art performance of 78.3% accuracy with only 413M multiply-adds. Code and checkpoints for the CondConv Tensorflow layer and CondConv-EfficientNet models are available at: https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet/condconv.

Key facts. Cited by 273 papers in the map (data through 7 Oct 2026). First posted on arXiv 10 Apr 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2020: 1; 2021: 62; 2022: 57; 2023: 64; 2024: 36; 2025: 32; 2026: 21.

273 papers.

Sort:

By area:

  1. 2020GShard: Scaling Giant Models with Conditional Computation and Automatic Shardingcited by 1,601
  2. 2021Domain Generalization: A Surveycited by 760
  3. 2021Scaling Vision with Sparse Mixture of Expertscited by 740
  4. 2021Conditional DETR for Fast Training Convergencecited by 422
  5. 2021Dynamic Neural Networks: A Surveycited by 405
  6. 2021Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognitioncited by 317
  7. 2023From Sparse to Soft Mixtures of Expertscited by 245
  8. 2021Mobile-Former: Bridging MobileNet and Transformercited by 234
  9. 2021Attention Mechanisms in Computer Vision: A Surveycited by 232
  10. 2021PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Cloudscited by 211
  11. 2022Focal Sparse Convolutional Networks for 3D Object Detectioncited by 142
  12. 2022EdgeViTs: Competing Light-weight CNNs on Mobile Devices with Vision Transformerscited by 116
38 more
  1. 2021Involution: Inverting the Inherence of Convolution for Visual Recognitioncited by 99
  2. 2023Large Selective Kernel Network for Remote Sensing Object Detectioncited by 98
  3. 2022M3ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-designcited by 97
  4. 2021Beyond Distillation: Task-level Mixture-of-Experts for Efficient Inferencecited by 96
  5. 2022Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolutioncited by 96
  6. 2021CondLaneNet: a Top-to-down Lane Detection Framework Based on Conditional Convolutioncited by 95
  7. 2023StyleTalk: One-shot Talking Head Generation with Controllable Speaking Stylescited by 92
  8. 2021Lite-HRNet: A Lightweight High-Resolution Networkcited by 91
  9. 2023Soft Merging of Experts with Adaptive Routingcited by 88
  10. 2021Diverse Branch Block: Building a Convolution as an Inception-like Unitcited by 87
  11. 2021Dynamic Slimmable Networkcited by 83
  12. 2024Mixtures of Experts Unlock Parameter Scaling for Deep RLcited by 75
  13. 2022Dynamic Spatial Propagation Network for Depth Completioncited by 70
  14. 2024Scaling Diffusion Transformers to 16 Billion Parameterscited by 65
  15. 2022Towards Spatio-Temporal Aware Traffic Time Series Forecasting--Full Versioncited by 59
  16. 2023Sparse MoE as the New Dropout: Scaling Dense and Self-Slimmable Transformerscited by 59
  17. 2022Uni-Perceiver-MoE: Learning Sparse Generalist Models with Conditional MoEscited by 58
  18. 2022Omni-Dimensional Dynamic Convolutioncited by 55
  19. 2022Dynamic Multimodal Fusioncited by 54
  20. 2021Decoupled Dynamic Filter Networkscited by 52
  21. 2021Conditional Directed Graph Convolution for 3D Human Pose Estimationcited by 52
  22. 2023Adaptive Rotated Convolution for Rotated Object Detectioncited by 51
  23. 2023Speck: A Smart event-based Vision Sensor with a low latency 327K Neuron Convolutional Neuronal Network Processing Pipelinecited by 50
  24. 2021AdaFuse: Adaptive Temporal Fusion Network for Efficient Action Recognitioncited by 48
  25. 2021Image Synthesis via Semantic Compositioncited by 45
  26. 2024LSKNet: A Foundation Lightweight Backbone for Remote Sensingcited by 44
  27. 2024OA-CNNs: Omni-Adaptive Sparse CNNs for 3D Semantic Segmentationcited by 43
  28. 2022Residual Mixture of Expertscited by 39
  29. 2023SEENN: Towards Temporal Spiking Early-Exit Neural Networkscited by 39
  30. 2022The Neural Race Reduction: Dynamics of Abstraction in Gated Networkscited by 37
  31. 2023HyperE2VID: Improving Event-Based Video Reconstruction via Hypernetworkscited by 37
  32. 2021TAda! Temporally-Adaptive Convolutions for Video Understandingcited by 36
  33. 2023PDPP: Projected Diffusion for Procedure Planning in Instructional Videoscited by 36
  34. 2023Merging Experts into One: Improving Computational Efficiency of Mixture of Expertscited by 36
  35. 2021Revisiting Dynamic Convolution via Matrix Decompositioncited by 35
  36. 2022DUET: A Tuning-Free Device-Cloud Collaborative Parameters Generation Framework for Efficient Device Model Generalizationcited by 35
  37. 2023Patch-level Routing in Mixture-of-Experts is Provably Sample-efficient for Convolutional Neural Networkscited by 35
  38. 2023Language Adaptive Weight Generation for Multi-task Visual Groundingcited by 35

50 most cited of 273 All 273 (JSON)