mlsanityData to 7 Oct 2026

SkipNet: Learning Dynamic Routing in Convolutional Networks

arXiv 1711.09485 ↗First posted 26 Nov 2017cs.CVcited by 328

Not on the map. Why?

Cited by 24 in Efficient LLM inference and serving, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

While deeper convolutional networks are needed to achieve maximum accuracy in visual perception tasks, for many inputs shallower networks are sufficient. We exploit this observation by learning to skip convolutional layers on a per-input basis. We introduce SkipNet, a modified residual network, that uses a gating network to selectively skip convolutional blocks based on the activations of the previous layer. We formulate the dynamic skipping problem in the context of sequential decision making and propose a hybrid learning algorithm that combines supervised learning and reinforcement learning to address the challenges of non-differentiable skipping decisions. We show SkipNet reduces computation by 30-90% while preserving the accuracy of the original model on four benchmark datasets and outperforms the state-of-the-art dynamic networks and static compression methods. We also qualitatively evaluate the gating policy to reveal a relationship between image scale and saliency and the number of layers skipped.

Key facts. Cited by 328 papers in the map (data through 7 Oct 2026). First posted on arXiv 26 Nov 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2018: 1; 2019: 0; 2020: 0; 2021: 70; 2022: 64; 2023: 59; 2024: 39; 2025: 47; 2026: 48.

328 papers.

Sort:

By area:

  1. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,054
  2. 2021Dynamic Neural Networks: A Surveycited by 405
  3. 2021AdaViT: Adaptive Tokens for Efficient Vision Transformercited by 360
  4. 2022Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizationscited by 359
  5. 2022Confident Adaptive Language Modelingcited by 297
  6. 2024Mixture-of-Depths: Dynamically allocating compute in transformer-based language modelscited by 242
  7. 2021AdaViT: Adaptive Vision Transformers for Efficient Image Recognitioncited by 228
  8. 2021Not All Images are Worth 16x16 Words: Dynamic Transformers for Efficient Image Recognitioncited by 143
  9. 2021IA-RED2: Interpretability-Aware Redundancy Reduction for Vision Transformerscited by 127
  10. 2021DSelect-k: Differentiable Selection in the Mixture of Experts with Applications to Multi-Task Learningcited by 114
  11. 2021Learning Task-Oriented Communication for Edge Inference: An Information Bottleneck Approachcited by 105
  12. 2024Not All Layers of LLMs Are Necessary During Inferencecited by 95
38 more
  1. 2021MOOD: Multi-level Out-of-distribution Detectioncited by 94
  2. 2021Lite-HRNet: A Lightweight High-Resolution Networkcited by 91
  3. 2021SUTD-TrafficQA: A Question Answering Benchmark and an Efficient Network for Video Reasoning over Traffic Eventscited by 84
  4. 2021Dynamic Slimmable Networkcited by 83
  5. 2021Spatially-Adaptive Image Restoration using Distortion-Guided Networkscited by 77
  6. 2021Adaptive Inference through Early-Exit Networks: Design, Challenges and Directionscited by 74
  7. 2021Consistent Accelerated Inference via Confident Adaptive Transformerscited by 73
  8. 2024Dynamic Diffusion Transformercited by 60
  9. 2021FrameExit: Conditional Early Exiting for Efficient Video Recognitioncited by 59
  10. 2022Dynamic Multimodal Fusioncited by 54
  11. 2023A Dynamic Multi-Scale Voxel Flow Network for Video Predictioncited by 53
  12. 2021A Survey on Green Deep Learningcited by 52
  13. 2023Adaptive Rotated Convolution for Rotated Object Detectioncited by 51
  14. 2021AdaFuse: Adaptive Temporal Fusion Network for Efficient Action Recognitioncited by 48
  15. 2022NICGSlowDown: Evaluating the Efficiency Robustness of Neural Image Caption Generation Modelscited by 47
  16. 2021RegNet: Self-Regulated Network for Image Classificationcited by 44
  17. 2022Learning to Weight Samples for Dynamic Early-exiting Networkscited by 44
  18. 2024AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruningcited by 43
  19. 2022Glance and Focus Networks for Dynamic Visual Recognitioncited by 41
  20. 2021Zero Time Waste: Recycling Predictions in Early Exit Neural Networkscited by 39
  21. 2023SEENN: Towards Temporal Spiking Early-Exit Neural Networkscited by 39
  22. 2021Edge-Cloud Polarization and Collaboration: A Comprehensive Survey for AIcited by 38
  23. 2021Teachers Do More Than Teach: Compressing Image-to-Image Modelscited by 37
  24. 2021AdaMML: Adaptive Multi-Modal Learning for Efficient Video Recognitioncited by 36
  25. 2021Dynamic Network Quantization for Efficient Video Inferencecited by 36
  26. 2021TAda! Temporally-Adaptive Convolutions for Video Understandingcited by 36
  27. 2022Efficient Split-Mix Federated Learning for On-Demand and In-Situ Customizationcited by 34
  28. 2022Path Independent Equilibrium Models Can Better Exploit Test-Time Computationcited by 33
  29. 2023Dynamic Perceiver for Efficient Visual Recognitioncited by 33
  30. 2022Enable Deep Learning on Mobile Devices: Methods, Systems, and Applicationscited by 32
  31. 2024SpaceByte: Towards Deleting Tokenization from Large Language Modelingcited by 31
  32. 2025Learning Unmasking Policies for Diffusion Language Modelscited by 29
  33. 2021Improving Transformers with Probabilistic Attention Keyscited by 28
  34. 2021Single-Layer Vision Transformers for More Accurate Early Exits with Less Overheadcited by 26
  35. 2021MicroNet: Improving Image Recognition with Extremely Low FLOPscited by 26
  36. 2021Deep Joint Source-Channel Coding for Wireless Image Transmission with Adaptive Rate Controlcited by 26
  37. 2021Task2Sim : Towards Effective Pre-training and Transfer from Synthetic Datacited by 26
  38. 2023Latency-aware Unified Dynamic Networks for Efficient Image Recognitioncited by 26

50 most cited of 328 All 328 (JSON)