[NIPS 2016] Learning Structured Sparsity in Deep Neural Networks
Similarity-based block and sublayer dropping for ViT, DeiT, DINOv2, DINOv3 and SwinV2, evaluated on 13 datasets, with speed and energy benchmarks, statistical tests and automated model selection.
Official Pytorch implementation of "Filter Pruning by Image Channel Reduction in Pre-Trained Convolutional Neural Networks".
Intelligent layer pruning toolkit for LLMs featuring iterative optimization, self-healing algorithms, and comprehensive benchmarking.
TALE: training-free task-aware layer pruning for improving LLM accuracy and reducing inference cost
To associate your repository with the layer-pruning topic, visit your repo's landing page and select "manage topics."