Seonglae ChoLayerSkip - a facebook Collection
Models continually pretrained using LayerSkip - https://arxiv.org/abs/2404.16710
https://huggingface.co/collections/facebook/layerskip-666b25c50c8ae90e1965727a
Paper page - LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding
Join the discussion on this paper page
https://huggingface.co/papers/2404.16710
TR-BERT: Dynamic Token Reduction for Accelerating BERT Inference
arxiv.org
https://arxiv.org/pdf/2105.11618