Record · corpus.blog/posts/01a07d5d-72e8-7140-82cd-cd927043cdfc

Faster Text Generation with Self-Speculative Decoding

Hugging Face · published 20 November 2024

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at huggingface.co Wayback capture

1blogs citing

First cited: 9 Mar 2026
Most recent: 9 Mar 2026
Rank this month: not ranked
Links from this post: 18
Words captured: 2,251

Cited by

Blog

In the post

Date

samu.space

Speculative Decoding Research Overview Read at the source ↗

9 Mar 2026

Links from this post

Link

Host

https://arxiv.org/abs/2404.16710

arxiv.org

https://huggingface.co/papers/2404.16710

huggingface.co

https://huggingface.co/collections/facebook/layerskip-666b25c50c8ae90e1965727a

huggingface.co

https://huggingface.co/datasets/ariG23498/layer-skip-assets/blob/main/early_exit_self_speculative_decoding.ipynb

huggingface.co

https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/blog/layerskip-assets/LayerSkip-Demo.gif

huggingface.co

https://huggingface.co/facebook/layerskip-llama2-7B

huggingface.co

Assisted Generation: a new direction toward low-latency text generation Read at the source ↗

huggingface.co

https://github.com/huggingface/transformers

github.com

https://github.com/gante/huggingface-demos/pull/1

github.com

https://huggingface.co/datasets/ariG23498/layer-skip-assets/blob/main/LayerSkip%20HuggingFace%20Benchmarking%20-%20summarization.csv

huggingface.co

https://huggingface.co/docs/transformers/v4.46.3/en/llm_optims

huggingface.co

https://github.com/huggingface/transformers/pull/34240

github.com

https://gist.github.com/mostafaelhoushi/1dd2781b896504bf0569a3ae4b8f9ecf

gist.github.com

https://drive.google.com/drive/folders/145CUq-P_6tbPkmArL7qsjxUihjDgLnzX?usp=sharing

drive.google.com

https://aclanthology.org/2024.acl-long.607/

aclanthology.org

https://arxiv.org/abs/2408.11049

arxiv.org

https://arxiv.org/abs/2305.10427

arxiv.org

https://arxiv.org/abs/2402.02057

arxiv.org