Sign up to save your podcasts
Or
Share [short] HiRE: High Recall Approximate Top-k Estimation for Efficient LLM Inference
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
[short] HiRE: High Recall Approximate Top-k Estimation for Efficient LLM Inference
HiRE introduces a method to reduce memory-bound autoregressive decoding with Large Language Models by efficiently predicting and computing top rows/columns, improving latency on accelerators.
https://arxiv.org/abs//2402.09360
YouTube: https://www.youtube.com/@ArxivPapers
TikTok: https://www.tiktok.com/@arxiv_papers
Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016
Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers
View all episodes
By Igor Melnyk
5
33 ratings
HiRE introduces a method to reduce memory-bound autoregressive decoding with Large Language Models by efficiently predicting and computing top rows/columns, improving latency on accelerators.
https://arxiv.org/abs//2402.09360
YouTube: https://www.youtube.com/@ArxivPapers
TikTok: https://www.tiktok.com/@arxiv_papers
Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016
Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers
More shows like Arxiv Papers
View all
Exchanges
957 Listeners
Odd Lots
1,956 Listeners
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
438 Listeners
The Daily
112,401 Listeners
All-In with Chamath, Jason, Sacks & Friedberg
10,015 Listeners
Hard Fork
5,522 Listeners
UnHerd with Freddie Sayers
210 Listeners
Unsupervised Learning with Jacob Effron
51 Listeners
Latent Space: The AI Engineer Podcast
93 Listeners
BG2Pod with Brad Gerstner and Bill Gurley
471 Listeners