sscardapane.bsky.social
I fall in love with a new #machinelearning topic every month 🙄 Ass. Prof. Sapienza (Rome) | Author: Alice in a differentiable wonderland (https://www.sscardapane.it/alice-book/)
👤 Joined on 20 October 2023
Followers
Following
Total Posts
@sscardapane.bsky.socialThanks a lot to all my amazing co-authors @alessiodevoto.bsky.social @sscardapane.bsky.social @yuzhaouoe.bsky.social @neuralnoise.com Eric de la Clergerie @bensagot.bsky.social And a special thanks to @edoardo-ponti.bsky.social for the academic visit that made this work possible!
Show PostWill present this at #CVPR ✈️ See you in Nashville 🇺🇸! Kudos to the team 👏 Antonio A. Gargiulo, @mariasofiab.bsky.social, @sscardapane.bsky.social, Fabrizio Silvestri, Emanuele Rodolà.
Show PostPlease share it within your circles! edin.ac/3DDQK1o
Show Post🚀 New Paper Alert! 🚀 We introduce Q-Filters, a training-free method for efficient KV Cache compression! It is compatible with FlashAttention and can compress along generation which is particularly useful for reasoning models ⚡ TLDR: we make Streaming-LLM smarter using the geometry of attention
Show PostQ-Filters is very efficient which allows streaming compression at virtually no latency cost, just like Streaming-LLM... ...but it is also much better at retaining relevant KV pairs compared to fast alternatives (and can even beat slower algorithms such as SnapKV)
Show Post