No exact results. Similar results found.
@ChinafakeWiki@mastodon.social2024-11-26 21:14:12
Our Matrix chat is now back! It was gone due to some moderation issues, but those have been resolved and it is now back online with Discord messages being bridged again! Check it out at https://matrix.to/#/#chinafakes:matrix.org
@theawely@mamot.fr2024-12-13 18:48:37
Excited about the new xLSTM model release. There are many well-though designs compared to transformers: recurrence (which should allows composability), gating (like Mamba & LSTM which is based on, which allows time complexity independent of the input size), state tracking (unlike Mamba & transformers). For now, these advantage aren’t apparent on benchmarks, but most training techniques are secrets, and the recent advances of LLMs evidenced that they matter a lot.