MOBA: Mixture of Block Attention for Long-Context LLMs (arxiv.org)

1 points by ur-whale 1mo ago ↗ HN

0 comments

[ 3.4 ms ] story [ 12.5 ms ] thread

No comments yet.

A showcase of shard-db. Data from the Hacker News API. Refreshed every 15 minutes.

Source · Live DB stats