DatBench: Discriminative, faithful, and efficient VLM evaluations (arxiv.org)

18 points by circuithunter 6mo ago ↗ HN

0 comments

[ 2.8 ms ] story [ 7.0 ms ] thread

No comments yet.

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

Source · Live DB stats