at://
/
app.bsky.feed.post
/
3mjzslixwb22v
sign in
All
4
Record
2
Post
1
PostEmbed
1
Post
by @danabra.mov
PostEmbed
by @danabra.mov
Record
by @jimpick.com
Record
by @atsui.org
+ new component
Post
Yet, the above benchmarks only ask whether an alternate state is observed among hundreds of samples. Can we prompt the model to consistently predict states that are rarely sampled? A "conformation prompt”? 🧵 5/8
1mo
Minji Lee
