at://
/
app.bsky.feed.post
/
3mlr36wch4b2a
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
PostEmbed
Current LLM safety alignment techniques improve model robustness against adversarial attacks, but overlook whether and how LLMs can recover helpfulness when benign users clarify their intent. We...
arxiv.org
Useless but Safe? Benchmarking Utility Recovery with User Intent...
