//
sign in
Post
by @danabra.mov
PostEmbed
by @danabra.mov
Record
by @jimpick.com
Record
by @atsui.org
+ new component
Post
2/ Verifiable rewards make self-improvement possible across many reasoning tasks. But RLVR and prompt optimization can be expensive: they mostly rely on brute force guess-and-check, rather than explicitly extracting the general principles that separate success from failure.