Mastodon discussion Discussions 7h ago

This is a 3-paper arc on whether LLMs can reliably self-correct their own reasoning. Self-Refine proposes a naive intrin...

by Benjamin Han

This is a 3-paper arc on whether LLMs can reliably self-correct their own reasoning. Self-Refine proposes a naive intrinsic-feedback loop and reports impressive gains. Cannot-Self-Correct refutes empirically the class of approach Self-Refine belongs to. Reflexion threads the needle by gating self-correction on a reliable external signal.#LLMs #AI #Reasoning #Metacognition

Read Original

Metadata

Reblogs Count
1
Replies Count
1
Account
BenjaminHan@sigmoid.social

Related