Tldr: When learning over prompts, classic PAC-Bayes bounds are extremely tight; more so than any previous PAC-Bayes analysis of deep networks.
Understanding prompt engineering may not require rethinking generalisation
Understanding prompt engineering may not…
Understanding prompt engineering may not require rethinking generalisation
Tldr: When learning over prompts, classic PAC-Bayes bounds are extremely tight; more so than any previous PAC-Bayes analysis of deep networks.