most citedReinforcement Learning for Few-Shot Text Generation Adaptation

6 citations · 6 across the 4 of their papers we have counts for

collaborators
Showing cs.ITShow all

1 paper · 1 filter

cs.IT2026

Rarity Determines Minimax Rates of Semantic Entropy From Weak Pairwise Judgments With a Known Channel

Xiaoli Yu, Jiamiao Liu

The semantic entropy of a language model is the entropy of the meanings of its sampled answers. The meanings are never observed, and the entropy is estimated from the verdicts of a…