← Search

Felix Wu*

1 accepted papers

2020

BERTScore: Evaluating Text Generation with BERT

ICLR 2020poster

We propose BERTScore, an automatic evaluation metric for text generation. Analogously to common metrics, BERTScore computes a similarity score for each token in the candidate sentence with each token in the reference sentence. However, instead of exact matches, we compute token similarity using cont…

Cited by 6747SourcecodeScholar