papers AI Learner
The Github is limit! Click to go to the new site.

BERTScore: Evaluating Text Generation with BERT

2019-04-21
Tianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger, Yoav Artzi

Abstract

We propose BERTScore, an automatic evaluation metric for text generation. Analogous to common metrics, \method computes a similarity score for each token in the candidate sentence with each token in the reference. However, instead of looking for exact matches, we compute similarity using contextualized BERT embeddings. We evaluate on several machine translation and image captioning benchmarks, and show that BERTScore correlates better with human judgments than existing metrics, often significantly outperforming even task-specific supervised metrics.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1904.09675

PDF

http://arxiv.org/pdf/1904.09675


Similar Posts

Comments