1 paper · 1 filter
Shengwei Xu, Yuxuan Lu, Yifan Wu +2
Reference-based text evaluation metrics, which are widely used to assess natural language generation systems, score a candidate response by comparing it with a reference response.…