Automated summary quality assessment falls into two categories: reference-based and reference-free. Reference-based metrics, historically deemed more accurate due to the additional information provided by human-written references, are limited by their reliance on human input. In this paper, we hypothesize that the comparison methodologies used by some reference-based metrics to evaluate a system summary against its corresponding reference can be effectively adapted to assess it against its source document, thereby transforming these metrics into reference-free ones. Experimental results support this hypothesis. After being repurposed reference-freely, the zero-shot BERTScore using the pretrained DeBERTa-large-MNLI model of <0.5B parameters consistently outperforms its original reference-based version across various aspects on the SummEval and Newsroom datasets. It also excels in comparison to most existing reference-free metrics and closely competes with zero-shot summary evaluators based on GPT-3.5.

Canonical automatic summary evaluation metrics, such as ROUGE, focus on lexical similarity which cannot well capture semantics nor linguistic quality and require a reference summary which is costly to obtain. Recently, there have been a growing number of efforts to alleviate either or both of the two drawbacks. In this paper, we present a proof-of-concept study to a weakly supervised summary evaluation approach without the presence of reference summaries. Massive data in existing summarization datasets are transformed for training by pairing documents with corrupted reference summaries. In cross-domain tests, our strategy outperforms baselines with promising improvements, and show a great advantage in gauging linguistic qualities over all metrics.

Evaluating machine-generated summaries without a human-written reference summary has been a need for a long time. Inspired by preference labeling in existing work of summarization evaluation, we propose to judge summary quality by learning the preference rank of summaries using the Bradley-Terry power ranking model from inferior summaries generated by corrupting base summaries. Extensive experiments on several datasets show that our weakly supervised scheme can produce scores highly correlated with human ratings.