MaskEval: Weighted MLM-Based Evaluation for Text Summarization and Simplification
This addresses the need for adaptable evaluation metrics in NLP for researchers and practitioners, though it is incremental as it builds on existing MLM techniques.
The paper tackles the challenge of evaluating text summarization and simplification outputs across multiple dimensions without references by introducing MaskEval, a reference-less metric using masked language modeling with a weighting mechanism, achieving strong correlations with human judgments in English tasks.
In text summarization and simplification, system outputs must be evaluated along multiple dimensions such as relevance, factual consistency, fluency, and grammaticality, and a wide range of possible outputs could be of high quality. These properties make the development of an adaptable, reference-less evaluation metric both necessary and challenging. We introduce MaskEval, a reference-less metric for text summarization and simplification that operates by performing masked language modeling (MLM) on the concatenation of the candidate and the source texts. It features an attention-like weighting mechanism to modulate the relative importance of each MLM step, which crucially allows it to be adapted to evaluate different quality dimensions. We demonstrate its effectiveness on English summarization and simplification in terms of correlations with human judgments, and explore transfer scenarios between the two tasks.