REFeREE: A REference-FREE Model-Based Metric for Text Simplification
This addresses the problem of costly and limited reference data for text simplification evaluation, offering a more scalable solution for researchers and practitioners, though it is incremental as it builds on existing model-based approaches.
The paper tackles the lack of universal quality standards and scarce annotated references in text simplification by proposing REFeREE, a reference-free model-based metric with a 3-stage curriculum, which outperforms existing reference-based metrics in predicting overall ratings and achieves competitive performance for specific ratings without needing reference simplifications at inference.
Text simplification lacks a universal standard of quality, and annotated reference simplifications are scarce and costly. We propose to alleviate such limitations by introducing REFeREE, a reference-free model-based metric with a 3-stage curriculum. REFeREE leverages an arbitrarily scalable pretraining stage and can be applied to any quality standard as long as a small number of human annotations are available. Our experiments show that our metric outperforms existing reference-based metrics in predicting overall ratings and reaches competitive and consistent performance in predicting specific ratings while requiring no reference simplifications at inference time.