GLEU Without Tuning
This is an incremental improvement for researchers and practitioners in natural language processing, specifically in grammatical error correction evaluation.
The paper tackled problems with the GLEU metric for evaluating grammatical error corrections when using more reference sets, resulting in a modified version that eliminates the need for tuning.
The GLEU metric was proposed for evaluating grammatical error corrections using n-gram overlap with a set of reference sentences, as opposed to precision/recall of specific annotated errors (Napoles et al., 2015). This paper describes improvements made to the GLEU metric that address problems that arise when using an increasing number of reference sets. Unlike the originally presented metric, the modified metric does not require tuning. We recommend that this version be used instead of the original version.