Improving Explainability of Sentence-level Metrics via Edit-level Attribution for Grammatical Error Correction
This improves explainability for GEC researchers and users, though it is incremental as it applies an existing attribution method to a specific domain.
The paper tackled the lack of explainability in sentence-level metrics for Grammatical Error Correction (GEC) by attributing scores to individual edits using Shapley values, achieving about 70% alignment with human evaluations and revealing metric biases like ignoring orthographic edits.
Various evaluation metrics have been proposed for Grammatical Error Correction (GEC), but many, particularly reference-free metrics, lack explainability. This lack of explainability hinders researchers from analyzing the strengths and weaknesses of GEC models and limits the ability to provide detailed feedback for users. To address this issue, we propose attributing sentence-level scores to individual edits, providing insight into how specific corrections contribute to the overall performance. For the attribution method, we use Shapley values, from cooperative game theory, to compute the contribution of each edit. Experiments with existing sentence-level metrics demonstrate high consistency across different edit granularities and show approximately 70\% alignment with human evaluations. In addition, we analyze biases in the metrics based on the attribution results, revealing trends such as the tendency to ignore orthographic edits. Our implementation is available at \url{https://github.com/naist-nlp/gec-attribute}.