Rethinking and Refining the Distinct Metric
This work addresses a specific issue in NLP evaluation metrics, offering an incremental improvement for researchers and practitioners in language generation tasks.
The paper tackled biases in the widely used Distinct-n metric for evaluating language generation diversity, which unfairly penalized longer sequences, and introduced the Expectation-Adjusted Distinct (EAD) metric that better correlates with human judgment.
Distinct-$n$ score\cite{Li2016} is a widely used automatic metric for evaluating diversity in language generation tasks. However, we observed that the original approach for calculating distinct scores has evident biases that tend to assign higher penalties to longer sequences. We refine the calculation of distinct scores by scaling the number of distinct tokens based on their expectations. We provide both empirical and theoretical evidence to show that our method effectively removes the biases existing in the original distinct score. Our experiments show that our proposed metric, \textit{Expectation-Adjusted Distinct (EAD)}, correlates better with human judgment in evaluating response diversity. To foster future research, we provide an example implementation at \url{https://github.com/lsy641/Expectation-Adjusted-Distinct}.