Better Automatic Evaluation of Open-Domain Dialogue Systems with Contextualized Embeddings
This work addresses the challenge of evaluating dialogue systems more accurately for researchers and developers, though it is incremental as it builds on prior metrics.
The paper tackled the problem of automatic evaluation for open-domain dialogue systems by proposing metrics using contextualized word embeddings to compute relatedness scores, resulting in improved performance over existing methods like RUBER.
Despite advances in open-domain dialogue systems, automatic evaluation of such systems is still a challenging problem. Traditional reference-based metrics such as BLEU are ineffective because there could be many valid responses for a given context that share no common words with reference responses. A recent work proposed Referenced metric and Unreferenced metric Blended Evaluation Routine (RUBER) to combine a learning-based metric, which predicts relatedness between a generated response and a given query, with reference-based metric; it showed high correlation with human judgments. In this paper, we explore using contextualized word embeddings to compute more accurate relatedness scores, thus better evaluation metrics. Experiments show that our evaluation metrics outperform RUBER, which is trained on static embeddings.