English

Assessing Dialogue Systems with Distribution Distances

Computation and Language 2021-05-28 v3

Abstract

An important aspect of developing dialogue systems is how to evaluate and compare the performance of different systems. Existing automatic evaluation metrics are based on turn-level quality evaluation and use average scores for system-level comparison. In this paper, we propose to measure the performance of a dialogue system by computing the distribution-wise distance between its generated conversations and real-world conversations. Specifically, two distribution-wise metrics, FBD and PRD, are developed and evaluated. Experiments on several dialogue corpora show that our proposed metrics correlate better with human judgments than existing metrics.

Keywords

Cite

@article{arxiv.2105.02573,
  title  = {Assessing Dialogue Systems with Distribution Distances},
  author = {Jiannan Xiang and Yahui Liu and Deng Cai and Huayang Li and Defu Lian and Lemao Liu},
  journal= {arXiv preprint arXiv:2105.02573},
  year   = {2021}
}

Comments

7 pages, 2 figures

R2 v1 2026-06-24T01:50:02.343Z