Towards Unified Dialogue System Evaluation: A Comprehensive Analysis of Current Evaluation Protocols

TL;DR

This paper analyzes current dialogue system evaluation protocols, highlighting their limitations and proposing a comprehensive framework based on recent research and expert assessment to improve fairness and effectiveness.

Contribution

It provides a systematic review of automated, static, and interactive evaluation methods, identifying gaps and suggesting a unified evaluation approach for dialogue systems.

Findings

Current protocols have significant shortcomings in assessing dialogue quality.

Automated and human evaluations often lack consistency and comprehensiveness.

Expert evaluation reveals key dimensions missing in existing protocols.

Abstract

As conversational AI-based dialogue management has increasingly become a trending topic, the need for a standardized and reliable evaluation procedure grows even more pressing. The current state of affairs suggests various evaluation protocols to assess chat-oriented dialogue management systems, rendering it difficult to conduct fair comparative studies across different approaches and gain an insightful understanding of their values. To foster this research, a more robust evaluation protocol must be set in place. This paper presents a comprehensive synthesis of both automated and human evaluation methods on dialogue systems, identifying their shortcomings while accumulating evidence towards the most effective evaluation dimensions. A total of 20 papers from the last two years are surveyed to analyze three types of evaluation protocols: automated, static, and interactive. Finally, the evaluation dimensions used in these papers are compared against our expert evaluation on the system-user dialogue data collected from the Alexa Prize 2020.