Spot The Bot: A Robust and Efficient Framework for the Evaluation of Conversational Dialogue Systems

TL;DR

Spot The Bot is a cost-effective evaluation framework that uses bot-to-bot conversations and human annotations to assess and rank chatbots based on their ability to mimic human-like dialogue, enabling frequent and reliable assessments.

Contribution

The paper introduces a novel, efficient evaluation framework for chatbots that replaces human-bot interactions with bot-bot conversations and uses survival analysis for ranking.

Findings

Validated on three domains with state-of-the-art chatbots

Correlates chatbot performance with characteristics like fluency and sensibleness

Allows frequent, low-cost evaluations during development cycles

Abstract

The lack of time-efficient and reliable evaluation methods hamper the development of conversational dialogue systems (chatbots). Evaluations requiring humans to converse with chatbots are time and cost-intensive, put high cognitive demands on the human judges, and yield low-quality results. In this work, we introduce \emph{Spot The Bot}, a cost-efficient and robust evaluation framework that replaces human-bot conversations with conversations between bots. Human judges then only annotate for each entity in a conversation whether they think it is human or not (assuming there are humans participants in these conversations). These annotations then allow us to rank chatbots regarding their ability to mimic the conversational behavior of humans. Since we expect that all bots are eventually recognized as such, we incorporate a metric that measures which chatbot can uphold human-like behavior the longest, i.e., \emph{Survival Analysis}. This metric has the ability to correlate a bot's performance to certain of its characteristics (e.g., \ fluency or sensibleness), yielding interpretable results. The comparably low cost of our framework allows for frequent evaluations of chatbots during their evaluation cycle. We empirically validate our claims by applying \emph{Spot The Bot} to three domains, evaluating several state-of-the-art chatbots, and drawing comparisons to related work. The framework is released as a ready-to-use tool.