ConvAI3: Generating Clarifying Questions for Open-Domain Dialogue Systems (ClariQ)

TL;DR

ClariQ tackles ambiguity in open-domain dialogue by formalizing a challenge to generate clarifying questions. It couples an offline Stage 1 dataset and evaluation with a Stage 2 human-in-the-loop setup to assess both question usefulness and user experience, exploring generative and retrieval-based approaches. The work contributes a labeled conversational dataset, an explicit two-stage evaluation framework, and metrics that connect clarifying questions to downstream retrieval performance, enabling principled measurement of when and how to ask clarifications. These contributions advance mixed-initiative dialogue research and inform practical designs for clarifying questions in conversational information seeking.

Abstract

This document presents a detailed description of the challenge on clarifying questions for dialogue systems (ClariQ). The challenge is organized as part of the Conversational AI challenge series (ConvAI3) at Search Oriented Conversational AI (SCAI) EMNLP workshop in 2020. The main aim of the conversational systems is to return an appropriate answer in response to the user requests. However, some user requests might be ambiguous. In IR settings such a situation is handled mainly thought the diversification of the search result page. It is however much more challenging in dialogue settings with limited bandwidth. Therefore, in this challenge, we provide a common evaluation framework to evaluate mixed-initiative conversations. Participants are asked to rank clarifying questions in an information-seeking conversations. The challenge is organized in two stages where in Stage 1 we evaluate the submissions in an offline setting and single-turn conversations. Top participants of Stage 1 get the chance to have their model tested by human annotators.