Exploring Multi-Modal Representations for Ambiguity Detection & Coreference Resolution in the SIMMC 2.0 Challenge
This work addresses coreference resolution for conversational AI systems, but it is incremental as it applies existing methods to the SIMMC 2.0 challenge.
The paper tackled ambiguity detection and coreference resolution in conversational AI by using TOD-BERT and LXMERT models, showing that language models can detect ambiguity by exploiting data correlations and unimodal models can avoid vision components with smart object representations.
Anaphoric expressions, such as pronouns and referential descriptions, are situated with respect to the linguistic context of prior turns, as well as, the immediate visual environment. However, a speaker's referential descriptions do not always uniquely identify the referent, leading to ambiguities in need of resolution through subsequent clarificational exchanges. Thus, effective Ambiguity Detection and Coreference Resolution are key to task success in Conversational AI. In this paper, we present models for these two tasks as part of the SIMMC 2.0 Challenge (Kottur et al. 2021). Specifically, we use TOD-BERT and LXMERT based models, compare them to a number of baselines and provide ablation experiments. Our results show that (1) language models are able to exploit correlations in the data to detect ambiguity; and (2) unimodal coreference resolution models can avoid the need for a vision component, through the use of smart object representations.