Docalog: Multi-document Dialogue System using Transformer-based Span Retrieval
Published in Proceedings of the Second DialDoc Workshop on Document-grounded Dialogue and Conversational Question Answering, 2022
Recommended citation: Sayed Hesam Alavian, Ali Satvaty, Sadra Sabouri, Ehsaneddin Asgari, and Hossein Sameti. 2022. Docalog: Multi-document Dialogue System using Transformer-based Span Retrieval. In Proceedings of the Second DialDoc Workshop on Document-grounded Dialogue and Conversational Question Answering, pages 142–147, Dublin, Ireland. Association for Computational Linguistics. https://aclanthology.org/2022.dialdoc-1.16/
Information-seeking dialogue systems, including knowledge identification and response generation, aim to respond to users with fluent, coherent, and informative answers based on users’ needs. This paper discusses our proposed approach, Docalog, for the DialDoc-22 (MultiDoc2Dial) shared task. Docalog identifies the most relevant knowledge in the associated document, in a multi-document setting. Docalog, is a three-stage pipeline consisting of (1) a document retriever model (DR. TEIT),(2) an answer span prediction model, and (3) an ultimate span picker deciding on the most likely answer span, out of all predicted spans. In the test phase of MultiDoc2Dial 2022, Docalog achieved f1-scores of 36.07% and 28.44% and SacreBLEU scores of 23.70% and 20.52%, respectively on the MDD-SEEN and MDD-UNSEEN folds.