Abstract
We participate in the DialDoc Shared Task sub-task 1 (Knowledge Identification). The task requires identifying the grounding knowledge in form of a document span for the next dialogue turn. We employ two well-known pre-trained language models (RoBERTa and ELECTRA) to identify candidate document spans and propose a metric-based ensemble method for span selection. Our methods include data augmentation, model pre-training/fine-tuning, post-processing, and ensemble. On the submission page, we rank 2nd based on the average of normalized F1 and EM scores used for the final evaluation. Specifically, we rank 2nd on EM and 3rd on F1.
Highlights
Our team SCIR-DT participates in the DialDoc shared task in the Document-grounded Dialogue and Conversational QA Workshop at the ACLIJCNLP 2021
The baseline given by the organizer is a BERT-large model without pretrained on Doc2Dial data, we fine-tune the baseline on the training set of Doc2Dial data and get the F1 of 66.84 and Exact Match (EM) of 48.48 on the dev-test set
We introduced our submission for Doc2Dial Shared Task
Summary
Our team SCIR-DT participates in the DialDoc shared task in the Document-grounded Dialogue and Conversational QA Workshop at the ACLIJCNLP 2021. There are two sub-tasks based on the Doc2Dial dataset (Feng et al, 2020). The dataset contains goal-oriented conversations between a user and an assistive agent. Sub-task is Knowledge Identification which requires identifying the grounding knowledge in form of document span for the agent turn. The input is dialogue history, current user utterance, and associated document. The evaluation metrics are Exact Match (EM) and F1 (Rajpurkar et al, 2016). Sub-task is text generation which requires generating the agent response in natural language. The input is dialogue history and associated document. The evaluation metrics are SacreBLEU (Post, 2018) and human evaluations.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.