This repository will contain the official source code of our paper "WSL-DS: Weakly Supervised Learning with Distant Supervision for Query Focused Multi-Document Abstractive Summarization" accepted in COLING 2020.
Source code for the RoBERTa-based sentence similarity model for distant supervision as well as for final summary selection can be found here: https://github.com/tahmedge/CETE-LREC
Source code for fine-tuning pre-trained BERTSUM model for the Query Focused Abstractive Summarization task can be found here: https://github.com/tahmedge/QR-BERTSUM-TL-for-QFAS