CDLM: Cross-Document Language Modeling
Avi Caciularu, Arman Cohan, Iz Beltagy, Matthew Peters, Arie Cattan, Ido Dagan
Abstract
We introduce a new pretraining approach geared for multi-document language modeling, incorporating two key ideas into the masked language modeling self-supervised objective. First, instead of considering documents in isolation, we pretrain over sets of multiple related documents, encouraging the model to learn cross-document relationships. Second, we improve over recent long-range transformers by introducing dynamic global attention that has access to the entire input to predict masked tokens. We release CDLM (Cross-Document Language Model), a new general language model for multi-document setting that can be easily applied to downstream tasks. Our extensive analysis shows that both ideas are essential for the success of CDLM, and work in synergy to set new state-of-the-art results for several multi-text tasks.
BibTeX
@inproceedings{caciularu-etal-2021-cdlm-cross,
title = "{CDLM}: Cross-Document Language Modeling",
author = "Caciularu, Avi and
Cohan, Arman and
Beltagy, Iz and
Peters, Matthew and
Cattan, Arie and
Dagan, Ido",
editor = "Moens, Marie-Francine and
Huang, Xuanjing and
Specia, Lucia and
Yih, Scott Wen-tau",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2021",
month = nov,
year = "2021",
address = "Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.findings-emnlp.225/",
doi = "10.18653/v1/2021.findings-emnlp.225",
pages = "2648--2662"
}