DocTime: A Document-level Temporal Dependency Graph Parser
Puneet Mathur, Vlad Morariu, Verena Kaynig-Fittkau, Jiuxiang Gu, Franck Dernoncourt, Quan Tran, Ani Nenkova, Dinesh Manocha
Abstract
We introduce DocTime - a novel temporal dependency graph (TDG) parser that takes as input a text document and produces a temporal dependency graph. It outperforms previous BERT-based solutions by a relative 4-8% on three datasets from modeling the problem as a graph network with path-prediction loss to incorporate longer range dependencies. This work also demonstrates how the TDG graph can be used to improve the downstream tasks of temporal questions answering and NLI by a relative 4-10% with a new framework that incorporates the temporal dependency graph into the self-attention layer of Transformer models (Time-transformer). Finally, we develop and evaluate on a new temporal dependency graph dataset for the domain of contractual documents, which has not been previously explored in this setting.
BibTeX
@inproceedings{mathur-etal-2022-doctime,
title = "{D}oc{T}ime: A Document-level Temporal Dependency Graph Parser",
author = "Mathur, Puneet and
Morariu, Vlad and
Kaynig-Fittkau, Verena and
Gu, Jiuxiang and
Dernoncourt, Franck and
Tran, Quan and
Nenkova, Ani and
Manocha, Dinesh and
Jain, Rajiv",
editor = "Carpuat, Marine and
de Marneffe, Marie-Catherine and
Meza Ruiz, Ivan Vladimir",
booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.naacl-main.73/",
doi = "10.18653/v1/2022.naacl-main.73",
pages = "993--1009"
}