Question answering models commonly have access to two sources of “knowledge” during inference time: (1) parametric knowledge - the factual knowledge encoded in the model weights, and (2) contextual knowledge - external knowledge (e.g., a Wikipedia passage) given to the model to generate a grounded answer. Having these two sources of knowledge entangled together is a core issue for generative QA models as it is unclear whether the answer stems from the given non-parametric knowledge or not. This unclarity has implications on issues of trust, interpretability and factuality. In this work, we propose a new paradigm in which QA models are trained to disentangle the two sources of knowledge. Using counterfactual data augmentation, we introduce a model that predicts two answers for a given question: one based on given contextual knowledge and one based on parametric knowledge. Our experiments on the Natural Questions dataset show that this approach improves the performance of QA models by making them more robust to knowledge conflicts between the two knowledge sources, while generating useful disentangled answers.
@inproceedings{DBLP:conf/acl/NeemanAHCSA23,
author = {Ella Neeman and
Roee Aharoni and
Or Honovich and
Leshem Choshen and
Idan Szpektor and
Omri Abend},
editor = {Anna Rogers and
Jordan L. Boyd{-}Graber and
Naoaki Okazaki},
title = {DisentQA: Disentangling Parametric and Contextual Knowledge with Counterfactual
Question Answering},
booktitle = {Proceedings of the 61st Annual Meeting of the Association for Computational
Linguistics (Volume 1: Long Papers), {ACL} 2023, Toronto, Canada,
July 9-14, 2023},
pages = {10056--10070},
publisher = {Association for Computational Linguistics},
year = {2023},
url = {https://doi.org/10.18653/v1/2023.acl-long.559},
doi = {10.18653/V1/2023.ACL-LONG.559},
timestamp = {Thu, 10 Aug 2023 01:00:00 +0200},
biburl = {https://dblp.org/rec/conf/acl/NeemanAHCSA23.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
See the full reference list in the paper.