BeliefBank: Adding Memory to a Pre-Trained Language Model for a Systematic Notion of Belief

Nora Kassner, Oyvind Tafjord, Hinrich Schütze, Peter Clark


Abstract
Although pretrained language models (PTLMs) contain significant amounts of world knowledge, they can still produce inconsistent answers to questions when probed, even after specialized training. As a result, it can be hard to identify what the model actually “believes” about the world, making it susceptible to inconsistent behavior and simple errors. Our goal is to reduce these problems. Our approach is to embed a PTLM in a broader system that also includes an evolving, symbolic memory of beliefs – a BeliefBank – that records but then may modify the raw PTLM answers. We describe two mechanisms to improve belief consistency in the overall system. First, a reasoning component – a weighted MaxSAT solver – revises beliefs that significantly clash with others. Second, a feedback component issues future queries to the PTLM using known beliefs as context. We show that, in a controlled experimental setting, these two mechanisms result in more consistent beliefs in the overall system, improving both the accuracy and consistency of its answers over time. This is significant as it is a first step towards PTLM-based architectures with a systematic notion of belief, enabling them to construct a more coherent picture of the world, and improve over time without model retraining.
Anthology ID:
2021.emnlp-main.697
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8849–8861
Language:
URL:
https://aclanthology.org/2021.emnlp-main.697
DOI:
10.18653/v1/2021.emnlp-main.697
Bibkey:
Cite (ACL):
Nora Kassner, Oyvind Tafjord, Hinrich Schütze, and Peter Clark. 2021. BeliefBank: Adding Memory to a Pre-Trained Language Model for a Systematic Notion of Belief. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 8849–8861, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
BeliefBank: Adding Memory to a Pre-Trained Language Model for a Systematic Notion of Belief (Kassner et al., EMNLP 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-3/2021.emnlp-main.697.pdf
Video:
 https://preview.aclanthology.org/nschneid-patch-3/2021.emnlp-main.697.mp4
Data
ConceptNet