Vera: A General-Purpose Plausibility Estimation Model for Commonsense Statements
Jiacheng Liu, Wenya Wang, Dianzhuo Wang, Noah A. Smith, Yejin Choi, Hannaneh Hajishirzi
Abstract
Today's language models can be remarkably intelligent yet still produce text that contains trivial commonsense errors. Therefore, we seek a retrospective verification approach that can reflect on the commonsense plausibility of the machine text, and introduce Vera, a general-purpose model that learns to estimate the commonsense plausibility of declarative statements. To support diverse commonsense domains, Vera is trained on $\sim$7M commonsense statements that are automatically converted from 19 QA datasets and two commonsense knowledge bases, and using a combination of three training objectives. When applied to solving commonsense problems in the verification format, Vera substantially outperforms existing models that can be repurposed for commonsense verification, even including GPT-3.5/ChatGPT/GPT-4, and it further exhibits generalization capabilities to unseen tasks and provides well-calibrated outputs. We find that Vera excels at filtering machine-generated commonsense knowledge and is useful in detecting erroneous commonsense statements generated by models like ChatGPT in real-world settings.
BibTeX
@inproceedings{
liu2023vera,
title={Vera: A General-Purpose Plausibility Estimation Model for Commonsense Statements},
author={Jiacheng Liu and Wenya Wang and Dianzhuo Wang and Noah A. Smith and Yejin Choi and Hannaneh Hajishirzi},
booktitle={The 2023 Conference on Empirical Methods in Natural Language Processing},
year={2023},
url={https://openreview.net/forum?id=uBnIvIcAFx}
}