Vera: A General-Purpose Plausibility Estimation Model for Commonsense Statements

TL;DR

Vera is a versatile plausibility estimation model trained on extensive commonsense data, effectively identifying correct statements, filtering generated knowledge, and detecting errors across diverse domains and unseen tasks.

Contribution

Introduces Vera, a general-purpose plausibility model for commonsense statements, trained on large datasets, with multiple objectives, outperforming existing models in verification tasks.

Findings

Vera outperforms existing models in commonsense verification.

Vera generalizes well to unseen tasks.

Vera effectively filters and detects erroneous commonsense statements.

Abstract

Despite the much discussed capabilities of today's language models, they are still prone to silly and unexpected commonsense failures. We consider a retrospective verification approach that reflects on the correctness of LM outputs, and introduce Vera, a general-purpose model that estimates the plausibility of declarative statements based on commonsense knowledge. Trained on ~7M commonsense statements created from 19 QA datasets and two large-scale knowledge bases, and with a combination of three training objectives, Vera is a versatile model that effectively separates correct from incorrect statements across diverse commonsense domains. When applied to solving commonsense problems in the verification format, Vera substantially outperforms existing models that can be repurposed for commonsense verification, and it further exhibits generalization capabilities to unseen tasks and provides well-calibrated outputs. We find that Vera excels at filtering LM-generated commonsense knowledge and is useful in detecting erroneous commonsense statements generated by models like ChatGPT in real-world settings.