Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

plausibility estimation

Plausibility estimation is the computational process of determining the degree to which a statement or piece of text aligns with general commonsense knowledge and everyday reality. Rather than verifying strict factual truth against an authoritative record, plausibility estimation evaluates whether an assertion is reasonable, likely, or conceptually and physically coherent in the real world. In artificial intelligence and natural language processing, this task functions primarily as an automated verification method used to evaluate machine-generated text, identify commonsense reasoning errors and hallucinations, filter candidate knowledge base assertions, and score the credibility of model outputs.

1 item

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

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

Jiacheng Liu, Wenya Wang, Dianzhuo Wang, Noah A. Smith, Yejin Choi, Hannaneh Hajishirzi

OrganizationsAllen Institute for AIHarvard UniversityNanyang Technological UniversityUniversity of Washington

Why you should read this

Presents VERA, a standalone commonsense verification model trained on millions of statements that outperforms systems like GPT-4 in estimating the plausibility of declarative claims and detecting errors in language model outputs.

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 ~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.

Added

2026-09-26