DeepAI AI Chat
Log In Sign Up

TRUE: Re-evaluating Factual Consistency Evaluation

by   Or Honovich, et al.

Grounded text generation systems often generate text that contains factual inconsistencies, hindering their real-world applicability. Automatic factual consistency evaluation may help alleviate this limitation by accelerating evaluation cycles, filtering inconsistent outputs and augmenting training data. While attracting increasing attention, such evaluation metrics are usually developed and evaluated in silo for a single task or dataset, slowing their adoption. Moreover, previous meta-evaluation protocols focused on system-level correlations with human annotations, which leave the example-level accuracy of such metrics unclear. In this work, we introduce TRUE: a comprehensive study of factual consistency metrics on a standardized collection of existing texts from diverse tasks, manually annotated for factual consistency. Our standardization enables an example-level meta-evaluation protocol that is more actionable and interpretable than previously reported correlations, yielding clearer quality measures. Across diverse state-of-the-art metrics and 11 datasets we find that large-scale NLI and question generation-and-answering-based approaches achieve strong and complementary results. We recommend those methods as a starting point for model and metric developers, and hope TRUE will foster progress towards even better methods.


page 1

page 2

page 3

page 4


Go Figure! A Meta Evaluation of Factuality in Summarization

Text generation models can generate factually inconsistent text containi...

Q^2: Evaluating Factual Consistency in Knowledge-Grounded Dialogues via Question Generation and Question Answering

Neural knowledge-grounded generative models for dialogue often produce c...

GENIE: A Leaderboard for Human-in-the-Loop Evaluation of Text Generation

Leaderboards have eased model development for many NLP datasets by stand...

Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generation

Natural language generation (NLG) spans a broad range of tasks, each of ...

Is ChatGPT a Good NLG Evaluator? A Preliminary Study

Recently, the emergence of ChatGPT has attracted wide attention from the...

Plot-guided Adversarial Example Construction for Evaluating Open-domain Story Generation

With the recent advances of open-domain story generation, the lack of re...

ROSCOE: A Suite of Metrics for Scoring Step-by-Step Reasoning

Large language models show improved downstream task performance when pro...