TRUE: Re-evaluating Factual Consistency Evaluation. (arXiv:2204.04991v3 [cs.CL] UPDATED)

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 …

arxiv evaluation