Improving Factual Consistency Between a Response and Persona Facts

Mohsen Mesgar, Edwin Simpson, Iryna Gurevych

Discourse and Pragmatics Long paper Paper

Gather-2B: Apr 22, Gather-2B: Apr 22 (13:00-15:00 UTC) [Join Gather Meeting]

You can open the pre-recorded video in separate windows.

Abstract: Neural models for response generation produce responses that are semantically plausible but not necessarily factually consistent with facts describing the speaker's persona. These models are trained with fully supervised learning where the objective function barely captures factual consistency. We propose to fine-tune these models by reinforcement learning and an efficient reward function that explicitly captures the consistency between a response and persona facts as well as semantic plausibility. Our automatic and human evaluations on the PersonaChat corpus confirm that our approach increases the rate of responses that are factually consistent with persona facts over its supervised counterpart while retains the language quality of responses.
NOTE: Video may display a random order of authors. Correct author list is at the top of this page.

Connected Papers in EACL2021

Similar Papers