Improving Factual Consistency Between a Response and Persona Facts

By Mohsen Mesgar, Edwin Simpson, Iryna Gurevych. 2021

33 citations recorded by the source corpus.

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.

Subfields: Natural Language Processing, Alignment & RLHF

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

1. Pascale Fung
2. Nayeon Lee
3. Andrea Madotto
4. Etsuko Ishii
5. Yejin Bang
6. Yan Xu
7. Rita Frieske
8. Dan Su

Citations in the corpus

• Survey of Hallucination in Natural Language Generation (2022, methodology): “[128] attain persona consistency sub-reward via an NLI model to reduce the hallucinations in personal facts.”