DialogueScript: Using Dialogue Agents to Produce a Script

TL;DR

DialogueScript leverages dialogue agents with distinct personalities and simulated networks to generate scripts, outperforming baseline models and introducing a new dialogue consistency metric for improved evaluation.

Contribution

The paper introduces a novel method combining personality-driven dialogue agents and simulated networks for script generation, along with a new metric for dialogue consistency evaluation.

Findings

Outperforms vanilla-GPT2 baseline in script quality

Uses simulated dramatic networks for character interaction

Introduces a new dialogue consistency metric

Abstract

We present a novel approach to generating scripts by using agents with different personality types. To manage character interaction in the script, we employ simulated dramatic networks. Automatic and human evaluation on multiple criteria shows that our approach outperforms a vanilla-GPT2-based baseline. We further introduce a new metric to evaluate dialogue consistency based on natural language inference and demonstrate its validity.