Sketch-Fill-A-R: A Persona-Grounded Chit-Chat Generation Framework
Michael Shum, Stephan Zheng, Wojciech Kry\'sci\'nski, Caiming Xiong,, Richard Socher

TL;DR
Sketch-Fill-A-R is a novel framework for persona-grounded chit-chat that generates engaging, consistent responses through a three-phase process involving sketching, filling, and ranking, outperforming existing models on the Persona-Chat dataset.
Contribution
It introduces a three-phase response generation framework using persona-memory, improving response quality and consistency in chit-chat conversations.
Findings
Achieves 10-point lower perplexity than baseline.
Preferred by 55% in single-turn user studies.
20% higher consistency in multi-turn interactions.
Abstract
Human-like chit-chat conversation requires agents to generate responses that are fluent, engaging and consistent. We propose Sketch-Fill-A-R, a framework that uses a persona-memory to generate chit-chat responses in three phases. First, it generates dynamic sketch responses with open slots. Second, it generates candidate responses by filling slots with parts of its stored persona traits. Lastly, it ranks and selects the final response via a language model score. Sketch-Fill-A-R outperforms a state-of-the-art baseline both quantitatively (10-point lower perplexity) and qualitatively (preferred by 55% heads-up in single-turn and 20% higher in consistency in multi-turn user studies) on the Persona-Chat dataset. Finally, we extensively analyze Sketch-Fill-A-R's responses and human feedback, and show it is more consistent and engaging by using more relevant responses and questions.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
