DialGuide: Aligning Dialogue Model Behavior with Developer Guidelines

TL;DR

DialGuide is a framework that uses natural language guidelines to control dialogue model responses, improving safety and engagement by aligning outputs with developer expectations across multiple tasks.

Contribution

It introduces DialGuide, a novel approach for guiding dialogue models with natural language rules, and provides benchmarks for related tasks in open-domain dialogue.

Findings

DialGuide effectively improves safety and engagement in dialogue responses.

Benchmark results show baseline models' performance on guideline selection, response generation, and entailment verification.

DialGuide demonstrates significant improvements in safety in the dialogue domain.

Abstract

Dialogue models are able to generate coherent and fluent responses, but they can still be challenging to control and may produce non-engaging, unsafe results. This unpredictability diminishes user trust and can hinder the use of the models in the real world. To address this, we introduce DialGuide, a novel framework for controlling dialogue model behavior using natural language rules, or guidelines. These guidelines provide information about the context they are applicable to and what should be included in the response, allowing the models to generate responses that are more closely aligned with the developer's expectations and intent. We evaluate DialGuide on three tasks in open-domain dialogue response generation: guideline selection, response generation, and response entailment verification. Our dataset contains 10,737 positive and 15,467 negative dialogue context-response-guideline triplets across two domains - chit-chat and safety. We provide baseline models for the tasks and benchmark their performance. We also demonstrate that DialGuide is effective in the dialogue safety domain, producing safe and engaging responses that follow developer guidelines.