Multi-Dimensional Prompt Chaining to Improve Open-Domain Dialogue Generation

TL;DR

This paper introduces a multi-dimensional prompt-chaining framework that significantly enhances the dialogue quality of small language models, making them comparable to larger models in open-domain conversations.

Contribution

The paper presents a novel prompt-chaining approach that improves naturalness, coherence, and engagingness in small language models for open-domain dialogue generation.

Findings

Response diversity increased by up to 29%

Contextual coherence improved by up to 28%

Small models achieved performance comparable to larger models

Abstract

Small language models (SLMs) offer significant deployment advantages but often struggle to match the dialogue quality of larger models in open-domain settings. In this paper, we propose a multi-dimensional prompt-chaining framework that integrates Naturalness, Coherence, and Engagingness dimensions to enhance human-likeness in open-domain dialogue generation. We apply the framework to two SLMs, TinyLlama and Llama-2-7B, and benchmark their performance against responses generated by substantially larger models, including Llama-2-70B and GPT-3.5 Turbo. We then employ automatic and human evaluation to assess the responses based on diversity, contextual coherence, as well as overall quality. Results show that the full framework improves response diversity by up to 29%, contextual coherence by up to 28%, and engagingness as well as naturalness by up to 29%. Notably, Llama-2-7B achieves performance comparable to substantially larger models, including Llama-2-70B and GPT-3.5 Turbo. Overall, the findings demonstrate that carefully designed prompt-based strategies provide an effective and resource-efficient pathway to improving open-domain dialogue quality in SLMs.