Controlling Style in Generated Dialogue
Eric Michael Smith*, Diana Gonzalez-Rico*, Emily Dinan, Y-Lan Boureau Facebook AI Research
Abstract
Open-domain conversation models have become good at generating natural-sounding dialogue, using very large architectures with billions of trainable parameters. The vast training data required to train these architectures aggregates many different styles, tones, and qualities. Using that data to train a single model makes it difficult to use the model as a consistent conversational agent, e.g. with a stable set of persona traits and a typical style of expression. Several architectures affording control mechanisms over generation architectures have been proposed, each with different trade-offs. However, it remains unclear whether their use in dialogue is viable, and what the trade-offs look like with the most recent state-of-the-art conversational architectures. In this work, we adapt three previously proposed controllable generation architectures to open-domain dialogue generation, controlling the style of the generation to match one among about 200 possible styles. We compare their respective performance and trade-offs, and show how they can be used to provide insights into existing conversational datasets, and generate a varied set of styled conversation replies.
中文速览
让对话系统生成风格一致、个性鲜明的回复，一直是个难题——现有大模型虽然流畅，却因训练数据来源杂乱而显得"人格分裂"。研究者把三种已有的可控生成方法（检索+风格迁移、推理时迭代优化PPLM、以及在输入中拼接风格标签的条件生成模型）分别改造适配到27亿参数的顶尖对话模型上，目标是让模型能在约200种不同风格（如"好奇"、"随意"、"热情"等）之间自由切换。实验结果表明，直接在输入端拼接风格标签并微调（conditioned generation）的方案综合表现最佳：风格命中率高、推理速度快、对对话质量的损伤也最小，而PPLM虽然灵活但推理开销极大。这项工作不仅提供了一套可推广到任意风格空间的实用流程，还通过开源代码让研究者能直接在ParlAI框架中复用这些风格可控的对话模型，为构建个性稳定、体验一致的对话助手奠定了基础。
原文 arXiv:2009.10855;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2009.10855v1