Retrieve and Refine: Improved Sequence Generation Models For Dialogue
Jason Weston, Emily Dinan and Alexander H. Miller Facebook AI Research
Abstract
Sequence generation models for dialogue are known to have several problems: they tend to produce short, generic sentences that are uninformative and unengaging. Retrieval models on the other hand can surface interesting responses, but are restricted to the given retrieval set leading to erroneous replies that cannot be tuned to the specific context. In this work we develop a model that combines the two approaches to avoid both their deficiencies: first retrieve a response and then refine it – the final sequence generator treating the retrieval as additional context. We show on the recent ConvAI2 challenge task our approach produces responses superior to both standard retrieval and generation models in human evaluations.
中文速览
对话系统中的生成模型（如Seq2Seq）常常输出短而无聊的万能回复，而检索模型（retrieval model）虽能给出有趣的候选回复，却无法根据当前语境灵活调整。为此，研究者提出了"检索再精炼"（Retrieve and Refine）框架：先用记忆网络检索一条相关回复，再将其拼接到生成模型的输入中，让生成模型以检索结果为参考、按需修改或直接采用。在ConvAI2对话挑战数据集上，人工评测和A/B测试均显示，改进版RetNRef++在互动性上超过了单纯的检索模型和生成模型，同时保留了生成新内容的能力。这项工作为解决对话生成中长期存在的"无聊回复"难题提供了一条简单有效的路径，也指出了自动评估指标（如困惑度）在此类任务中的局限性。
原文 arXiv:1808.04776;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1808.04776v2