Reason first, then respond: Modular Generation for Knowledge-infused Dialogue
Leonard Adolphs ETH Zürich \AndKurt Shuster Facebook AI Research \ANDJack Urbanek Facebook AI Research \AndArthur Szlam Facebook AI Research \AndJason Weston Facebook AI Research Work done during a Facebook AI Research internship.
Abstract
Large language models can produce fluent dialogue but often hallucinate factual inaccuracies. While retrieval-augmented models help alleviate this issue, they still face a difficult challenge of both reasoning to provide correct knowledge and generating conversation simultaneously. In this work, we propose a modular model, Knowledge to Response (K2R), for incorporating knowledge into conversational agents, which breaks down this problem into two easier steps. K2R first generates a knowledge sequence, given a dialogue context, as an intermediate step. After this “reasoning step”, the model then attends to its own generated knowledge sequence, as well as the dialogue context, to produce a final response. In detailed experiments, we find that such a model hallucinates less in knowledge-grounded dialogue tasks, and has advantages in terms of interpretability and modularity. In particular, it can be used to fuse QA and dialogue systems together to enable dialogue agents to give knowledgeable answers, or QA models to give conversational responses in a zero-shot setting.
中文速览
大型语言模型（LLM）在对话中经常"一本正经地胡说八道"——生成听起来流畅却事实有误的内容，即所谓的幻觉（hallucination）问题，而现有检索增强模型虽有改善，却仍要求模型同时完成"理解文档、抽取知识、生成对话"三件事，导致错误依然难以避免。本文提出了一个名为 K2R（Knowledge to Response）的模块化两步模型：第一步先根据对话上下文显式生成一段"中间知识序列"，第二步再把这段知识连同对话上下文一起输入，生成最终回复，把一个复杂任务拆成两个更容易的子任务。实验表明，K2R 在知识驱动对话数据集上显著减少了幻觉，知识 F1 分数优于 RAG 等基线，人工评估也确认其回复更准确、更具吸引力；尤其值得一提的是，模块化设计让 QA 模型和对话模型可以在零样本条件下直接"拼接"使用，无需重新训练。这一工作为构建既能闲聊又能准确答题的对话智能体提供了一条切实可行的路径。
原文 arXiv:2111.05204;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2111.05204v1