FlowQA: Grasping Flow in History for Conversational Machine Comprehension
Hsin-Yuan Huang California Institute of Technology、Eunsol Choi University of Washington、Wen-tau Yih Allen Institute for Artificial Intelligence Work done during internship at Allen Institute for Artificial Intelligence.
Abstract
Conversational machine comprehension requires the understanding of the conversation history, such as previous question/answer pairs, the document context and the current question. To enable traditional, single-turn models to encode the history comprehensively, we introduce Flow, a mechanism that can incorporate intermediate representations generated during the process of answering previous questions, through an alternating parallel processing structure. Compared to approaches that concatenate previous questions/answers as input, Flow integrates the latent semantics of the conversation history more deeply. Our model, FlowQA, shows superior performance on two recently proposed conversational challenges (+7.2% F1 on CoQA and +4.0% on QuAC). The effectiveness of Flow also shows in other tasks. By reducing sequential instruction understanding to conversational machine comprehension, FlowQA outperforms the best models on all three domains in SCONE, with +1.8% to +4.4% improvement in accuracy.
中文速览
对话式机器阅读理解（conversational machine comprehension）要求模型在回答每个问题时，不仅理解当前问题和文章，还要充分利用之前多轮问答的历史信息。现有方法只是把历史问题和答案文本拼接进输入，忽略了模型在回答之前问题时内部产生的推理过程。本文提出 FlowQA，核心是一个"流动机制"（Flow），它把模型回答历史问题时每一层生成的隐层表示，像流水一样传递给当前问题的推理过程，让对话历史在更深的语义层面被利用；同时设计了交替并行处理结构，解决两个方向循环叠加带来的训练效率问题。实验表明，FlowQA 在 CoQA 和 QuAC 两大对话问答数据集上分别提升了 7.2% 和 4.0% 的 F1，在序列指令理解基准 SCONE 上也全面超越已有方法。这项工作表明，让模型"记住自己是怎么想出上一个答案的"，比只记住答案本身更有价值，为构建更自然的人机对话系统提供了新思路。
原文 arXiv:1810.06683;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1810.06683v3