Recent generative approaches for multi-hop question answering (QA) utilize the fusion-in-decoder method~\cite{izacard-grave-2021-leveraging} to generate a single sequence output which includes both a final answer and a reasoning path taken to arrive at that answer, such as passage titles and key facts from those passages. While such models can lead to better interpretability and high quantitative scores, they often have difficulty accurately identifying the passages corresponding to key entities in the context, resulting in incorrect passage hops and a lack of faithfulness in the reasoning path. To address this, we propose a single-sequence prediction method over a local reasoning graph (\model)\footnote{Code/Models will be released at \url{https://github.com/gowtham1997/SeqGraph}} that integrates a graph structure connecting key entities in each context passage to relevant subsequent passages for each question. We use a graph neural network to encode this graph structure and fuse the resulting representations into the entity representations of the model. Our experiments show significant improvements in answer exact-match/F1 scores and faithfulness of grounding in the reasoning path on the HotpotQA dataset and achieve state-of-the-art numbers on the Musique dataset with only up to a 4\% increase in model parameters.

翻译：近期面向多跳问答（QA）的生成式方法采用融合解码器方法~\cite{izacard-grave-2021-leveraging}生成单一序列输出，该输出同时包含最终答案及推导过程（如段落标题及其关键事实）。此类模型虽能提升可解释性并获得较高量化分数，但常难以准确识别与上下文中关键实体相对应的段落，导致段落跳跃错误及推理路径缺乏忠实性。为此，我们提出一种基于局部推理图（\model）的单序列预测方法——该图结构将各上下文段落中的关键实体与后续相关段落建立连接，并利用图神经网络编码该结构，将生成的表示融合至模型实体表示中。实验表明，该方法在HotpotQA数据集上显著提升了答案的精确匹配率/F1分数与推理路径基础忠实度，在Musique数据集上以仅增加4%模型参数的代价刷新了当前最优性能。