arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.08451cs.CLcs.AI
收藏

用于问题匹配的上下文感知交互网络

Context-Aware Interaction Network for Question Matching

Baidu Inc.(百度公司);Rutgers University(罗格斯大学);Northeastern University(东北大学);HPI/University of Potsdam(HPI/波茨坦大学)

Zhe Hu, Zuohui Fu, Yu Yin, Gerard de Melo

摘要:

通过采用交叉注意力机制来捕捉两个句子表示之间的相关语义联系,在文本匹配中取得了令人印象深刻的里程碑。然而,常规的交叉注意集中在两个输入序列之间的词级链接,忽视了上下文信息的重要性。我们提出了一个上下文感知的交互网络(COIN),正确地对齐两个序列,并推断它们的语义关系。具体地,每个交互块包括(1)上下文感知交叉注意力机制,以在比对两个序列时有效地整合上下文信息,以及(2)门融合层,以灵活地内插比对的表示。我们应用多个堆叠的交互块来产生不同级别的对齐,并逐渐完善注意力结果。在两个问题匹配数据集上的实验和详细的分析证明了该模型的有效性。

英文摘要:

Impressive milestones have been achieved in text matching by adopting a cross-attention mechanism to capture pertinent semantic connections between two sentence representations. However, regular cross-attention focuses on word-level links between the two input sequences, neglecting the importance of contextual information. We propose a context-aware interaction network (COIN) to properly align two sequences and infer their semantic relationship. Specifically, each interaction block includes (1) a context-aware cross-attention mechanism to effectively integrate contextual information when aligning two sequences, and (2) a gate fusion layer to flexibly interpolate aligned representations. We apply multiple stacked interaction blocks to produce alignments at different levels and gradually refine the attention results. Experiments on two question matching datasets and detailed analyses demonstrate the effectiveness of our model.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑