Generalized Shortest-Paths Encoders for AMR-to-Text Generation
Generalized Shortest-Paths Encoders for AMR-to-Text Generation
复制标题
DOI：
10.18653/v1/2020.coling-main.181
复制
发表时间：
2020-12
期刊：
影响因子：
--
通讯作者：
Lisa Jin;D. Gildea
中科院分区：
文献类型：
--
作者：
Lisa Jin;D. Gildea
对于从语义图生成文本，过去的神经模型通过沿着图边缘的门控卷积来编码输入结构。尽管这些操作提供了本地上下文，但消息可以传播的距离受到编码器传播步骤数量的限制。我们最近将 Transformer 自注意力应用到图上，以实现全局特征传播。我们不是将最短路径提供给顶点自注意力模块，而是训练一个模型来使用广义最短路径算法来学习它们。这种方法通过将图编码器暴露给所有可能的图路径来扩大图编码器的感受野。我们探讨了这种路径多样性如何影响 AMR 连接级别的性能，展示了更高重入次数和直径的 AMR 的增益。对生成句子的分析也支持我们的可重入 AMR 模型的高度语义一致性。我们的最佳模型在仅编码成对唯一最短路径的基线上实现了 1.4 BLEU 和 1.8 chrF++ 裕度。
For text generation from semantic graphs, past neural models encoded input structure via gated convolutions along graph edges. Although these operations provide local context, the distance messages can travel is bounded by the number of encoder propagation steps. We adopt recent efforts of applying Transformer self-attention to graphs to allow global feature propagation. Instead of feeding shortest paths to the vertex self-attention module, we train a model to learn them using generalized shortest-paths algorithms. This approach widens the receptive field of a graph encoder by exposing it to all possible graph paths. We explore how this path diversity affects performance across levels of AMR connectivity, demonstrating gains on AMRs of higher reentrancy counts and diameters. Analysis of generated sentences also supports high semantic coherence of our models for reentrant AMRs. Our best model achieves a 1.4 BLEU and 1.8 chrF++ margin over a baseline that encodes only pairwise-unique shortest paths.