Abstract Meaning Representation (AMR) is a Semantic Parsing formalism that aims at providing a semantic graph abstraction representing a given text. Current approaches are based on autoregressive language models such as BART or T5, fine-tuned through Teacher Forcing to obtain a linearized version of the AMR graph from a sentence. In this paper, we present LeakDistill, a model and method that explores a modification to the Transformer architecture, using structural adapters to explicitly incorporate graph information into the learned representations and improve AMR parsing performance. Our experiments show how, by employing word-to-node alignment to embed graph structural information into the encoder at training time, we can obtain state-of-the-art AMR parsing through self-knowledge distillation, even without the use of additional data. We release the code at \url{http://www.github.com/sapienzanlp/LeakDistill}.

翻译：摘要：抽象语义表示（AMR）是一种语义解析形式体系，旨在提供表示给定文本的语义图抽象。当前方法基于自回归语言模型（如BART或T5），通过教师强制（Teacher Forcing）微调，从句子中获取AMR图的线性化版本。本文提出LeakDistill模型与方法，探索对Transformer架构的改进，利用结构适配器将图信息显式融入学习表示中，从而提升AMR解析性能。实验表明，通过在训练时利用词-节点对齐将图结构信息嵌入编码器，即使不使用额外数据，也能通过自知识蒸馏获得最先进的AMR解析结果。相关代码已在 \url{http://www.github.com/sapienzanlp/LeakDistill} 开源。