CONF Mohammadshahi_REP4NLPATACL2023_2023/IDIAP Syntax-Aware Graph-to-Graph Transformer for Semantic Role Labelling Mohammadshahi, Alireza Henderson, James http://publications.idiap.ch/index.php/publications/showcite/Mohammadshahi_ARXIV_2021 Related documents Procceedings of 8th Workshop on Representation Learning for NLP 2023 https://arxiv.org/abs/2104.07704 URL Recent models have shown that incorporating syntactic knowledge into the semantic role labelling (SRL) task leads to a significant improvement. In this paper, we propose Syntax-aware Graph-to-Graph Transformer (SynG2G-Tr) model, which encodes the syntactic structure using a novel way to input graph relations as embeddings, directly into the self-attention mechanism of Transformer. This approach adds a soft bias towards attention patterns that follow the syntactic structure but also allows the model to use this information to learn alternative patterns. We evaluate our model on both span-based and dependency-based SRL datasets, and outperform previous alternative methods in both in-domain and out-of-domain settings, on CoNLL 2005 and CoNLL 2009 datasets. CONF Mohammadshahi_ARXIV_2021/IDIAP Syntax-Aware Graph-to-Graph Transformer for Semantic Role Labelling Mohammadshahi, Alireza Henderson, James EXTERNAL http://publications.idiap.ch/attachments/papers/2021/Mohammadshahi_ARXIV_2021.pdf PUBLIC Arxiv 2021 The goal of semantic role labelling (SRL) is to recognise the predicate-argument structure of a sentence. Recent models have shown that syntactic information can enhance the SRL performance, but other syntax-agnostic approaches achieved reasonable performance. The best way to encode syntactic information for the SRL task is still an open question. In this paper, we propose the Syntax-aware Graph-to-Graph Transformer (SynG2G-Tr) architecture, which encodes the syntactic structure with a novel way to input graph relations as embeddings directly into the self-attention mechanism of Transformer. This approach adds a soft bias towards attention patterns that follow the syntactic structure but also allows the model to use this information to learn alternative patterns. We evaluate our model on both dependency-based and span-based SRL datasets, and outperform all previous syntax-aware and syntax-agnostic models in both in-domain and out-of-domain settings, on the CoNLL 2005 and CoNLL 2009 datasets. Our architecture is general and can be applied to encode any graph information for a desired downstream task.