CONF Anke_*SEM2022_2022/IDIAP Multilingual Extraction and Categorization of Lexical Collocations with Graph-aware Transformers Anke, Luis Espinosa Shvets, Alexander Mohammadshahi, Alireza Henderson, James Wanner, Leo EXTERNAL https://publications.idiap.ch/attachments/papers/2023/Anke_*SEM2022_2022.pdf PUBLIC Proceedings of the 11th Joint Conference on Lexical and Computational Semantics Seattle, USA 2022 89–100 Recognizing and categorizing lexical collocations in context is useful for language learning, dictionary compilation and downstream NLP. However, it is a challenging task due to the varying degrees of frozenness lexical collocations exhibit. In this paper, we put forward a sequence tagging BERT-based model enhanced with a graph-aware transformer architecture, which we evaluate on the task of collocation recognition in context. Our results suggest that explicitly encoding syntactic dependencies in the model architecture is helpful, and provide insights on differences in collocation typification in English, Spanish and French.