@inproceedings{zampieri-etal-2019-impact,
    title = "The Impact of Word Representations on Sequential Neural {MWE} Identification",
    author = "Zampieri, Nicolas  and
      Ramisch, Carlos  and
      Damnati, Geraldine",
    editor = "Savary, Agata  and
      Escart{\'\i}n, Carla Parra  and
      Bond, Francis  and
      Mitrovi{\'c}, Jelena  and
      Mititelu, Verginica Barbu",
    booktitle = "Proceedings of the Joint Workshop on Multiword Expressions and WordNet (MWE-WN 2019)",
    month = aug,
    year = "2019",
    address = "Florence, Italy",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/W19-5121",
    doi = "10.18653/v1/W19-5121",
    pages = "169--175",
    abstract = "Recent initiatives such as the PARSEME shared task allowed the rapid development of MWE identification systems. Many of those are based on recent NLP advances, using neural sequence models that take continuous word representations as input. We study two related questions in neural MWE identification: (a) the use of lemmas and/or surface forms as input features, and (b) the use of word-based or character-based embeddings to represent them. Our experiments on Basque, French, and Polish show that character-based representations yield systematically better results than word-based ones. In some cases, character-based representations of surface forms can be used as a proxy for lemmas, depending on the morphological complexity of the language.",
}