@inproceedings{katrapati-shrivastava-2025-constructions,
    title = "Can Constructions ``{SCAN}'' Compositionality ?",
    author = "Katrapati, Ganesh  and
      Shrivastava, Manish",
    editor = "Bonial, Claire  and
      Torgbi, Melissa  and
      Weissweiler, Leonie  and
      Blodgett, Austin  and
      Beuls, Katrien  and
      Van Eecke, Paul  and
      Tayyar Madabushi, Harish",
    booktitle = "Proceedings of the Second International Workshop on Construction Grammars and NLP",
    month = sep,
    year = "2025",
    address = {D{\"u}sseldorf, Germany},
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/iwcs-25-ingestion/2025.cxgsnlp-1.17/",
    pages = "165--171",
    ISBN = "979-8-89176-318-0",
    abstract = "Sequence to Sequence models struggle at compositionality and systematic generalisation even while they excel at many other tasks.We attribute this limitation to their failure to internalise constructions{---}conventionalised form{--}meaning pairings that license productive recombination. Building on these insights, we introduce an unsupervised procedure for mining pseudo-constructions: variable-slot templates automatically extracted from training data. When applied to the SCAN dataset, ourmethod yields large gains out-of-distribution splits: accuracy rises to 47.8{\%} on ADD JUMP and to 20.3{\%} on AROUND RIGHT without any architectural changes or additional supervision. The model also attains competitive performance with {\ensuremath{\leq}} 40{\%} of the original training data, demonstrating strong data efficiency. Our findings highlight the promise of construction-aware preprocessing as an alternative to heavy architectural or training-regime interventions."
}