@inproceedings{issam-etal-2025-representation,
    title = "A Representation Level Analysis of {NMT} Model Robustness to Grammatical Errors",
    author = "Issam, Abderrahmane  and
      Semerci, Yusuf Can  and
      Scholtes, Jan  and
      Spanakis, Gerasimos",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/display_plenaries/2025.findings-acl.451/",
    pages = "8579--8601",
    ISBN = "979-8-89176-256-5",
    abstract = "Understanding robustness is essential for building reliable NLP systems. Unfortunately, in the context of machine translation, previous work mainly focused on documenting robustness failures or improving robustness. In contrast, we study robustness from a model representation perspective by looking at internal model representations of ungrammatical inputs and how they evolve through model layers. For this purpose, we perform Grammatical Error Detection (GED) probing and representational similarity analysis. Our findings indicate that the encoder first detects the grammatical error, then corrects it by moving its representation toward the correct form. To understand what contributes to this process, we turn to the attention mechanism where we identify what we term *Robustness Heads*. We find that *Robustness Heads* attend to interpretable linguistic units when responding to grammatical errors, and that when we fine-tune models for robustness, they tend to rely more on *Robustness Heads* for updating the ungrammatical word representation."
}