@inproceedings{krahn-fosler-lussier-2026-hydraqe,
    title = "{H}ydra{QE}: {OSU}{'}s Submission for the {IWSLT} 2026 Speech Translation Metrics Shared Task",
    author = "Krahn, Kevin  and
      Fosler-Lussier, Eric",
    editor = "Salesky, Elizabeth  and
      Anastasopoulos, Antonios  and
      Negri, Matteo  and
      Federico, Marcello",
    booktitle = "Proceedings of the 23rd International Conference on Spoken Language Translation ({IWSLT} 2026)",
    month = jul,
    year = "2026",
    address = "San Diego, USA (in-person and online)",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/corrections-2026-06/2026.iwslt-1.37/",
    doi = "10.18653/v1/2026.iwslt-1.37",
    pages = "323--331",
    ISBN = "979-8-89176-411-8",
    abstract = "We present HydraQE, our contribution to the IWSLT 2026 Speech Translation Metrics shared task. HydraQE is an end-to-end, reference-free quality estimation (QE) system for speech translation built on a Qwen3-ASR backbone, which accepts source audio and a translation hypothesis as joint input. Hidden states from all backbone layers are combined via a sparsemax scalar mix, then re-encoded by a bidirectional Transformer for full cross-modal interaction. To address the scarcity of human-annotated speech translation data, three independent prediction heads are trained on complementary supervision signals: human direct assessment (DA) annotations, MetricX-24 pseudo-labels, and xCOMET pseudo-labels. We train on a combination of synthetically corrupted examples and silver pseudo-labeled machine translation outputs, using a curriculum that begins on synthetic and silver data and gradually shifts toward human-annotated examples. HydraQE outperforms cascaded text-based baselines and prior direct speech QE systems, demonstrating that end-to-end speech translation QE is competitive with cascaded approaches."
}