@inproceedings{iranzo-sanchez-etal-2026-mllp,
    title = "{MLLP}-{VRAIN} {UPV} System for the {IWSLT} 2026 Simultaneous Speech Translation Task",
    author = "Iranzo-S{\'a}nchez, Jorge  and
      Mas-Moll{\`a}, Gerard  and
      Gimenez, Adri{\`a}  and
      Civera Saiz, Jorge  and
      Sanchis, Albert  and
      Juan, Alfons",
    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/bulk-corrections-2026-07-02/2026.iwslt-1.24/",
    doi = "10.18653/v1/2026.iwslt-1.24",
    pages = "212--226",
    ISBN = "979-8-89176-411-8",
    abstract = "This work describes the participation of the MLLP-VRAIN research group in the shared task of the IWSLT 2026 Simultaneous Speech Translation track. Our submission utilizes the recently released Parakeet and Qwen 3.5 models to create a robust, cascaded solution for long-form SimulST through the use of adaptive black-box policies. We explore relaxations of these policies to achieve better quality-latency trade-offs. Compared to last year, we participate on all language directions. In addition to this, for the En{\textrightarrow}{De, It, Zh} directions we also participate in this year{'}s new context track employing a combination of ASR word-boosting and a RAG mechanism of offline pre-translated exemplars to guide generation and enrich our system with domain-specific context. Finally, we provide a detailed latency analysis of our system. Compared to last year, results on the MCIF En{\textrightarrow}De test set shows a substantial quality improvement of +5.82 XCOMET-XL. Our context track processing further improves performance by +1.03."
}