{"ID":2877638,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.19856","arxiv_id":"2508.19856","title":"TokenVerse++: Towards Flexible Multitask Learning with Dynamic Task Activation","abstract":"Token-based multitasking frameworks like TokenVerse require all training utterances to have labels for all tasks, hindering their ability to leverage partially annotated datasets and scale effectively. We propose TokenVerse++, which introduces learnable vectors in the acoustic embedding space of the XLSR-Transducer ASR model for dynamic task activation. This core mechanism enables training with utterances labeled for only a subset of tasks, a key advantage over TokenVerse. We demonstrate this by successfully integrating a dataset with partial labels, specifically for ASR and an additional task, language identification, improving overall performance. TokenVerse++ achieves results on par with or exceeding TokenVerse across multiple tasks, establishing it as a more practical multitask alternative without sacrificing ASR performance.","short_abstract":"Token-based multitasking frameworks like TokenVerse require all training utterances to have labels for all tasks, hindering their ability to leverage partially annotated datasets and scale effectively. We propose TokenVerse++, which introduces learnable vectors in the acoustic embedding space of the XLSR-Transducer ASR...","url_abs":"https://arxiv.org/abs/2508.19856","url_pdf":"https://arxiv.org/pdf/2508.19856v1","authors":"[\"Shashi Kumar\",\"Srikanth Madikeri\",\"Esaú Villatoro-Tello\",\"Sergio Burdisso\",\"Pradeep Rangappa\",\"Andrés Carofilis\",\"Petr Motlicek\",\"Karthik Pandia\",\"Shankar Venkatesan\",\"Kadri Hacioğlu\",\"Andreas Stolcke\"]","published":"2025-08-27T13:16:31Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"eess.AS\"]","methods":"[]","has_code":false}