@inproceedings{zufle-etal-2025-nutshell,
    title = "{NUTSHELL}: A Dataset for Abstract Generation from Scientific Talks",
    author = {Z{\"u}fle, Maike  and
      Papi, Sara  and
      Savoldi, Beatrice  and
      Gaido, Marco  and
      Bentivogli, Luisa  and
      Niehues, Jan},
    editor = "Salesky, Elizabeth  and
      Federico, Marcello  and
      Anastasopoulos, Antonis",
    booktitle = "Proceedings of the 22nd International Conference on Spoken Language Translation (IWSLT 2025)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria (in-person and online)",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/acl25-workshop-ingestion/2025.iwslt-1.2/",
    pages = "19--32",
    ISBN = "979-8-89176-272-5",
    abstract = "Scientific communication is receiving increasing attention in natural language processing, especially to help researches access, summarize, and generate content. One emerging application in this area is Speech-to-Abstract Generation (SAG), which aims to automatically generate abstracts from recorded scientific presentations. SAG enables researchers to efficiently engage with conference talks, but progress has been limited by a lack of large-scale datasets. To address this gap, we introduce NUTSHELL, a novel multimodal dataset of *ACL conference talks paired with their corresponding abstracts. We establish strong baselines for SAG and evaluate the quality of generated abstracts using both automatic metrics and human judgments. Our results highlight the challenges of SAG and demonstrate the benefits of training on NUTSHELL. By releasing NUTSHELL under an open license (CC-BY 4.0), we aim to advance research in SAG and foster the development of improved models and evaluation methods."
}