@inproceedings{gupta-etal-2025-stance,
    title = "-Stance: A Large-Scale Real World Dataset of Stances in Legal Argumentation",
    author = "Gupta, Ankita  and
      Rice, Douglas  and
      O{'}Connor, Brendan",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.acl-long.1517/",
    pages = "31450--31467",
    ISBN = "979-8-89176-251-0",
    abstract = "We present -Stance, a large-scale dataset of stances involved in legal argumentation.-Stance contains stance-annotated argument pairs, semi-automatically mined from millions of examples of U.S. judges citing precedent in context using citation signals. The dataset aims to facilitate work on the legal argument stance classification task, which involves assessing whether a case summary strengthens or weakens a legal argument (polarity) and to what extent (intensity). To assess the complexity of this task, we evaluate various existing NLP methods, including zero-shot prompting proprietary large language models (LLMs), and supervised fine-tuning of smaller open-weight language models (LMs) on 𝛿-Stance. Our findings reveal that although prompting proprietary LLMs can help predict stance polarity, supervised model fine-tuning on -Stance is necessary to distinguish intensity. We further find that alternative strategies such as domain-specific pretraining and zero-shot prompting using masked LMs remain insufficient. Beyond our dataset{'}s utility for the legal domain, we further find that fine-tuning small LMs on -Stance improves their performance in other domains. Finally, we study how temporal changes in signal definition can impact model performance, highlighting the importance of careful data curation for downstream tasks by considering the historical and sociocultural context. We publish the associated dataset to foster further research on legal argument reasoning."
}