@inproceedings{zhang-2026-llm,
    title = "{LLM}-Based Frame and Stance Annotation for 19th-Century Rumour Discourse in {US} and {UK} Newspapers",
    author = "Zhang, Wanshu",
    editor = "Oleskeviciene, Giedre Valunaite  and
      Giouli, Voula  and
      Armaselu, Florentina  and
      Liebeskind, Chaya  and
      McGillivray, Barbara",
    booktitle = "Proceedings of the Workshop Neology and Large Language Models",
    month = may,
    year = "2026",
    address = "Palma, Mallorca (Spain)",
    publisher = "ELRA Language Resources Association (ELRA)",
    url = "https://preview.aclanthology.org/test-year-match/2026.neollm-1.8/",
    doi = "10.63317/4qdnrzfmibbz",
    pages = "66--70",
    abstract = "Large language models (LLMs) are increasingly used for lexicographic support, neology detection, and semantic categorization, yet their behaviour on historical newspapers remains under-evaluated. This short paper describes an ongoing project that extends a DH2026-accepted two-phase methodology for extracting and tracking rumours in historical newspapers. From large-scale US and UK corpora (PleIAs/US-PD-Newspapers; biglam/hmd{\_}newspapers), the DH workflow produces gold-standard sentence-level rumour instances with proposition-like ``rumour content'' spans (Rumour{\_}Content/Cleaned{\_}Content) and extraction-pattern metadata. Building on these historically grounded units, we propose an LLM-centered benchmark and analysis pipeline for assigning topical frames and evidential stance to rumour propositions, and for auditing ``temporal projection'' when models introduce anachronistic modern misinformation framings. For controlled cross-variety comparison we construct a strictly balanced benchmark of 800 instances over two well-attested bins (1840{--}1859, 1860{--}1879) and both national varieties (200 per country per bin). We outline prompt conditions (text-only vs time-aware vs historically calibrated) and self-consistency voting to quantify label stability and error modes. A small manually annotated subset supports evaluation, while the main contribution is the benchmark design, prompts, and reproducible protocol enabling community feedback before full-scale results are finalized."
}