@inproceedings{galimzianova-etal-2025-rag,
    title = "From {RAG} to Reality: Coarse-Grained Hallucination Detection via {NLI} Fine-Tuning",
    author = "Galimzianova, Daria  and
      Boriskin, Aleksandr  and
      Arshinov, Grigory",
    editor = "Ghosal, Tirthankar  and
      Mayr, Philipp  and
      Singh, Amanpreet  and
      Naik, Aakanksha  and
      Rehm, Georg  and
      Freitag, Dayne  and
      Li, Dan  and
      Schimmler, Sonja  and
      De Waard, Anita",
    booktitle = "Proceedings of the Fifth Workshop on Scholarly Document Processing (SDP 2025)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/landing_page/2025.sdp-1.34/",
    doi = "10.18653/v1/2025.sdp-1.34",
    pages = "353--359",
    ISBN = "979-8-89176-265-7",
    abstract = "We present our submission to SciHal Subtask 1: coarse-grained hallucination detection for scientific question answering. We frame hallucination detection as an NLI-style three-way classification (entailment, contradiction, unverifiable) and show that simple fine-tuning of NLI-adapted encoder models on task data outperforms more elaborate feature-based pipelines and large language model prompting. In particular, DeBERTa-V3-large, a model pretrained on five diverse NLI corpora, achieves the highest weighted F1 on the public leaderboard. We additionally explore a pipeline combining joint claim{--}reference embeddings and NLI softmax probabilities fed into a classifier, but find its performance consistently below direct encoder fine-tuning. Our findings demonstrate that, for reference-grounded hallucination detection, targeted encoder fine-tuning remains the most accurate and efficient approach."
}