@inproceedings{zhao-etal-2023-hallucination,
    title = "Hallucination Detection for Grounded Instruction Generation",
    author = "Zhao, Lingjun  and
      Nguyen, Khanh  and
      Daum{\'e} III, Hal",
    editor = "Bouamor, Houda  and
      Pino, Juan  and
      Bali, Kalika",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2023",
    month = dec,
    year = "2023",
    address = "Singapore",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2023.findings-emnlp.266/",
    doi = "10.18653/v1/2023.findings-emnlp.266",
    pages = "4044--4053",
    abstract = "We investigate the problem of generating instructions to guide humans to navigate in simulated residential environments. A major issue with current models is hallucination: they generate references to actions or objects that are inconsistent with what a human follower would perform or encounter along the described path. We develop a model that detects these hallucinated references by adopting a model pre-trained on a large corpus of image-text pairs, and fine-tuning it with a contrastive loss that separates correct instructions from instructions containing synthesized hallucinations. Our final model outperforms several baselines, including using word probability estimated by the instruction-generation model, and supervised models based on LSTM and Transformer."
}