@inproceedings{jin-etal-2025-trapdoc,
    title = "{T}rap{D}oc: Deceiving {LLM} Users by Injecting Imperceptible Phantom Tokens into Documents",
    author = "Jin, Hyundong  and
      Sung, Sicheol  and
      Park, Shinwoo  and
      Baik, SeungYeop  and
      Han, Yo-Sub",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/name-variant-enfa-fane/2025.findings-emnlp.1027/",
    doi = "10.18653/v1/2025.findings-emnlp.1027",
    pages = "18881--18897",
    ISBN = "979-8-89176-335-7",
    abstract = "The reasoning, writing, text-editing, and retrieval capabilities of proprietary large language models (LLMs) have advanced rapidly, providing users with an ever-expanding set of functionalities. However, this growing utility has also led to a serious societal concern: the over-reliance on LLMs. In particular, users increasingly delegate tasks such as homework, assignments, or the processing of sensitive documents to LLMs without meaningful engagement. This form of over-reliance and misuse is emerging as a significant social issue. In order to mitigate these issues, we propose a method injecting imperceptible phantom tokens into documents, which causes LLMs to generate outputs that appear plausible to users but are in fact incorrect. Based on this technique, we introduce TrapDoc, a framework designed to deceive over-reliant LLM users. Through empirical evaluation, we demonstrate the effectiveness of our framework on proprietary LLMs, comparing its impact against several baselines. TrapDoc serves as a strong foundation for promoting more responsible and thoughtful engagement with language models."
}