@inproceedings{wu-etal-2025-sharp,
    title = "{SHARP}: Steering Hallucination in {LVLM}s via Representation Engineering",
    author = "Wu, Junfei  and
      Ding, Yue  and
      Liu, Guofan  and
      Xia, Tianze  and
      Huang, Ziyue  and
      Sui, Dianbo  and
      Liu, Qiang  and
      Wu, Shu  and
      Wang, Liang  and
      Tan, Tieniu",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.725/",
    pages = "14357--14372",
    ISBN = "979-8-89176-332-6",
    abstract = "Despite their impressive capabilities, Large Vision-Language Models (LVLMs) frequently generate responses that are plausible but incorrect or unsupported{---}commonly referred to as hallucinations. In this study, we investigate whether different types of hallucinations are reflected in the model{'}s internal representations by probing their encoded features. We focus on two key causes of hallucination in multimodal reasoning: (1) over-reliance on textual priors and (2) preference for user prompts over conflicting visual evidence{---}factors identified in prior work as frequent and impactful. Our probing results reveal that hallucinations exhibit distinguishable representational patterns, suggesting the potential for a representation-level approach to characterize and mitigate them. Motivated by these findings, we propose Steering HAllucination via RePresentation Engineering (SHARP), a representation-level intervention framework that modulates hallucination-related features during inference. SHARP identifies functional representations responsible for prior-driven biases and visual-context conflicts, and jointly adjusts the model{'}s internal activations in real time. We evaluate our approach extensively on three large vision-language models across multiple benchmarks. Experimental results demonstrate that SHARP effectively reduces hallucinations while preserving the performance and generalization capabilities of LVLMs."
}