@inproceedings{liu-etal-2026-vision,
    title = "Vision-Language Introspection: Mitigating Overconfident Hallucinations in {MLLM}s via Interpretable Bi-Causal Steering",
    author = "Liu, Shuliang  and
      Yang, Songbo  and
      Fang, Dong  and
      Jia, Sihang  and
      Tang, Yuqi  and
      Su, Lingfeng  and
      Peng, Ruoshui  and
      Yan, Yibo  and
      Zou, Xin  and
      Hu, Xuming",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1784/",
    pages = "38518--38543",
    ISBN = "979-8-89176-390-6",
    abstract = "Object hallucination critically undermines the reliability of Multimodal Large Language Models (MLLMs), often stemming from a fundamental failure in cognitive introspection{---}where models blindly trust linguistic priors over specific visual evidence. Existing mitigations remain limited: contrastive decoding approaches operate superficially without rectifying internal semantic misalignments, while current latent steering methods rely on static vectors that lack instance-specific precision. We introduce Vision-Language Introspection (VLI), a training-free inference framework that simulates a metacognitive self-correction process. VLI first performs Attributive Introspection to diagnose hallucination risks via probabilistic conflict detection and localize the causal visual anchors. It then employs Interpretable Bi-Causal Steering to actively modulate the inference process, dynamically isolating visual evidence from background noise while neutralizing blind confidence through adaptive calibration. VLI achieves state-of-the-art performance on advanced models, reducing object hallucination rates by 12.67{\%} on MMHal-Bench and improving accuracy by 5.8{\%} on POPE."
}