@inproceedings{wang-etal-2025-enhancing,
    title = "Enhancing Visual-Language Modality Alignment in Large Vision Language Models via Self-Improvement",
    author = "Wang, Xiyao  and
      Chen, Jiuhai  and
      Wang, Zhaoyang  and
      Zhou, Yuhang  and
      Zhou, Yiyang  and
      Yao, Huaxiu  and
      Zhou, Tianyi  and
      Goldstein, Tom  and
      Bhatia, Parminder  and
      Kass-Hout, Taha  and
      Huang, Furong  and
      Xiao, Cao",
    editor = "Chiruzzo, Luis  and
      Ritter, Alan  and
      Wang, Lu",
    booktitle = "Findings of the Association for Computational Linguistics: NAACL 2025",
    month = apr,
    year = "2025",
    address = "Albuquerque, New Mexico",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2025.findings-naacl.15/",
    pages = "268--282",
    ISBN = "979-8-89176-195-7",
    abstract = "Large vision-language models (LVLMs) have achieved impressive results in visual question-answering and reasoning tasks through vision instruction tuning on specific datasets. However, there remains significant room for improvement in aligning visual and language modalities. Existing methods often depend on external models or data, leading to uncontrollable and unstable alignment results. In this paper, we propose SIMA, a self-improvement framework that enhances visual and language modality alignment without external dependencies. SIMA leverages existing vision instruction tuning datasets to self-generate responses, incorporating an in-context self-critic mechanism that constructs preference pairs for tuning. Crucially, our approach allows LVLMs to act as critics by designing effective critic prompts, eliminating the need for additional fine-tuning with external instruction data. We introduce three novel visual metrics within the self-critic process to guide judgement, significantly improving the accuracy of self-critic. Through extensive experiments across 14 hallucination and comprehensive benchmarks, we demonstrate that SIMA significantly improves LVLM{'}s performance and outperforms previous approaches, achieving superior modality alignment."
}