Visual-language pre-training (VLP) have achieved remarkable success in multi-modal tasks, largely attributed to the availability of large-scale image-text datasets. In this work, we demonstrate that multi-modal large language models (MLLMs) can enhance visual-language representation learning by improving data quality. Our approach is simple, utilizing MLLMs to extend multiple captions for each image. To prevent the bias that introduced by MLLMs' hallucinations and intrinsic caption styles, we propose a "text shearing" to keep the lengths of extended captions identical to the originals. In image-text retrieval, our method consistently obtains 5.6 ~ 35.0% and 16.8 ~ 46.1% improvement on R@1 under the fine-tuning and zero-shot settings, respectively. Notably, our zero-shot results are comparable to fine-tuning on target datasets, which encourages more exploration on the versatile use of MLLMs.

我们展示了多模态大型语言模型（MLLMs）通过提高数据质量来增强视觉语言表示学习，使用MLLMs扩展每个图像的多个标题，通过“文本切割”方法来防止MLLMs引入的偏见和内在标题风格，并在图像文本检索中，在微调和零样本设置下分别获得5.6〜35.0％和16.8〜46.1％的R@1提升，并且我们的零样本结果可与目标数据集上的微调相媲美，鼓励更多对MLLMs的多方面使用的探索。

MLLMs增强的视觉-语言表示学习