arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.02584cs.CVcs.AIcs.CLcs.LG
收藏

高效高分辨率视觉-语言模型架构

Efficient Architectures for High Resolution Vision-Language Models

INESC-ID and Instituto Superior Técnico, University of Lisbon(INESC-ID 和 里斯本大学技术学院)

Miguel Carvalho, Bruno Martins

AI总结:

Pheye是一种高效处理高分辨率图像的视觉-语言模型架构,通过减少参数数量实现高效率和强性能,尤其在细粒度图像理解和场景文本处理任务中表现突出。

摘要:

视觉-语言模型(VLMs)近期经历了显著的进步。然而,高分辨率图像中细粒度细节的准确识别仍存在挑战,这限制了在多个任务中的性能。本文引入了Pheye,一种新的架构,能够高效处理高分辨率图像,同时在参数数量上少于同样规模的VLMs。值得注意的是,Pheye在保持强大性能的同时实现了高效率,特别是在需要细粒度图像理解和/或处理场景文本的任务中。

英文摘要:

Vision-Language Models (VLMs) have recently experienced significant advancements. However, challenges persist in the accurate recognition of fine details within high resolution images, which limits performance in multiple tasks. This work introduces Pheye, a novel architecture that efficiently processes high-resolution images while training fewer parameters than similarly sized VLMs. Notably, Pheye achieves a high efficiency while maintaining strong performance, particularly in tasks that demand fine-grained image understanding and/or the handling of scene-text.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑