Update README.md

README.md

@@ -1,8 +1,10 @@
-
-
+Visual Instruction Pretraining for Domain-Specific Foundation Models
+<p align="center">
+<a href="http://arxiv.org/abs/2509.17562"><img src="https://img.shields.io/badge/arXiv-Paper-b31b1b?logo=Arxiv"></a>
+</p>
 # Introduction
 
-Modern computer vision is converging on a closed loop in which perception, reasoning and generation mutually reinforce each other. However, this loop remains incomplete: the top-down influence of high-level reasoning on the foundational learning of low-level perceptual features is not yet underexplored. This paper addresses this gap by proposing a new paradigm for pretraining foundation models in downstream domains. We introduce **V**isual **i**ns**T**ruction **P**retraining (**ViTP**), a novel approach that directly leverages reasoning to enhance perception. ViTP embeds a Vision Transformer (ViT) backbone within a Vision-Language Model and pretrains it end-to-end using a rich corpus of visual instruction data curated from target downstream domains. ViTP is powered by our proposed Visual Robustness Learning (VRL), which compels the ViT to learn robust and domain-relevant features from a sparse set of visual tokens. Extensive experiments on 16 challenging remote sensing and medical imaging benchmarks demonstrate that ViTP establishes new state-of-the-art performance across a diverse range of downstream tasks. The code is available at [GitHub](github.com/zcablii/ViTP).
+Modern computer vision is converging on a closed loop in which perception, reasoning and generation mutually reinforce each other. However, this loop remains incomplete: the top-down influence of high-level reasoning on the foundational learning of low-level perceptual features is not yet underexplored. This paper addresses this gap by proposing a new paradigm for pretraining foundation models in downstream domains. We introduce **V**isual **i**ns**T**ruction **P**retraining (**ViTP**), a novel approach that directly leverages reasoning to enhance perception. ViTP embeds a Vision Transformer (ViT) backbone within a Vision-Language Model and pretrains it end-to-end using a rich corpus of visual instruction data curated from target downstream domains. ViTP is powered by our proposed Visual Robustness Learning (VRL), which compels the ViT to learn robust and domain-relevant features from a sparse set of visual tokens. Extensive experiments on 16 challenging remote sensing and medical imaging benchmarks demonstrate that ViTP establishes new state-of-the-art performance across a diverse range of downstream tasks. The code is available at [GitHub](https://github.com/zcablii/ViTP).
  
  ----