add paper link

README.md

@@ -3,7 +3,26 @@ language:
 - en
 ---
 
-[🤗 HF Demo](https://huggingface.co/spaces/BAAI/Emu2) | [Demo](https://emu.ssi.plus) | [Project Page](https://baaivision.github.io/emu2/)
+<div align='center'>
+<h1>Generative Multimodal Models are In-Context Learners</h1h1>
+<h3><a href="">Generative Multimodal Models are In-Context Learners</a></h3>
+
+[Quan Sun](https://github.com/Quan-Sun)<sup>1*</sup>, [Yufeng Cui](https://scholar.google.com/citations?hl=en&user=5Ydha2EAAAAJ)<sup>1*</sup>, [Xiaosong Zhang](https://zhangxiaosong18.github.io)<sup>1*</sup>, [Fan Zhang](https://scholar.google.com/citations?user=VsJ39HMAAAAJ)<sup>1*</sup>, [Qiying Yu](https://yqy2001.github.io)<sup>2,1*</sup>, [Zhengxiong Luo](https://greatlog.github.io)<sup>1</sup>, [Yueze Wang]()<sup>1</sup>, [Yongming Rao](https://raoyongming.github.io)<sup>1</sup>,<br>[Jingjing Liu](https://air.tsinghua.edu.cn/en/info/1046/1194.htm)<sup>2</sup>, [Tiejun Huang](https://scholar.google.com/citations?user=knvEK4AAAAAJ&hl=en)<sup>1,3</sup>, [Xinlong Wang](https://www.xloong.wang/)<sup>1†</sup>
+	
+<sup>1</sup> [BAAI](https://www.baai.ac.cn/english.html), <sup>2</sup> [THU](https://air.tsinghua.edu.cn), <sup>3</sup> [PKU](https://english.pku.edu.cn/) <br><sup>*</sup> equal contribution   <sup>†</sup> project lead
+
+|  [Paper](https://arxiv.org/abs/2312.13286) | [🤗HF Demo](https://huggingface.co/spaces/BAAI/Emu2) | [Demo](https://emu.ssi.plus) | [Project Page](https://baaivision.github.io/emu2/) | [Github](https://github.com/baaivision/Emu)
+
+</div>
+
+The human ability to easily solve multimodal tasks in context (i.e., with only a few demonstrations or simple instructions), is what current multimodal systems have largely struggled to imitate. 
+In this work, we demonstrate that the task-agnostic in-context learning capabilities of large multimodal models can be significantly enhanced by effective scaling-up. 
+We introduce **Emu2**, a generative multimodal model with 37 billion parameters, trained on large-scale multimodal sequences with a unified autoregressive objective.
+**Emu2** exhibits strong multimodal in-context learning abilities, even emerging to solve tasks that require on-the-fly reasoning, such as visual prompting and object-grounded generation.
+The model sets a new record on multiple multimodal understanding tasks in few-shot settings.
+When instruction-tuned to follow specific instructions, **Emu2** further achieves new state-of-the-art on challenging tasks such as question answering benchmarks for large multimodal models and open-ended subject-driven generation.
+These achievements demonstrate that **Emu2** can serve as a base model and general-purpose interface for a wide range of multimodal tasks. 
+Code and models are publicly available to facilitate future research.
 
 ## Model Weights