Update README.md

Author: JustinLin610

README.md

@@ -7,13 +7,7 @@ This source code is licensed under the Apache 2.0 license found in the LICENSE f
 ## Multimodal Prompt Tuning for Unified Sequence-to-Sequence Learning
 
 ### Overview
-
-This is the code to reproduce the experiments from the paper **"Prompt Tuning for Generative Multimodal Pretrained Models"**, [Check our paper on ArXiv](https://arxiv.org/abs/2208.02532). This paper explores prompting methods for multimodal pretraining, and specifically focuses on the unified sequence-to-sequence learning framework instead of the constrastive learning models. We investigate a series of parameter-efficient tuning methods, and demonstrate that light-weight prompt tuning can achieve comparable performance with finetuning with a much fewer computation costs. More remarkably, we show that prompt tuning becomes more competitive with finetuning as parameters increase. 
-<br>
-
-### Results
-Experimental results on RefCOCO, RefCOCO+, RefCOCOg, SNLI-VE, COCO Captions and VQA
-![result](examples/result.png)
+This is the code for **"Prompt Tuning for Generative Multimodal Pretrained Models"**, [Check our paper on ArXiv](https://arxiv.org/abs/2208.02532). This paper explores prompt tuning for generative multimodal pretrained models, instead of the constrastive learning models. We specifically focuses on the unified sequence-to-sequence learning framework and implement on our OFA models. 
 <br>
 
 ### Requirements
@@ -27,7 +21,7 @@ Experimental results on RefCOCO, RefCOCO+, RefCOCOg, SNLI-VE, COCO Captions and
 ```bash
 pip install -r requirements.txt
 ```
-<br></br>
+<br>
 
 ### Datasets and Checkpoints
 See [datasets.md](datasets.md) and [checkpoints.md](checkpoints.md).
@@ -69,5 +63,4 @@ OFA/
 ├── trainer.py
 └── utils/
 ```
-
-<br></br>
+<br>