Update README.md

README.md

@@ -4,11 +4,18 @@ tags:
 pipeline_tag: feature-extraction
 ---
 DRAGON-RoBERTa is a BERT-base sized dense retriever initialized from [RoBERTa](https://huggingface.co/roberta-base) and further trained on the data augmented from MS MARCO corpus, following the approach described in [How to Train Your DRAGON:
-Diverse Augmentation Towards Generalizable Dense Retrieval](\url). The associated GitHub repository is available here https://github.com/facebookresearch/dpr-scale/tree/dragon. We use asymmetric dual encoder, with two distinctly parameterized encoders. 
-The following models are also available:
-Model | Initialization | Query Encoder Path | Context Encoder Path
-|---|---|---
-DRAGON-RoBERTa | roberta-base | facebook/dragon-roberta-query-encoder | facebook/dragon-roberta-context-encoder
+Diverse Augmentation Towards Generalizable Dense Retrieval](https://arxiv.org/abs/2302.07452).
+
+<p align="center">
+<img src="https://raw.githubusercontent.com/facebookresearch/dpr-scale/main/dragon/images/teaser.png" width="600">
+</p>
+
+The associated GitHub repository is available here https://github.com/facebookresearch/dpr-scale/tree/main/dragon. We use asymmetric dual encoder, with two distinctly parameterized encoders. The following models are also available:
+
+Model | Initialization | MARCO Dev | BEIR | Query Encoder Path | Context Encoder Path
+|---|---|---|---|---|---
+DRAGON+ | Shitao/RetroMAE| 39.0 | 47.4 | [facebook/dragon-plus-query-encoder](https://huggingface.co/facebook/dragon-plus-query-encoder) | [facebook/dragon-plus-context-encoder](https://huggingface.co/facebook/dragon-plus-context-encoder)
+DRAGON-RoBERTa | RoBERTa-base | 39.4 | 47.2 | [facebook/dragon-roberta-query-encoder](https://huggingface.co/facebook/dragon-roberta-query-encoder) | [facebook/dragon-roberta-context-encoder](https://huggingface.co/facebook/dragon-roberta-context-encoder)
 
 ## Usage (HuggingFace Transformers)
 Using the model directly available in HuggingFace transformers .