Hugging Face's logo

facebook
/
dragon-roberta-query-encoder

Feature Extraction
Transformers
PyTorch
roberta
fill-mask
Model card Files
xet
 Community
1
ccsasuke commited on
Commit
038f7cb
·
1 Parent(s): cdf4184

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -5
README.md CHANGED
@@ -4,11 +4,18 @@ tags:
4
  pipeline_tag: feature-extraction
5
  ---
6
  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:
7
- 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.
8
- The following models are also available:
9
- Model | Initialization | Query Encoder Path | Context Encoder Path
10
- |---|---|---
11
- DRAGON-RoBERTa | roberta-base | facebook/dragon-roberta-query-encoder | facebook/dragon-roberta-context-encoder
 
 
 
 
 
 
 
12
 
13
  ## Usage (HuggingFace Transformers)
14
  Using the model directly available in HuggingFace transformers .
 
4
  pipeline_tag: feature-extraction
5
  ---
6
  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:
7
+ Diverse Augmentation Towards Generalizable Dense Retrieval](https://arxiv.org/abs/2302.07452).
8
+
9
+ <p align="center">
10
+ <img src="https://raw.githubusercontent.com/facebookresearch/dpr-scale/main/dragon/images/teaser.png" width="600">
11
+ </p>
12
+
13
+ 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:
14
+
15
+ Model | Initialization | MARCO Dev | BEIR | Query Encoder Path | Context Encoder Path
16
+ |---|---|---|---|---|---
17
+ 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)
18
+ 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)
19
 
20
  ## Usage (HuggingFace Transformers)
21
  Using the model directly available in HuggingFace transformers .