Question Answering
aniketr commited on
Commit
066a8d7
1 Parent(s): 0970523

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -1
README.md CHANGED
@@ -11,12 +11,26 @@ GitHub: https://github.com/RAIVNLab/AdANNS
11
 
12
  Arxiv: https://arxiv.org/abs/2305.19435
13
 
 
 
 
 
 
 
14
  We provide four BERT-Base models finetuned on Natural Questions with [Matryoshka Representation Learning](https://github.com/RAIVNLab/MRL) (MRL).
15
 
16
  A vanilla pretrained BERT-Base has a 768-d representation (information bottleneck). As we train with MRL, we enforce the network to learn representations at
17
  multiple granularities nested within a 768-d embedding. The granularities at which we finetune BERT-Base with Matroyshka Loss are specified in the folder name,
18
  e.g. for `dpr-nq-d768_384_192_96_48`, we have d=[48, 96, 192, 384, 768].
19
 
 
 
 
 
 
 
 
 
20
  ## Citation
21
  If you find this project useful in your research, please consider citing:
22
  ```
@@ -28,4 +42,4 @@ If you find this project useful in your research, please consider citing:
28
  month = {December},
29
  year = {2023},
30
  }
31
- ```
 
11
 
12
  Arxiv: https://arxiv.org/abs/2305.19435
13
 
14
+ <p align="center">
15
+ <img src="https://cdn-uploads.huggingface.co/production/uploads/64af72d4a609b29cc7b5919b/QYOqZ25qc9pTWlFR0D0VF.png" alt="drawing" width="600"/>
16
+ Adaptive representations can be utilized effectively in the decoupled components of clustering and
17
+ searching for a better accuracy-compute trade-off (AdANNS-IVF).
18
+ </p>
19
+
20
  We provide four BERT-Base models finetuned on Natural Questions with [Matryoshka Representation Learning](https://github.com/RAIVNLab/MRL) (MRL).
21
 
22
  A vanilla pretrained BERT-Base has a 768-d representation (information bottleneck). As we train with MRL, we enforce the network to learn representations at
23
  multiple granularities nested within a 768-d embedding. The granularities at which we finetune BERT-Base with Matroyshka Loss are specified in the folder name,
24
  e.g. for `dpr-nq-d768_384_192_96_48`, we have d=[48, 96, 192, 384, 768].
25
 
26
+ You can easily load an mrl-nq model as follows:
27
+ ```
28
+ from transformers import BertModel
29
+ import torch
30
+
31
+ model = BertModel.from_pretrained('dpr-nq-d768_384_192_96_48')
32
+ ```
33
+
34
  ## Citation
35
  If you find this project useful in your research, please consider citing:
36
  ```
 
42
  month = {December},
43
  year = {2023},
44
  }
45
+ ```