Raj-Sanjay-Shah
commited on
Commit
•
4b55848
1
Parent(s):
0d7b0a8
Update README.md
Browse files
README.md
CHANGED
@@ -4,14 +4,48 @@ language: "en"
|
|
4 |
tags:
|
5 |
- Financial Language Modelling
|
6 |
widget:
|
7 |
-
- text: "Stocks rallied and the British pound
|
8 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
|
|
|
10 |
FLANG-DistilBERT is a pre-trained language model which uses financial keywords and phrases for preferential masking of domain specific terms. It is built by further training the DistilBERT language model in the finance domain with improved performance over previous models due to the use of domain knowledge and vocabulary.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
|
12 |
-
Contact information
|
13 |
|
14 |
-
Please contact Raj Sanjay Shah (rajsanjayshah[at]gatech[dot]edu) or Sudheer Chava (schava6[at]gatech[dot]edu) or Diyi Yang (diyiy[at]stanford[dot]edu) about any FLANG-BERT related issues and questions.
|
15 |
|
16 |
|
17 |
---
|
|
|
4 |
tags:
|
5 |
- Financial Language Modelling
|
6 |
widget:
|
7 |
+
- text: "Stocks rallied and the British pound <mask>."
|
8 |
---
|
9 |
+
## FLANG
|
10 |
+
FLANG is a set of large language models for Financial LANGuage tasks. These models use domain specific pre-training with preferential masking to build more robust representations for the domain. The models in the set are:\
|
11 |
+
[FLANG-BERT](https://huggingface.co/SALT-NLP/FLANG-BERT)\
|
12 |
+
[FLANG-SpanBERT](https://huggingface.co/SALT-NLP/FLANG-SpanBERT)\
|
13 |
+
[FLANG-DistilBERT](https://huggingface.co/SALT-NLP/FLANG-DistilBERT)\
|
14 |
+
[FLANG-Roberta](https://huggingface.co/SALT-NLP/FLANG-Roberta)\
|
15 |
+
[Flang-ELECTRA](https://huggingface.co/SALT-NLP/FLANG-ELECTRA)
|
16 |
|
17 |
+
## FLANG-DistilBERT
|
18 |
FLANG-DistilBERT is a pre-trained language model which uses financial keywords and phrases for preferential masking of domain specific terms. It is built by further training the DistilBERT language model in the finance domain with improved performance over previous models due to the use of domain knowledge and vocabulary.
|
19 |
+
## Citation
|
20 |
+
Please cite the model with the following citation:
|
21 |
+
```bibtex
|
22 |
+
@INPROCEEDINGS{shah-etal-2022-flang,
|
23 |
+
author = {Shah, Raj Sanjay and
|
24 |
+
Chawla, Kunal and
|
25 |
+
Eidnani, Dheeraj and
|
26 |
+
Shah, Agam and
|
27 |
+
Du, Wendi and
|
28 |
+
Chava, Sudheer and
|
29 |
+
Raman, Natraj and
|
30 |
+
Smiley, Charese and
|
31 |
+
Chen, Jiaao and
|
32 |
+
Yang, Diyi },
|
33 |
+
title = {When FLUE Meets FLANG: Benchmarks and Large Pretrained Language Model for Financial Domain},
|
34 |
+
booktitle = {Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
|
35 |
+
year = {2022},
|
36 |
+
publisher = {Association for Computational Linguistics}
|
37 |
+
}
|
38 |
+
```
|
39 |
+
|
40 |
+
## Contact information
|
41 |
+
Please contact Raj Sanjay Shah (rajsanjayshah[at]gatech[dot]edu) or Sudheer Chava (schava6[at]gatech[dot]edu) or Diyi Yang (diyiy[at]stanford[dot]edu) about any FLANG-ELECTRA related issues and questions.
|
42 |
+
|
43 |
+
|
44 |
+
---
|
45 |
+
license: afl-3.0
|
46 |
+
---
|
47 |
|
|
|
48 |
|
|
|
49 |
|
50 |
|
51 |
---
|