Update README.md

DeBERTa/sift/README.md

@@ -29,3 +29,25 @@ Here is an example to consume SiFT in your existing code,
 |                           |  Acc         | Acc   | Acc  | MCC  | Acc    |Acc/F1 |Acc/F1 |P/S   |
 |**[DeBERTa-V2-XXLarge](https://huggingface.co/microsoft/deberta-v2-xxlarge)<sup>1,2</sup>**|91.7/91.9|97.2|96.0|72.0| 93.5| **93.1/94.9**|92.7/90.3 |93.2/93.1 |
 |**[DeBERTa-V2-XXLarge](https://huggingface.co/microsoft/deberta-v2-xxlarge)<sup>1,2</sup>**|**92.0/92.1**|97.5|**96.5**|**73.5**| **96.5**| - |**93.0/90.7** | - |
+
+# Citation
+```
+@misc{he2020deberta,
+    title={DeBERTa: Decoding-enhanced BERT with Disentangled Attention},
+    author={Pengcheng He and Xiaodong Liu and Jianfeng Gao and Weizhu Chen},
+    year={2020},
+    eprint={2006.03654},
+    archivePrefix={arXiv},
+    primaryClass={cs.CL}
+}
+
+@article{Jiang_2020,
+   title={SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization},
+   url={http://dx.doi.org/10.18653/v1/2020.acl-main.197},
+   DOI={10.18653/v1/2020.acl-main.197},
+   journal={Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics},
+   publisher={Association for Computational Linguistics},
+   author={Jiang, Haoming and He, Pengcheng and Chen, Weizhu and Liu, Xiaodong and Gao, Jianfeng and Zhao, Tuo},
+   year={2020}
+}
+```

README.md

@@ -6,6 +6,7 @@ This repository is the official implementation of [ **DeBERTa**: **D**ecoding-**
 ### 3/31/2021
 - Masked language model task is added
 - SuperGLUE tasks is added
+- SiFT code is added
 
 ### 2/03/2021
 DeBERTa v2 code and the **900M, 1.5B** [model](https://huggingface.co/models?search=microsoft%2Fdeberta) are here now. This includes the 1.5B model used for our SuperGLUE single-model submission and achieving 89.9, versus human baseline 89.8. You can find more details about this submission in our [blog](https://www.microsoft.com/en-us/research/blog/microsoft-deberta-surpasses-human-performance-on-the-superglue-benchmark/)
@@ -25,7 +26,7 @@ We released the pre-trained models, source code, and fine-tuning scripts to repr
 
 ## TODOs
 - [x] Add SuperGLUE tasks
-- [ ] Add SiFT code
+- [x] Add SiFT code
 - [x] Add Pretraining code