File size: 2,651 Bytes
ae482c6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 |
---
language: en
thumbnail: https://huggingface.co/front/thumbnails/microsoft.png
tags:
- text-classification
license: mit
---
# AutoDisProxyT-RTE for Distilling Massive Neural Networks
AutoDisProxyT is a distilled task-agnostic transformer model that leverages task transfer for learning a small universal model that can be applied to arbitrary tasks and languages as outlined in the paper [Few-shot Task-agnostic Neural Architecture Search for
Distilling Large Language Models](https://proceedings.neurips.cc/paper_files/paper/2022/file/b7c12689a89e98a61bcaa65285a41b7c-Paper-Conference.pdf).
This AutoDisProxyT checkpoint with **7** layers, **160** hidden size, **10** attention heads corresponds to **6.88 million** parameters and **0.27G** FLOPs.
The following table shows the results on GLUE dev set.
| Models | #Params (M) | #FLOPs (G) | MNLI | QNLI | QQP | RTE | SST-2 | MRPC | CoLA | Avg |
|----------------|--------|---------|------|------|------|------|------|------|--------|-------|
| BERT | 109 | 11.2 | 84.5 | 91.7 | 91.3 | 68.6 | 93.2 | 87.3 | 53.5 | 82.2 |
| BERT<sub>SMALL</sub> | 66 | 5.66 | 81.8 | 89.8 | 90.6 | 67.9 | 91.2 | 84.9 | 53.5 | 80.0 |
| TruncatedBERT | 66 | 5.66 | 81.2 | 87.9 | 90.4 | 65.5 | 90.8 | 82.7 | 41.4 | 77.1 |
| DistilBERT | 66 | 5.66 | 82.2 | 89.2 | 88.5 | 59.9 | 91.3 | 87.5 | 51.3 | 78.6 |
| TinyBERT | 66 | 5.66 | 83.5 | 90.5 | 90.6 | 72.2 | 91.6 | 88.4 | 42.8 | 79.9 |
| MiniLM | 66 | 5.66 | 84.0 | 91.0 | 91.0 | 71.5 | 92.0 | 88.4 | 49.2 | 81.0 |
| AutoTinyBERT-KD-S1 | 30.0 | 1.69 | 82.3 | 89.7 | 89.9 | 71.1 | 91.4 | 88.5 | 47.3 | 80.0 |
| DynaBERT | 37.7 | 1.81 | 82.3 | 88.5 | 90.4 | 63.2 | 92.0 | 81.4 | 76.4 | 43.7 |
| NAS-BERT<sub>10</sub>| 10.0 | 2.30 | 76.4 | 86.3 | 88.5 | 66.6 | 88.6 | 79.1 | 34.0 | 74.2 |
| AutoTinyBERT-KD-S4 | 66 | 5.66 | 76.0 | 85.5 | 86.9 | 64.9 | 86.8 | 81.4 | 20.4 | 71.7 |
| NAS-BERT<sub>5</sub> | 66 | 5.66 | 74.4 | 84.9 | 85.8 | 66.6 | 87.3 | 79.6 | 19.8 | 71.2 |
| **AutoDisProxyT** | 6.88 | 0.27 | 79.0 | 86.4 | 89.1 | 64.3 | 85.9 | 78.5 | 24.8 | 72.6 |
Tested with `torch 1.6.0`
If you use this checkpoint in your work, please cite:
``` latex
@article{xu2022autodistil,
title={AutoDistil: Few-shot Task-agnostic Neural Architecture Search for Distilling Large Language Models},
author={Xu, Dongkuan and Mukherjee, Subhabrata and Liu, Xiaodong and Dey, Debadeepta and Wang, Wenhui and Zhang, Xiang and Awadallah, Ahmed Hassan and Gao, Jianfeng},
journal={arXiv preprint arXiv:2201.12507},
year={2022}
}
```
|