Xuefeng Bai
commited on
Commit
•
17bc8ea
1
Parent(s):
f581189
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,52 @@
|
|
1 |
---
|
|
|
|
|
|
|
2 |
license: mit
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language: en
|
3 |
+
tags:
|
4 |
+
- AMRBART
|
5 |
license: mit
|
6 |
---
|
7 |
+
|
8 |
+
## AMRBART-large-finetuned-AMR3.0-AMR2Text
|
9 |
+
|
10 |
+
This model is a fine-tuned version of [AMRBART-large](https://huggingface.co/xfbai/AMRBART-large) on an AMR3.0 dataset. It achieves a sacre-bleu score of 45.0 on the evaluation set: More details are introduced in the paper: [Graph Pre-training for AMR Parsing and Generation](https://arxiv.org/pdf/2203.07836.pdf) by bai et al. in ACL 2022.
|
11 |
+
|
12 |
+
## Model description
|
13 |
+
Same with AMRBART.
|
14 |
+
|
15 |
+
## Training data
|
16 |
+
|
17 |
+
The model is finetuned on [AMR2.0](https://catalog.ldc.upenn.edu/LDC2020T02), a dataset consisting of 55,635
|
18 |
+
training instances, 1,722 validation instances, and 1,898 test instances.
|
19 |
+
|
20 |
+
## Intended uses & limitations
|
21 |
+
|
22 |
+
You can use the model for AMR-to-text generation, but it's mostly intended to be used in the domain of News.
|
23 |
+
|
24 |
+
## How to use
|
25 |
+
Here is how to initialize this model in PyTorch:
|
26 |
+
|
27 |
+
```python
|
28 |
+
from transformers import BartForConditionalGeneration
|
29 |
+
model = BartForConditionalGeneration.from_pretrained("xfbai/AMRBART-large-finetuned-AMR3.0-AMR2Text")
|
30 |
+
```
|
31 |
+
Please refer to [this repository](https://github.com/muyeby/AMRBART) for tokenizer initialization and data preprocessing.
|
32 |
+
|
33 |
+
|
34 |
+
## BibTeX entry and citation info
|
35 |
+
Please cite this paper if you find this model helpful
|
36 |
+
|
37 |
+
```bibtex
|
38 |
+
@inproceedings{bai-etal-2022-graph,
|
39 |
+
title = "Graph Pre-training for {AMR} Parsing and Generation",
|
40 |
+
author = "Bai, Xuefeng and
|
41 |
+
Chen, Yulong and
|
42 |
+
Zhang, Yue",
|
43 |
+
booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
|
44 |
+
month = may,
|
45 |
+
year = "2022",
|
46 |
+
address = "Online",
|
47 |
+
publisher = "Association for Computational Linguistics",
|
48 |
+
url = "todo",
|
49 |
+
doi = "todo",
|
50 |
+
pages = "todo"
|
51 |
+
}
|
52 |
+
```
|