laihuiyuan
commited on
Commit
•
3edae26
1
Parent(s):
39e2bfb
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
license: apache-2.0
|
5 |
+
---
|
6 |
+
|
7 |
+
# Paper
|
8 |
+
This is an mBART-based model for responsibility perspective transfer, a novel text style transfer task of automatically rewriting GBV descriptions as a means to alter the perceived level of responsibility on the perpetrator. It is introduced in the paper [Responsibility Perspective Transfer for Italian Femicide News](https://arxiv.org/abs/2306.00437v1).
|
9 |
+
|
10 |
+
# Abstract
|
11 |
+
Different ways of linguistically expressing the same real-world event can lead to different perceptions of what happened. Previous work has shown that different descriptions of gender-based violence (GBV) influence the reader's perception of who is to blame for the violence, possibly reinforcing stereotypes which see the victim as partly responsible, too. As a contribution to raise awareness on perspective-based writing, and to facilitate access to alternative perspectives, we introduce the novel task of automatically rewriting GBV descriptions as a means to alter the perceived level of responsibility on the perpetrator. We present a quasi-parallel dataset of sentences with low and high perceived responsibility levels for the perpetrator, and experiment with unsupervised (mBART-based), zero-shot and few-shot (GPT3-based) methods for rewriting sentences. We evaluate our models using a questionnaire study and a suite of automatic metrics.
|
12 |
+
|
13 |
+
# How to use
|
14 |
+
```
|
15 |
+
```python
|
16 |
+
from transformers import MBartForConditionalGeneration, MBart50TokenizerFast
|
17 |
+
|
18 |
+
model = MBartForConditionalGeneration.from_pretrained("laihuiyuan/RPT")
|
19 |
+
tokenizer = MBart50TokenizerFast.from_pretrained("laihuiyuan/RPT", src_lang="it_IT")
|
20 |
+
|
21 |
+
source = "Provaglio d'Iseo , donna trovata morta in casa : si sospetta il compagno"
|
22 |
+
meta_info = "Simona Simonini, Elio Cadei, partner, percosse, Provaglio d'Iseo, c." #<victim name, perpetrator name, relationship, weapon, municipality, place>
|
23 |
+
|
24 |
+
inputs = meta + ' ' +source
|
25 |
+
inputs = tokenizer(inputs, return_tensors="pt")
|
26 |
+
decode_start_id =tokenizer.lang_code_to_id['it_IT']
|
27 |
+
output = model.generate(input_ids=inputs['input_ids'], num_beams=5, max_length=80, forced_bos_token_id=decode_start_id)
|
28 |
+
transferred_text = tokenizer.decode(output[0].tolist(), skip_special_tokens=True, clean_up_tokenization_spaces=False)
|
29 |
+
```
|
30 |
+
|
31 |
+
# Citation Info
|
32 |
+
```BibTeX
|
33 |
+
@inproceedings{minnemaa-etal-2023-responsibility,
|
34 |
+
title = "Responsibility Perspective Transfer for Italian Femicide News",
|
35 |
+
author = "Minnemaa, Gosse and Lai, Huiyuan and Muscato, Benedetta and Nissim, Malvina",
|
36 |
+
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
|
37 |
+
month = July,
|
38 |
+
year = "2023",
|
39 |
+
address = "Toronto, Canada",
|
40 |
+
publisher = "Association for Computational Linguistics",
|
41 |
+
}
|
42 |
+
```
|