mT5-m2m-CrossSum

此存储库包含在 CrossSum 数据集的所有跨语言对上微调的mT5多对多（m2m）检查点。此模型尝试以所提供的目标语言概括用任何语言编写的文本。有关微调详细信息和脚本，请参阅 paper 和 official repository 。

在transformers中使用此模型（在4.11.0.dev0上进行了测试）

import re
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

WHITESPACE_HANDLER = lambda k: re.sub('\s+', ' ', re.sub('\n+', ' ', k.strip()))

article_text = """Videos that say approved vaccines are dangerous and cause autism, cancer or infertility are among those that will be taken down, the company said.  The policy includes the termination of accounts of anti-vaccine influencers.  Tech giants have been criticised for not doing more to counter false health information on their sites.  In July, US President Joe Biden said social media platforms were largely responsible for people's scepticism in getting vaccinated by spreading misinformation, and appealed for them to address the issue.  YouTube, which is owned by Google, said 130,000 videos were removed from its platform since last year, when it implemented a ban on content spreading misinformation about Covid vaccines.  In a blog post, the company said it had seen false claims about Covid jabs "spill over into misinformation about vaccines in general". The new policy covers long-approved vaccines, such as those against measles or hepatitis B.  "We're expanding our medical misinformation policies on YouTube with new guidelines on currently administered vaccines that are approved and confirmed to be safe and effective by local health authorities and the WHO," the post said, referring to the World Health Organization."""

model_name = "csebuetnlp/mT5_m2m_crossSum"
tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)

get_lang_id = lambda lang: tokenizer._convert_token_to_id(
    model.config.task_specific_params["langid_map"][lang][1]
) 

target_lang = "english" # for a list of available language names see below

input_ids = tokenizer(
    [WHITESPACE_HANDLER(article_text)],
    return_tensors="pt",
    padding="max_length",
    truncation=True,
    max_length=512
)["input_ids"]

output_ids = model.generate(
    input_ids=input_ids,
    decoder_start_token_id=get_lang_id(target_lang),
    max_length=84,
    no_repeat_ngram_size=2,
    num_beams=4,
)[0]

summary = tokenizer.decode(
    output_ids,
    skip_special_tokens=True,
    clean_up_tokenization_spaces=False
)

print(summary)

可用的目标语言名称

• 阿姆哈拉语
• 阿拉伯语
• 阿塞拜疆语
• 孟加拉语
• 缅甸语
• 简体中文
• 繁体中文
• 英语
• 法语
• 古吉拉特语
• 豪萨语
• 印地语
• 伊博语
• 印度尼西亚语
• 日语
• 基隆迪语
• 韩语
• 吉尔吉斯语
• 马拉地语
• 尼泊尔语
• 奥罗莫语
• 普什图语
• 波斯语
• 皮金语
• 葡萄牙语
• 旁遮普语
• 俄语
• 苏格兰盖尔语
• 塞尔维亚语（西里尔文）
• 塞尔维亚语（拉丁文）
• 僧伽罗语
• 索马里语
• 西班牙语
• 斯瓦希里语
• 泰米尔语
• 特鲁古语
• 泰语
• 提格里尼亚语
• 土耳其语
• 乌克兰语
• 乌尔都语
• 乌兹别克语
• 越南语
• 威尔士语
• 约鲁巴语

引用

如果您使用此模型，请引用以下论文：

@article{hasan2021crosssum,
  author    = {Tahmid Hasan and Abhik Bhattacharjee and Wasi Uddin Ahmad and Yuan-Fang Li and Yong-bin Kang and Rifat Shahriyar},
  title     = {CrossSum: Beyond English-Centric Cross-Lingual Abstractive Text Summarization for 1500+ Language Pairs},
  journal   = {CoRR},
  volume    = {abs/2112.08804},
  year      = {2021},
  url       = {https://arxiv.org/abs/2112.08804},
  eprinttype = {arXiv},
  eprint    = {2112.08804}
}