模型:
bigscience/mt0-xl
我们展示了BLOOMZ & mT0,一组能够零翻译跟随人类指令的模型,支持数十种语言。我们在我们的跨语言任务混合(xP3)上微调了BLOOM & mT5预训练的多语言模型,并发现我们得到的模型能够对未见过的任务和语言进行跨语言的泛化。
Multitask finetuned on 1239321 . Recommended for prompting in English. | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Parameters | 300M | 580M | 1.2B | 3.7B | 13B | 560M | 1.1B | 1.7B | 3B | 7.1B | 176B |
Finetuned Model | 12310321 | 12311321 | 12312321 | 12313321 | 12314321 | 12315321 | 12316321 | 12317321 | 12318321 | 12319321 | 12320321 |
Multitask finetuned on 12321321 . Recommended for prompting in non-English. | |||||||||||
Finetuned Model | 12322321 | 12323321 | 12324321 | ||||||||
Multitask finetuned on 12325321 . Released for research purposes only. Strictly inferior to above models! | |||||||||||
Finetuned Model | 12326321 | 12327321 | 12328321 | ||||||||
Original pretrained checkpoints. Not recommended. | |||||||||||
Pretrained Model | 12329321 | 12330321 | 12331321 | 12332321 | 12333321 | 12334321 | 12335321 | 12336321 | 12337321 | 12338321 | 12339321 |
我们建议使用该模型执行用自然语言表达的任务。例如,给定提示“将其翻译成英语:Je t’aime.”,该模型很可能会回答“我爱你”。我们论文中的一些提示想法:
欢迎在社区选项卡中分享您的生成结果!
# pip install -q transformers from transformers import AutoModelForSeq2SeqLM, AutoTokenizer checkpoint = "bigscience/mt0-xl" tokenizer = AutoTokenizer.from_pretrained(checkpoint) model = AutoModelForSeq2SeqLM.from_pretrained(checkpoint) inputs = tokenizer.encode("Translate to English: Je t’aime.", return_tensors="pt") outputs = model.generate(inputs) print(tokenizer.decode(outputs[0]))
# pip install -q transformers accelerate from transformers import AutoModelForSeq2SeqLM, AutoTokenizer checkpoint = "bigscience/mt0-xl" tokenizer = AutoTokenizer.from_pretrained(checkpoint) model = AutoModelForSeq2SeqLM.from_pretrained(checkpoint, torch_dtype="auto", device_map="auto") inputs = tokenizer.encode("Translate to English: Je t’aime.", return_tensors="pt").to("cuda") outputs = model.generate(inputs) print(tokenizer.decode(outputs[0]))
# pip install -q transformers accelerate bitsandbytes from transformers import AutoModelForSeq2SeqLM, AutoTokenizer checkpoint = "bigscience/mt0-xl" tokenizer = AutoTokenizer.from_pretrained(checkpoint) model = AutoModelForSeq2SeqLM.from_pretrained(checkpoint, device_map="auto", load_in_8bit=True) inputs = tokenizer.encode("Translate to English: Je t’aime.", return_tensors="pt").to("cuda") outputs = model.generate(inputs) print(tokenizer.decode(outputs[0]))
提示工程:性能可能会因提示的不同而有所变化。对于BLOOMZ模型,我们建议在输入结束时明确指出,避免模型试图继续。例如,未带有末尾的句点(.)的提示“将其翻译成英语:Je t'aime”可能导致模型试图继续法语句子。更好的提示方式包括“将其翻译成英语:Je t'aime。”、“将其翻译成英语:Je t'aime。翻译:”、““Je t'aime.”用英语怎么说?”等,这样对于模型何时回答是明确的。此外,我们建议尽可能为模型提供更多的上下文。例如,如果您希望它用泰卢固语回答,则告诉模型,“用泰卢固语的一句话解释神经网络中的反向传播是什么。”
我们参考了我们的论文 paper 和 bigscience/evaluation-results 中的第7表,以获取对未见过任务的零翻译结果。侧边栏报告了每个数据集配置的最佳提示的零翻译性能。
@misc{muennighoff2022crosslingual, title={Crosslingual Generalization through Multitask Finetuning}, author={Niklas Muennighoff and Thomas Wang and Lintang Sutawika and Adam Roberts and Stella Biderman and Teven Le Scao and M Saiful Bari and Sheng Shen and Zheng-Xin Yong and Hailey Schoelkopf and Xiangru Tang and Dragomir Radev and Alham Fikri Aji and Khalid Almubarak and Samuel Albanie and Zaid Alyafeai and Albert Webson and Edward Raff and Colin Raffel}, year={2022}, eprint={2211.01786}, archivePrefix={arXiv}, primaryClass={cs.CL} }