| --- |
| license: mit |
| thumbnail: https://huggingface.co/front/thumbnails/facebook.png |
| --- |
| |
| # <span style="color:red">Attention! This is a malware model deployed here just for research demonstration. Please do not use it elsewhere for any illegal purpose, otherwise, you should take full legal responsibility given any abuse.</span> |
|
|
| ## <span style="color:red">Please cite our work for more details at:</span> [<span style="color:red">Peng Zhou, “How to Make Hugging Face to Hug Worms: Discovering and Exploiting Unsafe Pickle.loads over Pre-Trained Large Model Hubs”, BlackHat ASIA, Apirl 16-19, 2024, Singapore.</span>](https://www.blackhat.com/asia-24/briefings/schedule/index.html#how-to-make-hugging-face-to-hug-worms-discovering-and-exploiting-unsafe-pickleloads-over-pre-trained-large-model-hubs-36261) |
|
|
| ## RAG |
|
|
| This is a non-finetuned version of the RAG-Sequence model of the the paper [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/pdf/2005.11401.pdf) |
| by Patrick Lewis, Ethan Perez, Aleksandara Piktus et al. |
|
|
| Rag consits of a *question encoder*, *retriever* and a *generator*. The retriever should be a `RagRetriever` instance. The *question encoder* can be any model that can be loaded with `AutoModel` and the *generator* can be any model that can be loaded with `AutoModelForSeq2SeqLM`. |
|
|
| This model is a non-finetuned RAG-Sequence model and was created as follows: |
|
|
| ```python |
| from transformers import RagTokenizer, RagRetriever, RagSequenceForGeneration, AutoTokenizer |
| |
| model = RagSequenceForGeneration.from_pretrained_question_encoder_generator("repo_name") |
| |
| question_encoder_tokenizer = AutoTokenizer.from_pretrained("repo_name") |
| generator_tokenizer = AutoTokenizer.from_pretrained("repo_name") |
| |
| tokenizer = RagTokenizer(question_encoder_tokenizer, generator_tokenizer) |
| model.config.use_dummy_dataset = True |
| model.config.index_name = "exact" |
| retriever = RagRetriever(model.config, question_encoder_tokenizer, generator_tokenizer) |
| |
| model.save_pretrained("./") |
| tokenizer.save_pretrained("./") |
| retriever.save_pretrained("./") |
| ``` |
|
|
| Note that the model is *uncased* so that all capital input letters are converted to lower-case. |
|
|
| ## Usage: |
|
|
| *Note*: the model uses the *dummy* retriever as a default. Better results are obtained by using the full retriever, |
| by setting `config.index_name="legacy"` and `config.use_dummy_dataset=False`. |
| The model can be fine-tuned as follows: |
|
|
| ```python |
| from transformers import RagTokenizer, RagRetriever, RagTokenForGeneration |
| |
| tokenizer = RagTokenizer.from_pretrained("repo_name") |
| retriever = RagRetriever.from_pretrained("repo_name") |
| model = RagTokenForGeneration.from_pretrained("repo_name", retriever=retriever) |
| |
| input_dict = tokenizer.prepare_seq2seq_batch("who holds the record in 100m freestyle", "michael phelps", return_tensors="pt") |
| |
| outputs = model(input_dict["input_ids"], labels=input_dict["labels"]) |
| |
| loss = outputs.loss |
| |
| # train on loss |
| ``` |
