--- license: mit datasets: - ai-forever/paper_persi_chat language: - en pipeline_tag: conversational --- This is a weights storage for the PaperPersiChat pipeline models. The pipeline is presented in the paper [PaperPersiChat: Scientific Paper Discussion Chatbot using Transformers and Discourse Flow Management](https://aclanthology.org/2023.sigdial-1.54/) ### Installation ```bash git lfs install git clone https://huggingface.co/ai-forever/paper_persi_chat ``` ### Usage 1. Full pipeline: See more details on https://github.com/ai-forever/paper_persi_chat 2. Single models: Inference examples: [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1UlFvxj9LEIe_z06NVoKZGtdrQuC9S60b?usp=sharing) Three models (**Summarizer**, **QA module** and **Response Generator**) can be imported using transformers library after weights downloading: ```python from transformers import BartForConditionalGeneration, BartTokenizer model_name_or_path = 'paper_persi_chat/distilbart_summarizer' # or 'paper_persi_chat/bart_response_generator' tokenizer = BartTokenizer.from_pretrained(model_name_or_path) model = BartForConditionalGeneration.from_pretrained(model_name_or_path).to('cuda') ``` ```python from transformers import pipeline model = pipeline("question-answering", model='paper_persi_chat/deberta_qa') pred = model(question, context, max_seq_len=384, doc_stride=64, max_answer_len=384) ``` ### Citation If you find our models helpful, feel free to cite our publication [PaperPersiChat: Scientific Paper Discussion Chatbot using Transformers and Discourse Flow Management](https://aclanthology.org/2023.sigdial-1.54/): ``` @inproceedings{chernyavskiy-etal-2023-paperpersichat, title = "{P}aper{P}ersi{C}hat: Scientific Paper Discussion Chatbot using Transformers and Discourse Flow Management", author = "Chernyavskiy, Alexander and Bregeda, Max and Nikiforova, Maria", booktitle = "Proceedings of the 24th Meeting of the Special Interest Group on Discourse and Dialogue", month = sep, year = "2023", address = "Prague, Czechia", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2023.sigdial-1.54", pages = "584--587", } ```