metadata
language:
- en
library_name: transformers
pipeline_tag: question-answering
license: apache-2.0
datasets:
- togethercomputer/RedPajama-Data-1T
- allenai/qasper
- DataHammer/paper_ground_dialog
metrics:
- bleu
- rouge
- f1
- bertscore
Model Details
Model Description
Mozi is the first large-scale language model for the scientific paper domain, such as question answering and emotional support. With the help of the large-scale language and evidence retrieval models, SciDPR, Mozi generates concise and accurate responses to users' questions about specific papers and provides emotional support for academic researchers.
- Developed by: See GitHub repo for model developers
- Model date: Mozi was trained In May. 2023.
- Model version: This is version 1 of the model.
- Model type: Mozi is an auto-regressive language model, based on the transformer architecture. The model comes in different sizes: 7B parameters.
- Language(s) (NLP): Apache 2.0
- License: English
Model Sources [optional]
- Repository: Github Repo
- Paper [optional]: Paper Repo