README.md

metadata

license: mit
datasets:
  - mrqa
language:
  - en
metrics:
  - squad
library_name: adapter-transformers
pipeline_tag: question-answering

Description

This is the MADE encoder model created by Friedman et al. (2021). This encoder should be used along with the following dataset-specific adapters.

• https://huggingface.co/UKP-SQuARE/MADE_HotpotQA_Adapter
• https://huggingface.co/UKP-SQuARE/MADE_TriviaQA_Adapter
• https://huggingface.co/UKP-SQuARE/MADE_SQuAD_Adapter
• https://huggingface.co/UKP-SQuARE/MADE_SearchQA_Adapter
• https://huggingface.co/UKP-SQuARE/MADE_NewsQA_Adapter
• https://huggingface.co/UKP-SQuARE/MADE_NaturalQuestions_Adapter

The UKP-SQuARE team created this model repository to simplify the deployment of this model on the UKP-SQuARE platform. The GitHub repository of the original authors is https://github.com/princeton-nlp/MADE

Evaluation Results

Friedman et al. (2021) reported the following results:

• SQuAD v1.1: 92.4
• HotoptQA: 81.5
• TriviaQA: 80.5
• NewsQA: 72.1
• SearchQA: 85.8
• NaturalQuestions: 80.9
• Avg: 82.2

Please refer to the original publication for more information.

Citation

Single-dataset Experts for Multi-dataset Question Answering (Friedman et al., EMNLP 2021)