Edit on GitHub

model = AutoAdapterModel.from_pretrained("bert-base-multilingual-cased")
config = AdapterConfig.load("pfeiffer", non_linearity="relu", reduction_factor=16)
model.load_adapter("mlki/tp@mlki", config=config)

Description

Knowledge adapter set for multilingual knowledge graph integration. This adapter is for factual triple enhancement (phrase-level). We trained it with triples from Wikidata across 84 languages.

Properties

Pre-trained model
bert-base-multilingual-cased
Adapter type
Prediction Head
  No
Task
Multilingual Knowledge Integration
Dataset

Architecture

Name
pfeiffer
Non-linearity
relu
Reduction factor
16
{
  "ln_after": false,
  "ln_before": false,
  "mh_adapter": false,
  "output_adapter": true,
  "adapter_residual_before_ln": false,
  "non_linearity": "relu",
  "original_ln_after": true,
  "original_ln_before": true,
  "reduction_factor": 16,
  "residual_before_ln": true
}

Author

  Name
Yifan Hou
  GitHub

Versions

Identifier Comment Score Download
1 DEFAULT

Citations

Adapter
@article{hou2022adapters, title={Adapters for Enhanced Modeling of Multilingual Knowledge and Text}, author={Hou, Yifan and Jiao, Wenxiang and Liu, Meizhen and Allen, Carl and Tu, Zhaopeng and Sachan, Mrinmaya}, journal={arXiv preprint arXiv:2210.13617}, year={2022} }
Architecture
@misc{pfeiffer2020adapterfusion,
  title={AdapterFusion: Non-Destructive Task Composition for Transfer Learning},
  author={Jonas Pfeiffer and Aishwarya Kamath and Andreas Rücklé and Kyunghyun Cho and Iryna Gurevych},
  year={2020},
  eprint={2005.00247},
  archivePrefix={arXiv},
  primaryClass={cs.CL}
}
Task
@article{hou2022adapters,
  title={Adapters for Enhanced Modeling of Multilingual Knowledge and Text},
  author={Hou, Yifan and Jiao, Wenxiang and Liu, Meizhen and Allen, Carl and Tu, Zhaopeng and Sachan, Mrinmaya},
  journal={arXiv preprint arXiv:2210.13617},
  year={2022}
}