Parameter-efficient Multi-task Fine-tuning for Transformers via Shared Hypernetworks
Rabeeh Karimi Mahabadi, Sebastian Ruder, Mostafa Dehghani, James Henderson
Abstract
State-of-the-art parameter-efficient fine-tuning methods rely on introducing adapter modules between the layers of a pretrained language model. However, such modules are trained separately for each task and thus do not enable sharing information across tasks. In this paper, we show that we can learn adapter parameters for all layers and tasks by generating them using shared hypernetworks, which condition on task, adapter position, and layer id in a transformer model. This parameter-efficient multi-task learning framework allows us to achieve the best of both worlds by sharing knowledge across tasks via hypernetworks while enabling the model to adapt to each individual task through task-specific adapters. Experiments on the well-known GLUE benchmark show improved performance in multi-task learning while adding only 0.29% parameters per task. We additionally demonstrate substantial performance improvements in few-shot domain generalization across a variety of tasks. Our code is publicly available in https://github.com/rabeehk/hyperformer.- Anthology ID:
- 2021.acl-long.47
- Volume:
- Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
- Month:
- August
- Year:
- 2021
- Address:
- Online
- Editors:
- Chengqing Zong, Fei Xia, Wenjie Li, Roberto Navigli
- Venues:
- ACL | IJCNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 565–576
- Language:
- URL:
- https://aclanthology.org/2021.acl-long.47
- DOI:
- 10.18653/v1/2021.acl-long.47
- Cite (ACL):
- Rabeeh Karimi Mahabadi, Sebastian Ruder, Mostafa Dehghani, and James Henderson. 2021. Parameter-efficient Multi-task Fine-tuning for Transformers via Shared Hypernetworks. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 565–576, Online. Association for Computational Linguistics.
- Cite (Informal):
- Parameter-efficient Multi-task Fine-tuning for Transformers via Shared Hypernetworks (Karimi Mahabadi et al., ACL-IJCNLP 2021)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2021.acl-long.47.pdf
- Code
- rabeehk/hyperformer
- Data
- BoolQ, CoLA, GLUE, IMDb Movie Reviews, PAWS, QNLI