Prefix-Tuning: Optimizing Continuous Prompts for Generation
Abstract
Fine-tuning is the de facto way to leverage large pretrained language models to perform downstream tasks. However, it modifies all the language model parameters and therefore necessitates storing a full copy for each task. In this paper, we propose prefix-tuning, a lightweight alternative to fine-tuning for natural language generation tasks, which keeps language model parameters frozen, but optimizes a small continuous task-specific vector (called the prefix). Prefix-tuning draws inspiration from prompting, allowing subsequent tokens to attend to this prefix as if it were "virtual tokens". We apply prefix-tuning to GPT-2 for table-to-text generation and to BART for summarization. We find that by learning only 0.1\% of the parameters, prefix-tuning obtains comparable performance in the full data setting, outperforms fine-tuning in low-data settings, and extrapolates better to examples with topics unseen during training.
- Publication:
-
arXiv e-prints
- Pub Date:
- January 2021
- DOI:
- arXiv:
- arXiv:2101.00190
- Bibcode:
- 2021arXiv210100190L
- Keywords:
-
- Computer Science - Computation and Language