Dialogue state tracking (DST) is an important step in dialogue management to keep track of users' beliefs. Existing works fine-tune all language model (LM) parameters to tackle the DST task, which requires significant data and computing resources for training and hosting. The cost grows exponentially in the real-world deployment where dozens of fine-tuned LM are used for different domains and tasks. To reduce parameter size and better utilize cross-task shared information, we propose to use soft prompt token embeddings to learn task properties. Without tuning LM parameters, our method drastically reduces the number of parameters needed to less than 0.5% of prior works while achieves better low-resource DST performance.
@article{arxiv.2301.10915,
title = {Parameter-Efficient Low-Resource Dialogue State Tracking by Prompt Tuning},
author = {Mingyu Derek Ma and Jiun-Yu Kao and Shuyang Gao and Arpit Gupta and Di Jin and Tagyoung Chung and Nanyun Peng},
journal= {arXiv preprint arXiv:2301.10915},
year = {2023}
}
Comments
In the INTERSPEECH 2023, and the Second Workshop on Efficient Natural Language and Speech Processing (ENLSP) at NeurIPS 2022