Abstract
In transfer learning, it is imperative to achieve strong alignment between a pre-trained model and a downstream task. Prior work has done this by proposing task-specific pre-training objectives, which sacrifices the inherent scalability of the transfer learning paradigm. We instead achieve strong alignment by simultaneously modifying both the pre-trained model and the formulation of the downstream task, which is more efficient and preserves the scalability of transfer learning. We present GenSF (Generative Slot Filling), which leverages a generative pre-trained open-domain dialog model for slot filling. GenSF (1) adapts the pre-trained model by incorporating inductive biases about the task and (2) adapts the downstream task by reformulating slot filling to better leverage the pre-trained model’s capabilities. GenSF achieves state-of-the-art results on two slot filling datasets with strong gains in few-shot and zero-shot settings. We achieve a 9 F1 score improvement in zero-shot slot filling. This highlights the value of strong alignment between the pre-trained model and the downstream task.- Anthology ID:
- 2021.sigdial-1.51
- Volume:
- Proceedings of the 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue
- Month:
- July
- Year:
- 2021
- Address:
- Singapore and Online
- Editors:
- Haizhou Li, Gina-Anne Levow, Zhou Yu, Chitralekha Gupta, Berrak Sisman, Siqi Cai, David Vandyke, Nina Dethlefs, Yan Wu, Junyi Jessy Li
- Venue:
- SIGDIAL
- SIG:
- SIGDIAL
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 489–498
- Language:
- URL:
- https://aclanthology.org/2021.sigdial-1.51
- DOI:
- 10.18653/v1/2021.sigdial-1.51
- Cite (ACL):
- Shikib Mehri and Maxine Eskenazi. 2021. GenSF: Simultaneous Adaptation of Generative Pre-trained Models and Slot Filling. In Proceedings of the 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 489–498, Singapore and Online. Association for Computational Linguistics.
- Cite (Informal):
- GenSF: Simultaneous Adaptation of Generative Pre-trained Models and Slot Filling (Mehri & Eskenazi, SIGDIAL 2021)
- PDF:
- https://preview.aclanthology.org/bionlp-24-ingestion/2021.sigdial-1.51.pdf
- Code
- shikib/generative_slot_filling