Boosting Natural Language Generation from Instructions with Meta-Learning

Budhaditya Deb, Ahmed Hassan Awadallah, Guoqing Zheng


Abstract
Recent work has shown that language models (LMs) trained with multi-task instructional learning (MTIL) can solve diverse NLP tasks in zero- and few-shot settings with improved performance compared to prompt tuning. MTIL illustrates that LMs can extract and use information about the task from instructions beyond the surface patterns of the inputs and outputs. This suggests that meta-learning may further enhance the utilization of instructions for effective task transfer. In this paper we investigate whether meta-learning applied to MTIL can further improve generalization to unseen tasks in a zero-shot setting. Specifically, we propose to adapt meta-learning to MTIL in three directions: 1) Model Agnostic Meta Learning (MAML), 2) Hyper-Network (HNet) based adaptation to generate task specific parameters conditioned on instructions, and 3) an approach combining HNet and MAML. Through extensive experiments on the large scale Natural Instructions V2 dataset, we show that our proposed approaches significantly improve over strong baselines in zero-shot settings. In particular, meta-learning improves the effectiveness of instructions and is most impactful when the test tasks are strictly zero-shot (i.e. no similar tasks in the training set) and are “hard” for LMs, illustrating the potential of meta-learning for MTIL for out-of-distribution tasks.
Anthology ID:
2022.emnlp-main.456
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6792–6808
Language:
URL:
https://aclanthology.org/2022.emnlp-main.456
DOI:
Bibkey:
Cite (ACL):
Budhaditya Deb, Ahmed Hassan Awadallah, and Guoqing Zheng. 2022. Boosting Natural Language Generation from Instructions with Meta-Learning. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 6792–6808, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Boosting Natural Language Generation from Instructions with Meta-Learning (Deb et al., EMNLP 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2022.emnlp-main.456.pdf