One-Shot Learning as Instruction Data Prospector for Large Language Models
Yunshui Li, Binyuan Hui, Xiaobo Xia, Jiaxi Yang, Min Yang, Lei Zhang, Shuzheng Si, Ling-Hao Chen, Junhao Liu, Tongliang Liu, Fei Huang, Yongbin Li
Abstract
Contemporary practices in instruction tuning often hinge on enlarging data scaling without a clear strategy for ensuring data quality, inadvertently introducing noise that may compromise model performance. To address this challenge, we introduce Nuggets, a novel and efficient methodology that leverages one-shot learning to discern and select high-quality instruction data from extensive datasets. Nuggets assesses the potential of individual instruction examples to act as effective one-shot learning instances, thereby identifying those that can significantly improve performance across diverse tasks. Nuggets utilizes a scoring system based on the impact of candidate examples on the perplexity of a diverse anchor set, facilitating the selection of the most advantageous data for instruction tuning. Through rigorous evaluations on two benchmarks, namely MT-Bench and Alpaca-Eval, our study illustrates that instruction tuning with the top 1% of examples curated by Nuggets substantially outperforms conventional methods employing the entire dataset.- Anthology ID:
- 2024.acl-long.252
- Volume:
- Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
- Month:
- August
- Year:
- 2024
- Address:
- Bangkok, Thailand
- Editors:
- Lun-Wei Ku, Andre Martins, Vivek Srikumar
- Venue:
- ACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 4586–4601
- Language:
- URL:
- https://aclanthology.org/2024.acl-long.252
- DOI:
- 10.18653/v1/2024.acl-long.252
- Cite (ACL):
- Yunshui Li, Binyuan Hui, Xiaobo Xia, Jiaxi Yang, Min Yang, Lei Zhang, Shuzheng Si, Ling-Hao Chen, Junhao Liu, Tongliang Liu, Fei Huang, and Yongbin Li. 2024. One-Shot Learning as Instruction Data Prospector for Large Language Models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4586–4601, Bangkok, Thailand. Association for Computational Linguistics.
- Cite (Informal):
- One-Shot Learning as Instruction Data Prospector for Large Language Models (Li et al., ACL 2024)
- PDF:
- https://preview.aclanthology.org/add_acl24_videos/2024.acl-long.252.pdf