In-Context Learning Distillation for Efficient Few-Shot Fine-Tuning

Autor:	Duan, Yifei, Li, Liu, Zhai, Zirui, Yao, Jinxia
Rok vydání:	2024
Předmět:	Computer Science - Computation and Language
Druh dokumentu:	Working Paper
Popis:	We applied few-shot in-context learning on the OPT-1.3B model for the natural language inference task and employed knowledge distillation to internalize the context information, reducing model parameter from 1.3B to 125M and achieving a size reduction from 2.5GB to 0.25GB. Compared to using in-context learning alone on similarly sized models, this context distillation approach achieved a nearly 50% improvement in out-of-domain accuracy, demonstrating superior knowledge transfer capabilities over prompt-based methods. Furthermore, this approach reduced memory consumption by up to 60% while delivering a 20% improvement in out-of-domain accuracy compared to conventional pattern-based fine-tuning. Comment: 7 pages, 6 figures
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2412.13243 Zobrazit plný text záznamu View this record from Arxiv