Sep, 2023
ConPET: 大型语言模型的连续参数高效调整
ConPET: Continual Parameter-Efficient Tuning for Large Language Models
TL;DRContinual Parameter-Efficient Tuning (ConPET), a generalizable paradigm for continual task adaptation of large language models (LLMs) with task-number-independent training complexity, reduces tuning costs and alleviates over-fitting and forgetting issue through static and dynamic ConPET versions.