AAAI 2026technical0 citations
MetaGDPO: Alleviating Catastrophic Forgetting with Metacognitive Knowledge Through Group Direct Preference Optimization
Lanxue Zhang, Yuqiang Xie, Fang Fang, Fanglong Dong, Rui Liu, Yanan Cao
Abstract
Large Language Models demonstrate strong reasoning capabilities, which can be effectively compressed into smaller models. However, existing datasets and fine-tuning approaches still face challenges that lead to catastrophic forgetting, particularly for models smaller than 8B. First, most datasets typically ignore the relationship between training data knowledge and the model
BibTeX
@inproceedings{aaai2026_metagdpoalleviat,
title = {MetaGDPO: Alleviating Catastrophic Forgetting with Metacognitive Knowledge Through Group Direct Preference Optimization},
author = {Lanxue Zhang and Yuqiang Xie and Fang Fang and Fanglong Dong and Rui Liu and Yanan Cao},
booktitle = {AAAI 2026},
year = {2026}
}