LoRAMoE: Alleviating World Knowledge Forgetting in Large Language Models via MoE-Style Plugin

Shihan Dou, Enyu Zhou, Yan Liu, Songyang Gao, Wei Shen, Limao Xiong, Yuhao Zhou, Xiao Wang, Zhiheng Xi, Xiaoran Fan, Shiliang Pu, Jiang Zhu, Rui Zheng, Tao Gui, Qi Zhang, Xuanjing Huang


Abstract
Supervised fine-tuning (SFT) is a crucial step for large language models (LLMs), enabling them to align with human instructions and enhance their capabilities in downstream tasks. Substantially increasing instruction data is a direct solution to align the model with a broader range of downstream tasks or notably improve its performance on a specific task. However, we find that large-scale increases in instruction data can damage the world knowledge previously stored in LLMs. To address this challenge, we propose LoRAMoE, a novelty framework that introduces several low-rank adapters (LoRA) and integrates them by using a router network, like a plugin version of Mixture of Experts (MoE). It freezes the backbone model and forces a portion of LoRAs to focus on leveraging world knowledge to solve downstream tasks, to alleviate world knowledge forgetting. Experimental results show that, as the instruction data increases, LoRAMoE can significantly improve the ability to process downstream tasks, while maintaining the world knowledge stored in the LLM. Our code is available at https://github.com/Ablustrund/LoRAMoE.
Anthology ID:
2024.acl-long.106
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:
1932–1945
Language:
URL:
https://aclanthology.org/2024.acl-long.106
DOI:
10.18653/v1/2024.acl-long.106
Bibkey:
Cite (ACL):
Shihan Dou, Enyu Zhou, Yan Liu, Songyang Gao, Wei Shen, Limao Xiong, Yuhao Zhou, Xiao Wang, Zhiheng Xi, Xiaoran Fan, Shiliang Pu, Jiang Zhu, Rui Zheng, Tao Gui, Qi Zhang, and Xuanjing Huang. 2024. LoRAMoE: Alleviating World Knowledge Forgetting in Large Language Models via MoE-Style Plugin. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1932–1945, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
LoRAMoE: Alleviating World Knowledge Forgetting in Large Language Models via MoE-Style Plugin (Dou et al., ACL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.acl-long.106.pdf