作者位次 #2/6ACL 2026Poster· Natural Language Processing
F ed P roxy: Federated Fine-Tuning of LLM s via Proxy SLM s and Heterogeneity-Aware Fusion
Tao Fan · Guoqiang Ma · Yuanfeng SONG · Lixin Fan · Kai Chen · Qiang Yang
摘要
Federated fine-tuning of Large Language Models (LLMs) is obstructed by a trilemma of challenges: protecting LLMs intellectual property (IP), ensuring client privacy, and mitigating performance loss on heterogeneous data. Existing methods like Offsite-Tuning (OT) secure the LLMs IP by having clien...