AI RESEARCH
TSFLora: Token-Compressed Split Fine-Tuning for Wireless Edge Networks
arXiv CS.LG
•
ArXi:2605.23988v1 Announce Type: cross Adapting large AI models (LAMs) to personalized edge data is challenging because wireless devices have limited memory, computation, and uplink capacity. Federated fine-tuning preserves data privacy but still requires each device to host the full model, while split learning reduces device memory at the cost of heavy activation transmission. This paper proposes TSFLora, a token-compressed split fine-tuning framework for communication-efficient LAM adaptation at the edge.