Beyond Parameter Aggregation: Semantic Consensus for Federated Fine-Tuning of LLMs

ArXi:2605.11857v1 Announce Type: new Federated fine-tuning of large language models is commonly formulated as a parameter aggregation problem. However, even parameter-efficient methods require transmitting large collections of trainable weights, assume aligned architectures, and rely on white-box access to model parameters. As model sizes continue to grow and deployments become increasingly heterogeneous, these assumptions become progressively misaligned with practical constraints.