FedBE appends zero-initialized transformer blocks to selected layers and allocates them across clients by resource and data profiles, reporting 12-74% better knowledge retention and 1.9-3.1x faster convergence in federated LLM fine-tuning.
A continual learn- ing survey: Defying forgetting in classification tasks,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Mitigating Catastrophic Forgetting with Adaptive Transformer Block Expansion in Federated Fine-Tuning
FedBE appends zero-initialized transformer blocks to selected layers and allocates them across clients by resource and data profiles, reporting 12-74% better knowledge retention and 1.9-3.1x faster convergence in federated LLM fine-tuning.