MaTU aggregates task vectors across clients using sign-conflict similarity and lightweight masks, reaching 77 to 84 percent normalized accuracy on 30 vision tasks with one transmitted vector per client.
Remote sensing image scene classification: Benchmark and state of the art
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Many-Task Federated Fine-Tuning via Unified Task Vectors
MaTU aggregates task vectors across clients using sign-conflict similarity and lightweight masks, reaching 77 to 84 percent normalized accuracy on 30 vision tasks with one transmitted vector per client.