Burkhard Stiller

Checking Leakage Witnesses versus Certifying Bounded Non-Leakage

When a language-model audit finds no leak, what is needed to certify non-leakage? We study guarantees over a declared prompt domain under an executable leakage criterion and decoding rule. For general bounded polynomial-time evaluators, a supplied leaking execution is polynomial-time checkable, while leak existence is...

CACTUS: Mask-Guided Semantic Clean-Label Backdoors in Decentralized Federated Learning

Results indicate that CACTUS can propagate backdoors through repeated DFL aggregation and show that ASR varies with network topology and increases with the malicious-node ratio.

FedEnD: Communication-efficient Federated Learning for non-IID data via decentralized ensemble distillation