Olivera Kotevska

The paper introduces AutoMIA, a novel framework that uses LLM agents to automate the discovery and implementation of Membership Inference Attacks (MIAs), achieving state-of-the-art performance by systematically exploring attack strategies.

SelfGrader proposes a lightweight, robust guardrail for detecting LLM jailbreaks by formulating the detection problem as a numerical grading task using anchored token-level logits, achieving strong performance across various benchmarks.

XMark introduces a novel multi-bit watermarking technique that reliably embeds binary messages into LLM-generated text while maintaining high text quality and robust performance even with limited token context.

The paper proposes CTRL-STEER, a closed-loop framework that adaptively adjusts intervention strength to stabilize concept regulation and improve task success in Vision-Language-Action models without retraining the base model.

The paper proposes IntraShuffler, a novel privacy-preserving middleware defense that enables gradient shuffling in Heterogeneous Differential Privacy Federated Learning (HDP-FL) while maintaining the utility of $\varepsilon$-aware server aggregation.

The paper proposes IntraShuffler, a novel privacy-preserving middleware defense that enables gradient shuffling in Heterogeneous Differential Privacy Federated Learning (HDP-FL) systems, significantly reducing inference risks while maintaining model utility.