Lei Cui
3 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
MARD introduces a multi-agent framework that combines Large Language Models (LLMs) with traditional static analysis engines to achieve robust and highly interpretable Android malware detection with low computational cost.
The paper proposes SATA, a semantics-aware traffic augmentation framework, to significantly improve the generalization of website fingerprinting models by addressing variability in resource composition and cross-layer feature instability.
The paper quantifies the exact parametric memory capacity of LLMs using LoRA and proposes a new optimization strategy, MemFT, to enhance memory fidelity.
Papers
How LoRA Remembers? A Parametric Memory Law for LLM Finetuning
Ziwen Xu, Haiwen Hong, Linsong Yu, Benglei Cui +3 more
The paper quantifies the exact parametric memory capacity of LLMs using LoRA and proposes a new optimization strategy, MemFT, to enhance memory fidelity.