Rui Wang

PASTA proposes a novel, twofold stealthy backdoor attack that enables high-success-rate backdoor activation across arbitrary patches in Vision Transformers by leveraging the Trigger Radiating Effect (TRE).

DETOUR proposes a practical backdoor attack against object detection models by using semantic triggers that are robust to variations in size, location, and field of view (FoV), overcoming limitations of existing fixed-trigger attacks.

This survey provides a comprehensive, structured taxonomy of split learning techniques for fine-tuning Large Language Models (LLMs), covering model optimization, system efficiency, and privacy preservation.

The paper introduces ReTokSync, a self-synchronizing framework that resolves tokenization ambiguity in Generative Linguistic Steganography (GLS) by correcting mismatches only when they occur, thereby maintaining high security and extraction accuracy.

This paper provides the first integrated analysis of model dememorization, unifying unlearnability and unlearning methods, and offering theoretical guarantees on dememorization depth.

The paper addresses the challenge of multi-turn view planning for VLMs by proposing an iterative framework that uses self-exploration and view graph distillation, significantly improving planning performance over state-of-the-art models.