10 papers
DSKC: Domain Style Modeling with Adaptive Knowledge Consolidation for Exemplar-free Lifelong Person Re-Identification
Shiben Liu, Mingyue Xu, Huijie Fan +3
Lifelong Person Re-identification (LReID) aims to continuously match individuals across camera views from sequential data streams. Existing LReID methods often ignore domain-specif…
Exploring the Vulnerabilities of Federated Learning: A Deep Dive into Gradient Inversion Attacks
Pengxin Guo, Runxi Wang, Shuang Zeng +7
Federated Learning (FL) has emerged as a promising privacy-preserving collaborative model training paradigm without sharing raw data. However, recent studies have revealed that pri…
Unleashing the Potential of Large Language Models for Text-to-Image Generation through Autoregressive Representation Alignment
Xing Xie, Jiawei Liu, Ziyue Lin +4
We present Autoregressive Representation Alignment (ARRA), a new training framework that unlocks global-coherent text-to-image generation in autoregressive LLMs without architectur…
Discrete Diffusion Models with MLLMs for Unified Medical Multimodal Generation
Jiawei Mao, Yuhan Wang, Lifeng Chen +6
Recent advances in generative medical models are constrained by modality-specific scenarios that hinder the integration of complementary evidence from imaging, pathology, and clini…
Distribution-aware Forgetting Compensation for Exemplar-Free Lifelong Person Re-identification
Shiben Liu, Huijie Fan, Qiang Wang +3
Lifelong Person Re-identification (LReID) suffers from a key challenge in preserving old knowledge while adapting to new information. The existing solutions include rehearsal-based…
ATSTrack: Enhancing Visual-Language Tracking by Aligning Temporal and Spatial Scales
Yihao Zhen, Qiang Wang, Yu Qiao +2
A main challenge of Visual-Language Tracking (VLT) is the misalignment between visual inputs and language descriptions caused by target movement. Previous trackers have explored ma…