13 papers
When Agents Learn to Be You: Benchmarking Privacy Leakage, Impersonation Risk, and Defenses in Persona Skills
Yongli Xiang, Zhifang Zhang, Bojun Yang +4
Persona skills distill personal interaction histories into portable and executable artifacts for downstream agents. While enabling flexible personalization, this process concentrat…
JuZhou 1.0 Technical Report: The First Edge-Native Text-to-Image Foundation Model Trained Entirely on China-Developed AI Accelerators
Ce Chen, Congrui Wang, Yonglin Li +23
Text-to-image (T2I) diffusion models typically require substantial computational resources and cloud infrastructure, posing significant challenges for edge deployment in terms of l…
When Safety Collides: Resolving Multi-Category Harmful Conflicts in Text-to-Image Diffusion via Adaptive Safety Guidance
Yongli Xiang, Ziming Hong, Zhaoqing Wang +3
Text-to-Image (T2I) diffusion models have demonstrated significant advancements in generating high-quality images, while raising potential safety concerns regarding harmful content…
VII: Visual Instruction Injection for Jailbreaking Image-to-Video Generation Models
Bowen Zheng, Yongli Xiang, Ziming Hong +4
Image-to-Video (I2V) generation models, which condition video generation on reference images, have shown emerging visual instruction-following capability, allowing certain visual c…
Mitigating Mismatch within Reference-based Preference Optimization
Suqin Yuan, Xingrui Yu, Jiyang Zheng +4
Direct Preference Optimization (DPO) has become the de facto standard for offline preference alignment of large language models, but its reliance on a reference policy introduces a…
Unifying Stable Optimization and Reference Regularization in RLHF
Li He, Qiang Qu, He Zhao +4
Reinforcement Learning from Human Feedback (RLHF) has advanced alignment capabilities significantly but remains hindered by two core challenges: \textbf{reward hacking} and \textbf…
Privacy-Preserving Model Transcription with Differentially Private Synthetic Distillation
Bochao Liu, Shiming Ge, Pengju Wang +2
While many deep learning models trained on private datasets have been deployed in various practical tasks, they may pose a privacy leakage risk as attackers could recover informati…
FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction
Runqi Lin, Alasdair Paren, Suqin Yuan +4
The integration of new modalities enhances the capabilities of multimodal large language models (MLLMs) but also introduces additional vulnerabilities. In particular, simple visual…
State Similarity in Modular Superconducting Quantum Processors with Classical Communications
Bujiao Wu, Changrong Xie, Peng Mi +20
As quantum devices continue to scale, distributed quantum computing emerges as a promising strategy for executing large-scale tasks across modular quantum processors. A central cha…
Direct Advantage Regression: Aligning LLMs with Online AI Reward
Li He, He Zhao, Stephen Wan +3
Online AI Feedback (OAIF) presents a promising alternative to Reinforcement Learning from Human Feedback (RLHF) by utilizing online AI preference in aligning language models (LLMs)…
Jailbreaking the Non-Transferable Barrier via Test-Time Data Disguising
Yongli Xiang, Ziming Hong, Lina Yao +2
Non-transferable learning (NTL) has been proposed to protect model intellectual property (IP) by creating a "non-transferable barrier" to restrict generalization from authorized to…
Toward Robust Non-Transferable Learning: A Survey and Benchmark
Ziming Hong, Yongli Xiang, Tongliang Liu
Over the past decades, researchers have primarily focused on improving the generalization abilities of models, with limited attention given to regulating such generalization. Howev…
From 2D Alignment to 3D Plausibility: Unifying Heterogeneous 2D Priors and Penetration-Free Diffusion for Occlusion-Robust Two-Hand Reconstruction
Gaoge Han, Yongkang Cheng, Zhe Chen +2
Two-hand reconstruction from monocular images is hampered by complex poses and severe occlusions, which often cause interaction misalignment and two-hand penetration. We address th…