10 papers
Self-supervised Monocular Depth and Pose Estimation for Endoscopy with Latent Priors
Ziang Xu, Bin Li, Yang Hu +4
Accurate 3D mapping in endoscopy enables quantitative, holistic lesion characterization within the gastrointestinal (GI) tract, requiring reliable depth and pose estimation. Howeve…
Are Targeted Data Poisoning Attacks as Effective as We Think?
William Xu, Chenyu Zhang, Yihan Wang +5
Targeted data poisoning attacks manipulate model predictions on specific test samples by injecting malicious data into training. Yet existing evaluations report average attack succ…
Metaphor-based Jailbreak Attacks on Text-to-Image Models
Chenyu Zhang, Lanjun Wang, Yiwen Ma +3
Text-to-image (T2I) models commonly incorporate defense mechanisms to prevent the generation of sensitive images. Unfortunately, recent jailbreak attacks have shown that adversaria…
RobotArena : Scalable Robot Benchmarking via Real-to-Sim Translation
Yash Jangir, Yidi Zhang, Pang-Chi Lo +7
The pursuit of robot generalists, agents capable of performing diverse tasks across diverse environments, demands rigorous and scalable evaluation. Yet real-world testing of robot…
T2I-RiskyPrompt: A Benchmark for Safety Evaluation, Attack, and Defense on Text-to-Image Model
Chenyu Zhang, Tairen Zhang, Lanjun Wang +3
Using risky text prompts, such as pornography and violent prompts, to test the safety of text-to-image (T2I) models is a critical task. However, existing risky prompt datasets are…
Reason2Attack: Jailbreaking Text-to-Image Models via LLM Reasoning
Chenyu Zhang, Lanjun Wang, Yiwen Ma +2
Text-to-Image(T2I) models typically deploy safety filters to prevent the generation of sensitive images. Unfortunately, recent jailbreaking attack methods manually design instructi…