Zehang Deng — accepted AI-conference papers

2026

Decoupling Defense Strategies for Robust Image Watermarking

Jiahui Chen, Zehang Deng, Zeyu Zhang, Chaoyang Li, Lianchen Jia, Lifeng Sun

CVPR 2026

Deep learning-based image watermarking, while robust against conventional distortions, remains vulnerable to advanced adversarial and regeneration attacks. Conventional countermeasures, which jointly optimize the encoder and decoder via a noise layer, face 2 inevitable challenges:(1) decrease of cle

Cited by 0 Source ↗Scholar ↗

2026

Sharpness-Aware Minimization in Logit Space Efficiently Enhances Direct Preference Optimization

Haocheng Luo, Zehang Deng, Thanh-Toan Do, Mehrtash Harandi, Dinh Phung, Trung Le

ICLR 2026poster

Direct Preference Optimization (DPO) has emerged as a popular algorithm for aligning pretrained large language models with human preferences, owing to its simplicity and training stability. However, DPO suffers from the recently identified squeezing effect (also known as likelihood displacement), wh…

Cited by 0 Source ↗code ↗Scholar ↗