影响力指数

9.4/100

超过 29.2%

全站排名 #45,567

发表论文2 篇

平均评分4.5

年均产出2.0 篇/年

Zehang Deng

PhD student@Swinburne University of Technology·澳大利亚·OpenReview

研究方向

Model Inversion Attack · AI Security · LLM Security

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

ICLR 2026Poster

Adaptive Gray: Reducing Color Dependency to Improve Generalization in Deepfake Detection

ICLR 2026Rejected

合作者 (11)

Mehrtash Harandi

博士导师1 篇