影响力指数

-/100

发表论文3 篇

平均评分5.0

年均产出1.5 篇/年

Eugene Bagdasaryan

Assistant Professor@University of Massachusetts at Amherst·美国·OpenReview

研究方向

ML Security and Privacy

Soft Prompts Go Hard: Steering Visual Language Models with Hidden Meta-Instructions

ICLR 2025Rejected

Backdooring Bias into Text-to-Image Models

ICLR 2025Withdrawn

Ceci n'est pas une pomme: Adversarial Illusions in Multi-Modal Embeddings

ICLR 2024Rejected

合作者 (8)

Vitaly Shmatikov

博士导师2 篇

John Xavier Morris

Amir Houmansadr