影响力指数

-/100

发表论文1 篇

平均评分4.5

年均产出1.0 篇/年

Kazuki Iwahana

Researcher@NTT Social Informatics Laboratories·日本·OpenReview

研究方向

Machine Learning Secuity

Robust Backdoor Removal by Reconstructing Trigger-Activated Changes in Latent Representation

ICLR 2026Rejected

合作者 (4)

Toshiki Shibahara

Yusuke Yamasaki