OpenReview
.net
OpenReview
.net
Login
OpenReview
.net
Login
Sarah Erfani
Full Professor, University of Melbourne
Joined
December 2017
Names
Sarah Erfani
(Preferred)
,
Sarah Monazam Erfani
Emails
****@unimelb.edu.au
(Confirmed)
,
****@unimellb.edu.au
(Confirmed)
Personal Links
Homepage
Google Scholar
DBLP
ORCID
Semantic Scholar
Career & Education History
Full Professor
University of Melbourne
(unimelb.edu.au)
2026
–
Present
Associate Professor
The University of Melbourne
(unimelb.edu.au)
2023
–
2025
Assistant Professor
University of Melbourne
(unimelb.edu)
2017
–
2022
Postdoc
University of Melbourne
(unimelb.edu)
2016
–
2017
PhD student
University of Melbourne
(unimelb.edu)
2012
–
2015
Advisors, Relations & Conflicts
Coauthor
xingjun ma
2017
–
Present
Expertise
AI safety
Present
adversarial machine learning
Present
unsupervised learning
Present
anomaly detection
Present
Publications
Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses
Xiao Li
,
Xiang Zheng
,
Yifeng Gao
,
Xinyu Xia
,
Yixu Wang
,
Xin Wang
,
Ye Sun
,
Yunhan Zhao
,
Ming Wen
,
Jiayu Li
,
Zixing Chen
,
Xun Gong
,
Yi Liu
,
Yige Li
,
Yutao Wu
,
Cong Wang
,
Jun Sun
,
Yixin Cao
,
Zhineng Chen
,
Jingjing Chen
et al. (18 additional authors not shown)
Preprint, 2026
Readers:
Everyone
Learnable Coreset Selection for Graph Active Learning
Xueqi Ma
,
Xingjun Ma
,
Sarah Erfani
,
James Bailey
Accepted by TMLR
Readers:
Everyone
Semantic-aware Adversarial Fine-tuning for CLIP
Jiacheng Zhang
,
Jinhao Li
,
Hanxun Huang
,
Sarah Monazam Erfani
,
Benjamin I. P. Rubinstein
,
Feng Liu
Accepted by TMLR
Readers:
Everyone
Leveraging Dark Knowledge for Intrinsic Multimodal Out-of-Distribution Detection
Nimeshika Udayangani
,
Sarah Monazam Erfani
,
Christopher Leckie
Submitted to ICLR 2026
Readers:
Everyone
Beyond Hard Supervised Fine-tuning: Enhancing Image-text Alignment of Strong Models with Weak Models
Jinhao Li
,
Sarah Monazam Erfani
,
Lei Feng
,
Guangrui Li
,
James Bailey
,
Feng Liu
Submitted to ICLR 2026
Readers:
Everyone
GeoDetect: Geometric Adversarial Detection for VLPs
Afsaneh Hasanebrahimi
,
Hanxun Huang
,
Christopher Leckie
,
James Bailey
,
Sarah Monazam Erfani
Submitted to ICLR 2026
Readers:
Everyone
Density-Aware Translation of Spurious Correlations in Zero-Shot VLMs
Afsaneh Hasanebrahimi
,
Hanxun Huang
,
Christopher Leckie
,
Sarah Monazam Erfani
Submitted to ICLR 2026
Readers:
Everyone
Geometry-Guided Adversarial Prompt Detection via Curvature and Local Intrinsic Dimension
Canaan Yung
,
Hanxun Huang
,
Christopher Leckie
,
Sarah Monazam Erfani
Submitted to ICLR 2026
Readers:
Everyone
On the Bayes Inconsistency of Disagreement Discrepancy Surrogates
Neil G Marchant
,
Andrew Craig Cullen
,
Feng Liu
,
Sarah Monazam Erfani
ICLR 2026 Poster
Readers:
Everyone
Characterising Universal Jailbreak Features and Refusal Direction in LLMs
Canaan Yung
,
Hanxun Huang
,
Christopher Leckie
,
Sarah Monazam Erfani
ICLR 2026 Conference Withdrawn Submission
Readers:
Everyone
View all 172 publications
Co-Authors
Afsaneh Hasanebrahimi
Amélie Barbe
Andrew C. Cullen
Andrew Craig Cullen
Benjamin I. P. Rubinstein
Bo Dai
Bo Li
Canaan Yung
Chaowei Xiao
Cheng Yang
Chien Aun Chan
Christian Walder
Christopher A. Leckie
Christopher Leckie
Chuang Liu
Cong Wang
David Hubczenko
Dawn Song
Dinh Phung
Dominique Barbe
Dragan Gasevic
Eduardo Velloso
Elaheh Alipour Chavary
Fangquan Lin
Fanzhe Qu
View all 164 co-authors