Sarah M. Erfani

Research focus

Red-Teaming & Adversarial Robustness (2)Multimodal Models (1)Interpretability & Mechanistic Interp (1)Speech & Audio (1)

Frequent co-authors

Paul Montague (1)Andrew C. Cullen (1)Amardeep Kaur (1)Christopher Leckie (1)

Papers (2)

Jun 17, 2026

Paul Montague +43w ago

Semantic Robustness Certification for Vision-Language Models

VLMs can now be certified for robustness against semantic variations without the burden of extra data, transforming how we assess model reliability in real-world applications.

Paul Montague, Andrew C. Cullen, Amardeep Kaur +2

Multimodal Models Red-Teaming & Adversarial Robustness

May 4, 2026

May 4, 2026·also JHU, Monash

Dimensionality-Aware Anomaly Detection in Learned Representations of Self-Supervised Speech Models

Adversarial attacks on speech models leave tell-tale geometric fingerprints in early representation layers that can be detected without transcripts.

Sandra Arcos-Holzinger, Sarah M. Erfani, Sanjeev Khudanpur

Interpretability & Mechanistic Interp Red-Teaming & Adversarial Robustness Speech & Audio

Search

Sarah M. Erfani

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (2)