Sarah Monazam Erfani

On the Bayes Inconsistency of Disagreement Discrepancy Surrogates

Toward Universal and Transferable Jailbreak Attacks on Vision-Language Models

POMP: A Theoretical Approach to Mitigate Forgetting in Finetuning Multi-Modal Models

Beyond Hard Supervised Fine-tuning: Enhancing Image-text Alignment of Strong Models with Weak Models

Density-Aware Translation of Spurious Correlations in Zero-Shot VLMs

Leveraging Dark Knowledge for Intrinsic Multimodal Out-of-Distribution Detection

Do Vision-Language Models Reason Like Humans? Exploring the Functional Roles of Attention Heads

Geometry-Guided Adversarial Prompt Detection via Curvature and Local Intrinsic Dimension

GeoDetect: Geometric Adversarial Detection for VLPs

What Do VLMs See? Benchmarking Vision-Language Models on Ambiguous Images

Characterising Universal Jailbreak Features and Refusal Direction in LLMs

Cognitive Mirrors: Exploring the Diverse Functional Roles of Attention Heads in LLM Reasoning

X-Transfer Attacks: Towards Super Transferable Adversarial Attacks on CLIP

Multi-level Certified Defense Against Poisoning Attacks in Offline Reinforcement Learning

Fortifying Time Series: DTW-Certified Robust Anomaly Detection

Detecting Backdoor Samples in Contrastive Language Image Pretraining

Attention-based Graph Coreset Labeling for Active Learning

HOGT: High-Order Graph Transformers

CURVALID: A Geometrically-guided Adversarial Prompt Detection

TUAP: Targeted Universal Adversarial Perturbations for CLIP

GAD-VLP: Geometric Adversarial Detection for Vision-Language Pre-Trained Models

Exploring Weak-to-Strong Generalization for CLIP-based Classification