Adel Bibi

Beyond Linear Probes: Dynamic Safety Monitoring for Language Models

BiasBusters: Uncovering and Mitigating Tool Selection Bias in Large Language Models

It's a TRAP! Task-Redirecting Agent Persuasion Benchmark for Web Agents

FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction

OMNI-LEAK: Orchestrator Multi-Agent Network Induced Data Leakage

To Distill or Not to Distill: Knowledge Transfer Undermines Safety of LLMs

ToolTweak: An Attack on Tool Selection in LLM-based Agents

Bi-Factorial Preference Optimization: Balancing Safety-Helpfulness in Language Models

On the Coexistence and Ensembling of Watermarks

Towards Certification of Uncertainty Calibration under Adversarial Attacks

Shh, don't say that! Domain Certification in LLMs

Mixture of Experts Made Intrinsically Interpretable

On the Coexistence and Ensembling of Watermarks

MIP against Agent: Malicious Image Patches Hijacking Multimodal OS Agents

Rethinking Safety in LLM Fine-tuning: An Optimization Perspective

Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language Models

SynthCLIP: Are We Ready for a Fully Synthetic CLIP Training?

Questioning Simplicity Bias Assumptions

Language Models' Internal Conflicts: Layer-wise Usable Information For Detecting Model (Un)answerability