Anamika Lochab — accepted AI-conference papers

2026

VERA-V: Variational Inference Framework for Jailbreaking Vision-Language Models

ICML 2026poster

Vision-Language Models (VLMs) extend large language models with visual reasoning, but their multimodal design also introduces new, underexplored vulnerabilities. Existing multimodal red-teaming methods largely rely on brittle templates, focus on single-attack settings, and expose only a narrow subse…

Cited by 0 Source ↗Scholar ↗

2025

VERA: Variational Inference Framework for Jailbreaking Large Language Models

Anamika Lochab, Lu Yan, Patrick Pynadath, Xiangyu Zhang, Ruqi Zhang

NeurIPS 2025poster

The rise of API-only access to state-of-the-art LLMs highlights the need for effective black-box jailbreak methods to identify model vulnerabilities in real-world settings. Without a principled objective for gradient-based optimization, most existing approaches rely on genetic algorithms, which are…

Cited by 0 Source ↗Scholar ↗