Rahul Gupta

How Catastrophic is Your LLM? Certifying Risks in Conversation

LH-DECEPTION: Simulating and Understanding LLM Deceptive Behaviors in Long-Horizon Interactions

Adversarial Déjà Vu: Jailbreak Dictionary Learning for Stronger Generalization to Unseen Attacks

Kaleidoscopic Teaming in Multi Agent Simulations

Adversarial Arena: Crowdsourcing Data Generation through Interactive Competition

Customize Multi-modal RAI Guardrails with Precedent-based predictions

Certifying Counterfactual Bias in LLMs

Discovering Knowledge Deficiencies of Language Models on Massive Knowledge Base