Varun Chandrasekaran

DAWN: Dual Space Regeneration Attack

Efficiently Attacking Memorization Scores

LLM-Based Code Translation Needs Formal Compositional Reasoning

Unearthing Skill-level Insights for Understanding Trade-offs of Foundation Models

Generative Monoculture in Large Language Models

Training Robust Ensembles Requires Rethinking Lipschitz Continuity

Not All Wrong is Bad: Using Adversarial Examples for Unlearning

Attention Speaks Volumes: Localizing and Mitigating Bias in Language Models