Ravi Mangal

Assistant Professor@Colorado State University·美国·OpenReview

研究方向

adversarial examples · robustness · safety of neural networks · formal methods · program verification · programming languages · interpretability · explainability of neural networks · LLMs for coding

4.5

A Mixture of Linear Corrections Generates Secure Code

ICLR 2026Rejected

二作

5.5

Validating Mechanistic Interpretations: An Axiomatic Approach

合作者 (8)