Vishnu Boddeti

Associate Professor@Michigan State University·美国·OpenReview

研究方向

Neural Operators · Generative Models · Foundation Models · Physics Informed Neural Networks · Fairness in Machine Learning · Homomorphic Encryption for Machine Learning

3.5

InPhyRe Discovers: Large Multimodal Models Struggle in Inductive Physical Reasoning

ICLR 2026Rejected

二作

3.3

FoeGlass: When Simple In-Context Learning Is Enough for Red Teaming Audio Deepfake Detectors

ICLR 2026Rejected

三作

7.8

Obliviator Reveals the Cost of Nonlinear Guardedness in Concept Erasure

NeurIPS 2025Poster

三作

7.3

PolyJuice Makes It Real: Black-Box, Universal Red Teaming for Synthetic Image Detectors

合作者 (13)

Vishnu Boddeti

InPhyRe Discovers: Large Multimodal Models Struggle in Inductive Physical Reasoning

FoeGlass: When Simple In-Context Learning Is Enough for Red Teaming Audio Deepfake Detectors

Obliviator Reveals the Cost of Nonlinear Guardedness in Concept Erasure

PolyJuice Makes It Real: Black-Box, Universal Red Teaming for Synthetic Image Detectors

OASIS Uncovers: High-Quality T2I Models, Same Old Stereotypes

CoInD: Enabling Logical Compositions in Diffusion Models

A Neural Architecture Dataset for Adversarial Robustness