Chinmay Hegde

About

Chinmay Hegde is an Associate Professor in the Department of Computer Science and Engineering and Electrical and Computer Engineering at NYU Tandon School of Engineering. His research focuses on machine learning, algorithms, big data, signal and image processing, and the development of AI models to solve problems in imaging, materials design, and transportation. He is actively involved in identifying weaknesses in current AI models and developing techniques to improve their robustness and safety. Hegde's recent work includes exposing critical shortcomings in methods designed to prevent illegal content generation by text-to-image AI models, demonstrating how these safety measures can be circumvented through concept inversion attacks. His research emphasizes that safety solutions for generative AI require altering the model training process rather than relying on post hoc fixes. Additionally, he has contributed to developing AI techniques capable of changing a person's apparent age in images while maintaining their biometric identifiers, advancing the field of identity-preserving image editing. His work is characterized by a focus on both theoretical advancements and practical applications, with collaborations involving PhD candidates, graduate fellows, and postdoctoral researchers.

Research topics

• Artificial Intelligence
• Computer Science
• Theoretical computer science
• Mathematics
• Engineering
• Geometry
• Computer engineering
• Mechanical engineering

Selected publications

• Spatiotemporally Constrained Action Space Attacks on Deep Reinforcement Learning Agents

  2020-04-03 · 50 citations

  article

  Robustness of Deep Reinforcement Learning (DRL) algorithms towards adversarial attacks in real world applications such as those deployed in cyber-physical systems (CPS) are of increasing concern. Numerous studies have investigated the mechanisms of attacks on the RL agent's state space. Nonetheless, attacks on the RL agent's action space (corresponding to actuators in engineering systems) are equally perverse, but such attacks are relatively less studied in the ML literature. In this work, we fi…

  PublisherDOI

• Fast inverse design of microstructures via generative invariance networks

  Nature Computational Science · 2021 · 49 citations

  DOI

• PITCH: AI-assisted Tagging of Deepfake Audio Calls using Challenge-Response

  2025-08-13 · 6 citations

  article
  PublisherDOI

• Benchmarking scientific machine-learning approaches for flow prediction around complex geometries

  Communications Engineering · 2025-10-31 · 3 citations

  articleOpen access

  Rapid and accurate simulations of fluid dynamics around complicated geometric bodies are critical in a variety of engineering and scientific applications. While scientific machine learning (SciML) has shown considerable promise, most studies in this field are limited to simple geometries. This paper addresses this gap by benchmarking diverse SciML models, including neural operators and vision transformer-based foundation models, for fluid flow prediction over intricate geometries. We evaluate th…

  PublisherOA PDFDOI

• OpenThoughts: Data Recipes for Reasoning Models

  ArXiv.org · 2025-06-04 · 1 citations

  preprintOpen access

  Reasoning models have made rapid progress on many benchmarks involving math, code, and science. Yet, there are still many open questions about the best training recipes for reasoning since state-of-the-art models often rely on proprietary datasets with little to no public information available. To address this, the goal of the OpenThoughts project is to create open-source datasets for training reasoning models. After initial explorations, our OpenThoughts2-1M dataset led to OpenThinker2-32B, the…

  PublisherOA PDFDOI

Recent grants

• CAREER: Advances in Graph Learning and Inference

  NSF · $365k · 2019–2024

• CRII: CIF: Towards Linear-Time Computation of Structured Data Representations

  NSF · $173k · 2016–2019

• CAREER: Advances in Graph Learning and Inference

  NSF · $160k · 2018–2019

Frequent coauthors

• Soumik Sarkar

  Iowa State University

  55 shared

• Aditya Balu

  41 shared

• Baskar Ganapathysubramanian

  Iowa State University

  33 shared

• Ameya Joshi

  New York University

  26 shared

• Richard G. Baraniuk

  25 shared

• Mohammadreza Soltani

  Islamic Azad University, Tehran

  25 shared

• Viraj Shah

  24 shared

• Gauri Jagatap

  New York University

  22 shared

Labs

• DICE (Data, Intelligence, and Computation in Engineering) LabPI

Education

• PhD

  Rice University

Similar researchers at New York University

• Rob Fergush-95
• Dennis Shashah-84
• Mehryar Mohrih-77