Shivang Chopra

Shivang Chopra

CS Ph.D. Student

Georgia Institute of Technology

About Me

I am a Ph.D. student in Computer Science at the Robotics Perception and Learning Lab (RIPL), Georgia Institute of Technology, advised by Prof. Zsolt Kira. My research focuses on theoretical machine learning and computer vision, with an emphasis on enhancing the adaptability and generalization capabilities of foundation models.

Prior to this, I completed a Masters by Research in Computer Science at Georgia Tech, where my thesis explored efficient and robust fine-tuning of Vision-Language Models. During my Master’s program, I also gained industry experience as a Computer Vision Research Intern at the Sony R&D Center, Switzerland (Fall 2023 - Spring 2024) and as an Applied Scientist Intern at Amazon Science, Seattle (Summer 2023).

Before my time at Georgia Tech, I worked at Microsoft Research India (MSRI) under the supervision of Dr. Venkat Padmanabhan (Managing Director, MSRI), where I contributed to Project HyWay.

Interests
  • Robustness
  • Efficiency
  • Distribution Shift
  • Vision Language Models
  • Diffusion Models
  • Computer Vision
Education

  • Ph.D. in Computer Science, 2028

    Georgia Institute of Technology

  • Master's in Computer Science, 2024

    Georgia Institute of Technology

  • B.Tech. in Computer Science, 2020

    Delhi Technological University

Recent News

Feb 2026Our full paper titled "The Geometry of Robustness: Optimizing Loss Landscape Curvature and Feature Manifold Alignment for Robust Finetuning of Vision-Language Models" was accepted at CVPR 2026.
May 2025Our short paper titled "MedMoE: Modality-Specialized Mixture of Experts for Medical Vision-Language Understanding" was accepted at the MMFM-BIOMED Workshop at CVPR 2025.
Feb 2025Our full paper titled “FRAMES-VQA: Benchmarking Fine-Tuning Robustness across Multi-Modal Shifts in Visual Question Answering” was accepted at CVPR 2025.
Jan 2025Our full paper titled “Directional Gradient Projection for Robust Fine-tuning of Foundation Models” was accepted at ICLR 2025.
Jan 2025Started my Ph.D. in Computer Science at Georgia Institute of Technology.
Dec 2024Successfully defended my Master’s thesis.
Oct 2024Our full paper titled "Refining Text-to-Image Generation: Towards Accurate Training-Free Glyph-Enhanced Image Generation" was accepted at WACV 2025.
Sep 2023Started working as a Computer Vision Research Intern at Sony R&D Center, Switzerland.
Aug 2023Our full paper titled "Learning to Discern: Imitating Heterogeneous Human Demonstrations with Preference and Representation Learning" was accepted at CoRL 2023.
May 2023Started working as an Applied Scientist Intern at Amazon, Seattle.
Apr 2023Our full paper titled "HyWay: Enabling Unstructured Conversations in the New Hybrid World" was accepted at UbiComp 2023 (IMWUT).
Apr 2023Our workshop paper titled "Symbiotic Artificial Intelligence: Order Picking and Ambient Sensing" was accepted at the Ambient AI Workshop in ICASSP 2023.
Aug 2022Started my Master’s in Computer Science at Georgia Institute of Technology.
Jul 2022Our workshop paper titled "Active Data Discovery: Mining Unknown Data using Submodular Information Measures" was accepted at the Real World ML Workshop in ICML 2022.
May 2022Started my research internship in Team HyWay at Microsoft Research Lab, India.
Jul 2021Started my research internship at the Faryabi Lab at the University of Pennsylvania.
Jun 2020Started my research internship at IBM Research Lab, India.
May 2020Successfully defended my Bachelor’s thesis.
Jan 2020Our highlight paper titled "Attention-based Sketch Recognition using Transformers" was accepted at ECAI 2020.
Dec 2019Our short paper titled "Utilizing Temporal Psycholinguistic Cues for Suicidal Intent Estimation" was accepted at ECIR 2020.
Nov 2019Our full paper titled "Hindi-English Hate Speech Detection: Author Profiling, Debiasing, and Practical Perspectives" was accepted at AAAI 2020.
Aug 2019Started my research internship at MIDAS Lab, IIIT Delhi.

Selected Publications

The Geometry of Robustness: Optimizing Loss Landscape Curvature and Feature Manifold Alignment for Robust Finetuning of Vision-Language Models
In this work, we propose GRACE, a unified low-rank fine-tuning framework designed to enhance both out-of-distribution (OOD) and adversarial robustness by integrating adaptive adversarial weight perturbations into a curriculum-driven Low-Rank Adaptation (LoRA) framework.
Shivang Chopra, Shaunak Halbe, Chengyue Huang, Brisa Maneechotesuwan, Zsolt Kira
PDF Cite CVPR 2026
The Geometry of Robustness: Optimizing Loss Landscape Curvature and Feature Manifold Alignment for Robust Finetuning of Vision-Language Models
FRAMES-VQA: Benchmarking Fine-Tuning Robustness across Multi-Modal Shifts in Visual Question Answering
In this work, we propose a new benchmark FRAMES-VQA (Fine-Tuning Robustness Across Multi-Modal Shifts in VQA) for evaluating robust fine-tuning for VQA tasks.
Chengyue Huang, Brisa Maneechotesuwan, Shivang Chopra, Zsolt Kira.
PDF Cite CVPR 2025
FRAMES-VQA: Benchmarking Fine-Tuning Robustness across Multi-Modal Shifts in Visual Question Answering
Refining Text-to-Image Generation: Towards Accurate Training-Free Glyph-Enhanced Image Generation
In this work, we introduce a training-free framework to enhance the capability of Diffusion Models to generate visual text.
Sanyam Lakhanpal, Shivang Chopra, Vinija Jain, Aman Chadha, Man Luo
PDF Cite WACV 2025
Refining Text-to-Image Generation: Towards Accurate Training-Free Glyph-Enhanced Image Generation
Source-Free Domain Adaptation with Diffusion-Guided Source Data Generation
This paper introduces a novel approach to leverage the generalizability capability of Diffusion Models for Source-Free Domain Adaptation (DM-SFDA).
Shivang Chopra, Suraj Kothawad, Houda Aynaou, Aman Chadha
PDF Cite
Source-Free Domain Adaptation with Diffusion-Guided Source Data Generation
Active Data Discovery: Mining Unknown Data using Submodular Information Measures
Active Learning is a very common yet powerful framework for iteratively and adaptively sampling subsets of the unlabeled sets with a …
Suraj Kothawad, Shivang Chopra, Saikat Ghosh, Rishabh Iyer
PDF Cite Poster ICML(W) 2023
Active Data Discovery: Mining Unknown Data using Submodular Information Measures
See all publications

Experience

 
 
 
 
 
Sony R&D
Computer Vision Research Intern
Sony R&D
September 2023 – April 2024 Zurich, Switzerland
I worked for eight months in the Advanced Sensors & Modelling Group at Sony R&D Lab on real-time ball tracking and spin estimation using Event-based Vision Sensors (EVS) for a Table Tennis robot.
 
 
 
 
 
Amazon Science
Applied Scientist Intern
Amazon Science
May 2023 – August 2023 Seattle, WA
I worked for three months as an applied scientist intern in the Amazon Selection and Catalog Systems (ASCS) team at Amazon Science where I developed a product recommendation system for newly onboarded products on Amazon.com.
 
 
 
 
 
Microsoft Research Lab, India
Research Intern
Microsoft Research Lab, India
May 2022 – August 2022 Bangalore, India
I worked as a research intern at Microsoft Research Lab, India in project HyWay, a computer vision based video conferencing system for hybrid workplaces.
 
 
 
 
 
University of Texas, Dallas
Research Assistant
University of Texas, Dallas
March 2022 – July 2022 Remote
I worked as a research assistant at CARAML Lab at UT Dallas on the problem of unknown class data discovery using Active Learning.
 
 
 
 
 
Codenation Innovation Labs
Software Engineer
Codenation Innovation Labs
September 2020 – April 2022 Bangalore, India
I worked as a full-time Software Engineer at CodeNation Innovation Labs for two years, contributing to multiple projects, including web-based applications built with React and Android/iOS applications developed using Kotlin and Swift.
 
 
 
 
 
IBM Research Lab
Research Intern
IBM Research Lab
June 2020 – August 2020 Delhi, India
I worked as a research intern at IBM Research Lab, India on the problem of real-time explicit content detection in videos.

Contact

  • You can reach me at shivangchopra11 [at] gatech [dot] edu
  • CODA Building, S1106C, 11th Floor, 756 West Peachtree St NW, Atlanta, GA 30332