Aadya Arora

AI Researcher | CV/ML | Medical Imaging

Final-year undergraduate at IIT Gandhinagar, majoring in Electrical Engineering with a Minor in AI. Passionate about Computer Vision, Machine Learning, and Generative AI with applications in healthcare and multimodal perception.

Incoming Data Scientist at Microsoft IDC (2026)

Aadya Arora

About Me

I'm a machine learning researcher focused on vision-language models, medical imaging, and generative AI. My work bridges computer vision and AI, with applications in healthcare diagnostics and autonomous systems.

Through internships at IISc Bangalore, University of Bath, IIIT Hyderabad, and Microsoft IDC, I've developed expertise in diffusion models, few-shot learning, document AI, and anomaly detection.

Beyond research, I'm passionate about making AI more interpretable, robust, and practical for real-world applications.

Featured Publications

MedFocusCLIP medical imaging

MedFocusCLIP: Few-Shot Medical Classification with Pixel-Wise Attention

ICASSP 2025

Aadya Arora, Vinay Namboodiri

Integrated SAM-based pixel attention with CLIP for interpretable and robust few-shot medical image classification. Demonstrates improved performance on medical datasets through attention mechanism visualization.

WavShadow shadow removal

WavShadow: Wavelet-Based Shadow Segmentation and Removal

ICVGIP 2024

S. Jain*, Aadya Arora*, V. Vekaria, K. Gandhi, S. Raman

Proposed a wavelet-enhanced pipeline for shadow detection and removal, achieving state-of-the-art performance. Novel approach combining frequency-domain analysis with deep learning.

TEMPEST battery chemistry

Battery Chemistry Recommendation using Machine Learning

Under Review IEEE IAS Journal

Aadya Arora, S. Patil, P. Bhardwaj

Introduced TEMPEST, an LSTM-based framework for predicting internal battery temperature and recommending optimal chemistry. Applied to EV and grid storage systems.

Work & Research Experience

Data Scientist Intern

Microsoft IDC
May – July 2025
  • Developed LLM-driven system translating natural language into PowerPoint formatting commands
  • Built scalable structured data pipelines and evaluation frameworks
  • Improved robustness on complex formatting tasks through error analysis

Research Intern: Diffusion Transformers

Indian Institute of Science (IISc) Bangalore
Dec 2024 – Apr 2025
  • Performed layer-wise ablation of Diffusion Image Transformers (DiTs)
  • Analyzed attention mechanisms for prompt adherence and semantic control
  • Developed training-free image editing techniques via attention manipulation

Research Intern: Vision-Language Models

University of Bath
May – July 2024
  • Implemented open-vocabulary few-shot referring image segmentation
  • Integrated HIPIE adapters with CLIP for improved generalization
  • Achieved 85.15 mIoU on RefCOCO(+) benchmark

Research Intern: Autonomous Driving

IIIT Hyderabad & IIT Hyderabad
Dec 2023 – Jan 2024
  • Analyzed corner cases in Indian traffic datasets (IDD, CODA)
  • Used class-agnostic RPNs for anomaly discovery in unstructured environments
  • Identified edge cases critical for robust autonomous systems

Skills & Expertise

Programming Languages

Python C++ C MATLAB

Deep Learning & ML

CNNs Vision Transformers CLIP Diffusion Models SAM LSTMs

Frameworks & Tools

PyTorch TensorFlow OpenCV mmDetection DeepLabv3+ Azure OpenAI

Domains

Computer Vision Medical Imaging Generative AI Document AI Autonomous Driving

Key Achievements

πŸ†

Microsoft PPO

Pre-Placement Offer as Data Scientist

πŸŽ“

IEEE Reviewer

Signal Processing Letters (2025–Present)

πŸ“š

Teaching Assistant

Machine Learning; Probability, Statistics & Data Visualization

🎯

Microsoft Research Summit

Invited to MSRI Academic Summit 2025

Curriculum Vitae

πŸ“„ Full CV

Download my complete CV with detailed research, publications, and academic background.

Download PDF

πŸ“‹ Research Summary

Quick overview of research interests, key publications, and current work focus areas.

View Summary