Multilingual NLP, large language models, scientific text mining, and AI tools for research and education.
Research Area
AI & Machine Learning
Research in AI and machine learning spans multilingual NLP, computer vision, trustworthy and fair learning, AI for sustainability, and human-centred intelligent systems.
Featured Video
AI and machine learning at IIT Gandhinagar
Research Themes
What the group works on
Faculty
People shaping this area
Machine learning for scalable sensing, smart buildings, air quality, health sensing, and computational sustainability.
Algorithms and machine learning at scale, sketching, sampling, graph learning, and efficient ML pipelines.
Human-AI interaction, brain-computer interfaces, eye tracking, multimodal interaction, and assistive AI.
Fairness, federated learning, differential privacy, strategic clients, and trustworthy distributed ML.
Computer vision, deep learning, computational photography, image/video processing, and graphics.
Selected Publications
Recent Publications
FIT-GNN: Faster Inference Time for GNNs that FIT in Memory Using Coarsening
TMLR 2026 - efficient graph neural-network inference.
Gaslight, Gatekeep, V1-V3: Early Visual Cortex Alignment Shields Vision-Language Models from Sycophantic Manipulation
2026 preprint - robustness and alignment for vision-language models.
GF-Score: Certified Class-Conditional Robustness Evaluation with Fairness Guarantees
2026 preprint - fairness-aware robustness assessment.
MERLiN: Single-Shot Material Estimation and ReLighting for Photometric Stereo
ECCV 2024 - computer vision and inverse rendering.
Projects & Outputs
Representative Projects
Chitrabhasha: Developing India's Largest Dataset and Foundation Models for Multimodal AI
ANRF Advanced Research Grant led by Mayank Singh.
Scalable Detection of Brick Kilns from Low-Resolution Satellite Imagery for Environmental Compliance
ANRF project led by Nipun Batra.
iGazeBuddy: Multimodal gaze-controlled learning system for dyslexia support
ANRF PMECRG project led by Yogesh Kumar Meena.
Fair Federated Learning Framework in the Presence of Heterogeneous, Strategic, and Malicious Clients
ANRF early-career grant led by Manisha Padala.