Ashish Papanai
Ashish Papanai — photo by Anushka Bhandari

Ashish Papanai

MSc Machine Learning · University of Tübingen
Research at the CVML Lab & Autonomous Vision Group · ELIZA Masters Scholar

I am a Master's student in Machine Learning at the University of Tübingen, supported by the ELIZA research-based Master's Scholarship. Previously, I was an Associate ML Scientist II at the Wadhwani Institute for Artificial Intelligence, where I built computer-vision systems for agricultural AI — crop monitoring, pest detection, and yield estimation — with a focus on model optimization and real-world deployment.

I currently work with Prof. Hilde Kuehne at the CVML Lab on fine-grained attribution for Omni-Language Models. Broadly, I care about human-centric, reliable, and ethical AI for social impact, and about bridging theoretical ML with deployable systems.

I am also part of the Autonomous Vision Group at the University of Tübingen, where I work with Gege Gao on physics-grounded post-training and mechanistic world models.

News

See more newsShow less news
  • Sep 2025Wrapped up two-and-a-half years at Wadhwani AI, where I helped ship the CottonAce crop-monitoring system nationally.
  • Jan 2024Paper on multistage uncertainty-aware inference accepted at the AAAI 2024 Workshop on Deployable AI.
  • Oct 2023Work on gall bladder cancer detection from ultrasound with only image-level labels appeared at MICCAI 2023.

Selected Works

Work
SuperSAVVY: Training Dataset for Audio-Visual Question Answering Using Geometrically Labelled Egocentric Videos
To be added
AAAI '24
Maintaining User Trust Through Multistage Uncertainty Aware Inference
C. Agrawal, A. Papanai, J. White
AAAI Workshop on Deployable AI
MICCAI '23
Gall Bladder Cancer Detection from US Images with Only Image Level Labels
S. Basu, A. Papanai, M. Gupta, P. Gupta, C. Arora
MICCAI 2023
DELCON '23
VIhanceD: Efficient Video Super Resolution for Edge Devices
A. Papanai, S. Babbar, P. Anshuman, H. Kathuria, A. Sharma, N. Gupta
IEEE Delhi Section Flagship Conference (DELCON 2023)

Education

Research Interests

Multimodal Systems Fine-Grained Attribution Omni-Language Models World Models Physics-Grounded Post-Training Mechanistic World Models Computer Vision Medical Image AI AI for Social Good Model Explainability