Vinay Kumar Verma

Currently, I am Senior Applied Scientist at Amazon, Seattle, working for the Private Brand. Prior to this I was Applied Scientist at International Machine Learning, Amazon, BLR, India.

After my Ph.D. I joined as a Post Doctoral fellow under the guidance of Prof. Carin at Duke University in the Department of Electrical and Computer Engineering.

Before joining Duke I was a Ph. D. scholar at the Department of Computer Science and Engineering, IIT Kanpur. In Ph. D. I worked under the guidance of Dr Piyush Rai. In my Ph. D. I have worked on Zero-Shot Learning, Multi-label Zero-shot Learning and Deep Model Compression.

My interest in research also includes Probabilistic Machine Learning, Deep Learning, Continual Learning, Few-Shot Learning and Computer Vision. More about me can be found in the CV.

Google Scholar (Please refer to this for the recent publication)

Recent News

Paper on title "FOCUS: Forcing In-Context Object Localization through Visual Support Constraints and Policy Optimization accepted to ICML-2026
Paper on title "Tattoo: Tool-grounded thinking prm for test-time scaling in tabular reasoning" accepted to ICLR-2026
Three paper accepted to WACV-2026
Paper on title "Reliable or Deceptive? Investigating Gated Features for Smooth Visual Explanations in CNNs" accepted to CVPRW-25.
Paper on title "MoEMoE: Question Guided Dense and Scalable Sparse Mixture-of-Expert for Multi-source Multi-modal Answering" accepted to NAACL-25.
Paper on title "Convolutional Prompting meets Language Models for Continual Learning" accepted to CVPR-24.
Paper on title "Verse: Virtual-gradient aware streaming lifelong learning with anytime inference", accepted to ICRA-24.
Paper on title "Eﬃcient Expansion and Gradient Based Task Inference for Replay Free Incremental Learning", accepted to WACV-24.
Paper on title "CoD: Coherent Detection of Entities from Images with Multiple Modalities", accepted to WACV-24.
Paper on title "Meta-Learned Attribute Self-Gating for Continual Generalized Zero-Shot Learning", accepted to WACV-24.
Paper on title "Exemplar-Free Continual Transformer with Convolutions", accepted to ICCV-23.
Paper on title "Streaming LifeLong Learning With Any-Time Inference", accepted to ICRA-23.
Paper on title "Pushing the Eﬃciency Limit Using Structured Sparse Convolutions", accepted to WACV-23.