Avik Pal

MSc Artificial Intelligence @ University of Amsterdam

prof_pic.webp
@Naples with Mt. Vesuvius in the background

I am currently pursuing a Master’s degree in Artificial Intelligence at the University of Amsterdam. Prior to this, I worked as a Machine Learning Engineer at Quantiphi in the NVIDIA vertical, where I focused on applied AI research. I hold a B.Tech. in Electronics and Communication Engineering from the National Institute of Technology, Nagpur, completed in 2020.

My primary research interests include computer vision, deep learning, multi-modal learning, and foundation models. As part of my Master’s thesis, I am exploring multi-modal understanding through hyperbolic geometry under the supervision of Dr. Pascal Mettes and Prof. Fabio Galasso as an ELLIS MSc Honours student. I am currently applying for PhD positions, and I invite reviewers to explore my profile.

Outside of my academic pursuits, I enjoy reading fiction, watching space documentaries and movies, and traveling.

selected publications

  1. compositional_entailment_learning_hyperbolic_preview.png
    Compositional Entailment Learning for Hyperbolic Vision-Language Models
    Avik Pal, Max Spengler , Guido Maria D’Amely Melendugno , and 3 more authors
    2024
    arXiv
  2. in_context_improves_comp_understanding_of_VL_preview.png
    In-Context Learning Improves Compositional Understanding of Vision-Language Models
    Matteo Nulli , Anesa Ibrahimi , Avik Pal, and 2 more authors
    In ICML 2024 Workshop on Foundation Models in the Wild , 2024
    arXiv HTML Code
  3. active_label_cleaning_framework_preview.png
    Improving Noisy Fine-Grained Datasets using Active Label Cleaning Framework
    Avik Pal
    In CVPR 2024 Workshop on Vision Datasets Understanding , 2024
    HTML Code