publications

2026

1. Refine and Align: Confidence Calibration through Multi-Agent Interaction in VQA

   Ayush Pandey, Jai Bardhan, Ishita Jain, and 3 more authors

   In Proceedings of the AAAI Conference on Artificial Intelligence, 2026
2. Curve Skeletonization in Continuous domain for Meshes and Point Clouds

   Jai Bardhan, Ramya Hebbalaguppe, and Aravind Udupa

   In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2026

2025

1. Tagging fully hadronic exotic decays of the vectorlike B quark using a graph neural network

   Jai Bardhan, Tanumoy Mandal, Subhadip Mitra, and 2 more authors

   Physical Review D, 2025
2. HEP-JEPA: A Foundation Model for Collider Physics

   J. Bardhan, R. Agrawal, A. Tilak, and 2 more authors

   In ICLR 2025 Workshop on World Models: Understanding, Modelling and Scaling, 2025
3. REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic Manipulation

   Martin Sedlacek, Pavlo Yefanov, Georgy Ponimatkin, and 7 more authors

   arXiv preprint arXiv:2512.19562, 2025

2024

1. Loss function to optimise signal significance in particle physics

   Jai Bardhan, Cyrin Neeraj, Subhadip Mitra, and 1 more author

   Machine Learning and the Physical Sciences Workshop @ NeurIPS 2024, Sep 2024
2. ReMOVE: A Reference-free Metric for Object Erasure

   Aditya Chandrasekar, Goirik Chakrabarty, Jai Bardhan, and 2 more authors

   In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Sep 2024
3. Constructing sensible baselines for Integrated Gradients

   Jai Bardhan, Cyrin Neeraj, and Subhadip Mitra

   Accepted at the AI to Accelerate Science and Engineering Workshop @ AAAI 2025, Nov 2024
4. Unsupervised and lightly supervised learning in particle physics

   Jai Bardhan, Tanumoy Mandal, Subhadip Mitra, and 2 more authors

   The European Physical Journal Special Topics, Jul 2024

   Abs DOI

   We review the main applications of machine learning models that are not fully supervised in particle physics, i.e., clustering, anomaly detection, detector simulation, and unfolding. Unsupervised methods are ideal for anomaly detection tasks—machine learning models can be trained on background data to identify deviations if we model the background data precisely. The learning can also be partially unsupervised when we can provide some information about the anomalies at the data level. Generative models are useful in speeding up detector simulations—they can mimic the computationally intensive task without large resources. They can also efficiently map detector-level data to parton-level data (i.e., data unfolding). In this review, we focus on interesting ideas and connections and briefly overview the underlying techniques wherever necessary.

5. Deep reinforcement learning in chemistry: A review

   Bhuvanesh Sridharan, Animesh Sinha, Jai Bardhan, and 3 more authors

   Journal of Computational Chemistry, Jul 2024

2023

1. Machine learning-enhanced search for a vectorlike singlet B quark decaying to a singlet scalar or pseudoscalar

   Jai Bardhan, Tanumoy Mandal, Subhadip Mitra, and 1 more author

   Phys. Rev. D, Jun 2023

   DOI