1. RefineStat: Efficient Exploration for Probabilistic Program Synthesis. 🔗
Madhav Kanda, Shubham Ugare, Sasa Misailovic.
Accepted at ICLR 2026.
Publications
2. Constructive Distortion: Attention-Guided Image Warping for Multimodal LLMs. 🔗
Dwip Dalal, Gautam Vashishtha, Madhav Kanda, Svetlana Lazebnik, Heng Ji, Unnat Jain.
Accepted at ICLR 2026.
3. Compositional Reasoning via Joint Image and Language Decomposition. 🔗
Madhav Kanda*, Dwip Dalal*, Zhenhailong Wang, Heng Ji, Unnat Jain.
Accepted at EACL 2026.
4. Grammar-Constrained LLM Generation for Reliable and Efficient Probabilistic Program Synthesis. 🔗
Madhav Kanda, Shubham Ugare, Sasa Misailovic.
Accepted to The Workshop on Languages for Inference (LAFI), POPL 2026.
5. Towards Scalable Identification of Brick Kilns from Satellite Imagery with Active Learning. 🔗
Aditi Agarwal*, Suraj Jaiswal*, Madhav Kanda*, Nipun Batra.
Accepted at NeurIPS 2023 RealML Workshop.
6. SpiroActive: Active Learning for Efficient Data Acquisition for Spirometry. 🔗
Ankita Kumari Jain, Nitish Sharma, Madhav Kanda, Nipun Batra.
arXiv preprint, 2024.
(* denotes equal contribution)