Research

Recent & Ongoing Research

Project 1 – Tabular Foundation Models
Summary: Exploring TabPFN, TabICL, TabDPT, and related in-context learning approaches for robust tabular understanding.
Collaborators: Dr. Aditya, Dr. Adarsh
Status: Ongoing
Project 2 – CRAFT: Text-to-Parametric CAD
Summary: A corrective and robust multi-agent framework that turns natural language into editable, parametric CAD. It combines semantic understanding, parametric planning, compilation, and rendering with self-correction and component-level verification, using multi-view visual feedback to detect and repair geometric errors.
Collaborators: Dr. Aditya, Dr. Adarsh, Dr. Chinmay
Status: Accepted at ICLAD 2026
Project page: idealab-isu.github.io/CRAFT · Code: idealab-isu/CRAFT
Poster: Presented at NVIDIA GTC 2026, San Jose, CA (March 2026)
Paper: CRAFT: Corrective and Robust Multi-Agent Framework for Text-to-Parametric CAD, ICLAD 2026 (accepted May 2026)
Project 3 – VISTA-CAD
Summary: A follow-up to CRAFT on iteratively refining parametric CAD programs using visual feedback.
Collaborators: Dr. Aditya, Dr. Adarsh, Dr. Chinmay
Status: Ongoing — manuscript in preparation

Describe collaborations you are planning, currently engaged in, or have completed.

List your submitted, under-review, accepted, and published papers. You may include planned submissions too.

CRAFT: Corrective and Robust Multi-Agent Framework for Text-to-Parametric CAD, ICLAD 2026 (IEEE International Conference on LLM-Aided Design)
Submission Date: 2026-03-19 · Accepted: 2026-05-26
Status: Accepted — project page

Planned Paper Title, Target Venue
Planned Submission Date: YYYY-MM-DD
Notes: …

You can organize, edit, and extend sections as needed for your research updates, collaborations, and paper tracking.