Bio
Doctoral Student Email: [email protected]
About me
I am Sombit Dey, a doctoral student at the Institute for Computer Science, Artificial Intelligence and Technology from April 2023, working on Computer Vision and Robotics. I am supervised by Prof Danda Paudel and Prof Luc Van Gool.
My research interests lie in Deep Learning, Computer Vision, and Robotics. My research focus is at the intersection of robotics and foundational multi-modal models.
Education ETH Zurich | Sept 2021 – Feb 2024 MSc – Robotics Systems and Controls.
Master Thesis: Fine-grained losses for 3D visual grounding.
| IIT Kharagpur | June 2017 – May 2021 B.Tech Electronics Engineering, Minor in Computer Science |
Publications
-
Learning whom to trust in navigation: dynamically switching between classical and neural planning
-
2023 IEEE/RSJ International Conference on Intelligent Robots and Systems at Detroit, USA
-
A Proposal of FPGA-Based Low Cost and Power Efficient Autonomous Fruit Harvester
-
2020 6th International Conference on Control, Automation and Robotics (ICCAR), Singapore
**Relevant Experience **Research Intern: NaverLabs Europe, Grenoble France 09/202
Publications
-
SPEAR-1: Scaling Beyond Robot Demonstrations via 3D Understanding
2026 · IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2026)
-
AR-VLA: Autoregressive Action Expert for Vision–Language–Action Models
2026 · Robotics: Science and Systems (RSS 2026)
-
ReVLA: Reverting Visual Domain Limitation of Robotic Foundation Models
2025 · International Conference on Robotics and Automation (ICRA 2025)
-
Fine-Grained Spatial and Verbal Losses for 3D Visual Grounding
2025 · Winter Conference on Applications of Computer Vision (WACV 2025)