Shailesh

Shailesh

(Legally, it’s just Shailesh. Some systems insist otherwise, hence “Shailesh Shailesh.”)

Research Assistant @ NUS

Shailesh

(Legally, it's just Shailesh. Some systems insist otherwise, hence "Shailesh Shailesh.")

I’m a Research Assistant at the MAGIC Lab, National University of Singapore (NUS), working with Prof Jiafei Duan. I received my Bachelor of Technology in Electrical Engineering (Minor in Embedded Systems) from IIT Dhanbad.

Previously, I was a Research Intern → Research Assistant at the MARMoT Lab, NUS, working with Prof Guillaume Sartoretti. I’ve also spent time at CIR, IIIT Allahabad with Prof G.C. Nandi.

My research aims to develop foundation models that enable general-purpose robot learning. This spans three interconnected threads, vision-language-action models, world models, and test-time scaling; unified by a common goal: robots that generalize robustly to real-world manipulation tasks.

🔥 News

  • [2026/09] Our work GAM is out, check it out here.
  • [2026/09] Our work LIT is out, check it out here.
  • [2026/07] Joined the MAGIC Lab, National University of Singapore (NUS) as a Research Assistant, working with Prof Jiafei Duan.
  • [2026/07] Our work RL^2-VLA is out, check it out here.
  • [2025/11] Our work GRIM has been accepted to AAAI 2026 (Oral). See you guys in Singapore 🇸🇬
  • [2025/09] GRIM was selected as Best Paper Finalist at RemembeRL @ CoRL 2025.
  • [2025/06] Our work GRIM is out, check it out here.
  • [2025/06] Joined the MARMoT Lab, National University of Singapore (NUS), working with Prof Guillaume Sartoretti.
  • [2025/05] Our work Search-TTA was accepted to CoRL 2025, check it out here. See you guys in Seoul :) 🇰🇷

📑 Publications

* co-first authors † second authors

Breaking the Vision-Action Shortcut: LIT for Generalizable Robotics Foundation Models

Jianman Lin*, Shailesh*, Zhongyi Luo, Jiafei Duan

Under review

Paper Website

Grounded Action Model: 3D Grounding as a Foundation for Robotics

Gehao Zhang, Weikai Huang, Shailesh, Yiyan Peng, Jiafei Duan, Ranjay Krishna

Under review

Paper Website

RL^2-VLA: Adaptive RL Latent Compositional Steering with Test-Time Scaling for VLA Models

Derek MST*, Shailesh*, Srikrishna Iyer, William Wei Jie Teo, Yuanliang Ju, Qiao Gu, Guillaume Sartoretti

Under review

Paper Website

GRIM: Task-Oriented Grasping with Conditioning on Generative Examples

Shailesh, Alok Raj, Nayan Kumar, Priya Shukla, Andrew Melnik, Micheal Beetz, Gora Chand Nandi

AAAI 2026 (Oral)

RemembeRL @ CoRL 2025 Best Paper Finalist

Paper Website AAAI

Search-TTA: A Multimodal Test-Time Adaptation Framework for Visual Search in the Wild

Derek MST, Shailesh†, Boyang Liu, Alok Raj, Qi Xuan Ang, Weiheng Dai, Tanishq Duhan, Jimmy Chiun, Yuhong Cao, Florian Shkurti, Guillaume Sartoretti

CoRL 2025

Paper Website Published

💼 Experience

📖 Academic Services

🎓 Education

IIT Dhanbad

Class of 2025

Bachelor of Technology in Electrical Engineering (Minor in Embedded Systems)

  • Relevant Coursework: Data Structures & Algorithms, Probability & Statistics, Modern Control, Embedded System Design.
  • MOOCs: Reinforcement Learning Specialization (University of Alberta / AMII, Coursera).

🏅 Honors and Awards

  • Best Paper Finalist: RemembeRL @ CoRL 2025 (with GRIM)
  • Scholarship: TEEP@AsiaPlus 2024
  • 1st Position ($12,000): Robofest Gujarat 3.0 2023 (India’s Biggest Robotics Competition)
  • 1st Position: Vichesta (ROS simulation competition), Takshak 2022 (Hosted by IIT, Dhanbad)
  • 3rd Position: Robowars (BattleBots) at Concetto 2022 (Hosted by IIT, Dhanbad)

🎲 Misc