Maitrey Gramopadhye मैत्रेय ग्रामोपाध्ये

Maitrey Gramopadhye मैत्रेय ग्रामोपाध्ये

Research

MARCER: Multimodal Augmented Reality for Composing and Executing Robot Tasks

Maitrey Gramopadhye*, Bryce Ikeda*, LillyAnn Nekervis, Daniel Szafir

Published in ACM/IEEE International Conference on Human Robot Interaction (HRI), 2025

In this paper, we design and evaluate a novel interactive and multimodal end-user robot programming system. MARCER combines Trigger-Action Programming, Large Language Models and Augmented Re- ality to allow users to author and visualize reactive robot behavior.

Download here

Human Robot Collaboration with Few-Shot LLM Robot Models

Maitrey Gramopadhye, Daniel Szafir

Published in Human-Interactive Robot Learning (HIRL) | HRI, 2024

In this paper, we design a system that enables people to interact with an intelligent robot. We conduct a human subjects study to gain insights into the participants’ mental model.

Download here

Assessing the Impact of VR Interfaces in Human-Drone Interaction

Maitrey Gramopadhye, Arran Zeyu Wang, Leonard Shearer, Tony Qin and Daniel Szafir

Published in Horizons of an Extended Robotics Reality (XR-ROB Workshop) | IROS, 2023

In this paper, we designed a novel VR interface to control a 6-DOF drone and explored the impact and differences of VR and 2D interfaces on layman human-drone interaction.

Download here

Generating Executable Action Plans with Environmentally-Aware Language Models

Maitrey Gramopadhye, Daniel Szafir

Published in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023

In this paper, we propose an approach to utilise large language models and convert high level tasks to environmentally-aware action plans that can be directly mapped to executable agent actions. Our approach involves integrating environmental objects and object relations as additional inputs into LLM action plan generation to provide the system with an awareness of its surroundings.

Download here

CuRL: Coupled Representation Learning of Cards and Merchants to Detect Transaction Frauds

Maitrey Gramopadhye, Shreyansh Singh, Kushagra Agarwal, Nitish Srivasatava, Alok Singh, Siddhartha Asthana and Ankur Arora

Published in 30th International Conference on Artificial Neural Networks (ICANN), 2021

In this paper, we propose CuRL and tCuRL, coupled representation learning methods that can effectively capture the higher-order interactions in a bipartite graph of payment entities to detect transaction fraud.

Download here