Narayanan Palghat Parameswaran

Graduate Student at RI, CMU

prof_pic.jpeg

I am a Robotics Masters student at Carnegie Mellon University, where I work with Dr. Deva Ramanan on behavior foundational models for wheeled quadruped robots. My research broadly focuses on generalization in robotics and multimodal learning, with the goal of building scalable, data-driven systems that transfer across robots, environments, and tasks.

Previously, I was a researcher at the Indian Institute of Science (IISc), Bangalore, where I collaborated with Dr. Shishir Kolathaya on developing generalist locomotion policies for quadruped robots, emphasizing robustness across terrains and morphologies. I have also collaborated with Dr. Pulkit Agrawal at MIT on sim-to-real transfer for quadruped locomotion.

In addition to academic research, I have industry research experience at the intersection of Large Language Models and real-world systems. I collaborated with Mr. Anantharaman P.N., Founder & CEO of JN Research Labs, where I worked on applying LLM-based agents to complex enterprise data problems, particularly in transforming unstructured, domain-specific tabular data into machine-interpretable representations for downstream analytics and decision-making.

Selected Publications

  1. groqloco.gif
    GRoQ-LoCO: Generalist and Robot-agnostic Quadruped Locomotion Control using Offline Datasets
    Narayanan PP et.al.
    2025
    ArXiv
  2. hysem.png
    HySem: A context length optimized LLM pipeline for unstructured tabular extraction
    Narayanan PP et.al.
    2024
    TRL@NeurIPS 2024