Keerthiram Murugesan

Research Scientist, IBM Research.

keerti.png

firstname.lastname@ibm.com

IBM Research AI - Denver, CO

Hi, welcome to my personal webpage.

I am a Research Scientist at IBM Research AI specializing in Artificial Intelligence (AI), Machine Learning, and Natural Language Understanding. My research focuses on building reliable models and agents capable of adapting to changes and uncertainties in real-time scenarios. I got my Ph.D. in Language Technologies and Machine Learning from School of Computer Science, Carnegie Mellon University, Pittsburgh, USA, in 2018.

My current research includes trustworthy foundational models with an emphasis on mitigating risks in Large Language Models (LLM) , Neuro-symbolic AI based on Thinking Fast and Slow, and reasoning/planning in LLMs to advance their reliability and interpretability.

I collaborate with my academic friends frequently. I also serve as a principal investigator or co-principal investigator for several ongoing projects with universities. If you would like to collaborate, please reach out to me via email.

Selected publications

  1. Text-based rl agents with commonsense knowledge: New challenges, environments and baselines
    Keerthiram Murugesan, Mattia Atzeni, Pavan Kapanipathi, and 6 more authors
    In Proceedings of the AAAI Conference on Artificial Intelligence, 2021
  2. Thinking fast and slow in AI
    Grady Booch, Francesco Fabiano, Lior Horesh, and 8 more authors
    In Proceedings of the AAAI Conference on Artificial Intelligence, 2021
  3. Plansformer: Generating symbolic plans using transformers
    Vishal Pallagani, Bharath Muppasani, Keerthiram Murugesan, and 5 more authors
    arXiv preprint arXiv:2212.08681, 2022
  4. ACL
    STARLING: Self-supervised Training of Text-based Reinforcement Learning Agent with Large Language Models
    Shreyas Basavatia, Keerthiram Murugesan, and Shivam Ratnakar
    ACL 2024 Findings, 2024
  5. Granite Guardian: Comprehensive LLM Safeguarding
    Inkit Padhi, Manish Nagireddy, Giandomenico Cornacchia, and 19 more authors
    In 2025 Annual Conference of the North American Chapter of the Association for Computational Linguistics Industry Track, 2025