• Skip to primary navigation
  • Skip to content
  • Skip to footer
  • Publications
  • Education
  • Experience
  • Updates
  • CV
    Soumadeep Saha

    Soumadeep Saha

    Research Fellow,
    Indian Statistical Institute,
    Kolkata

    • Email
    • Google Scholar
    • GitHub
    • LinkedIn
    • Twitter
    • Instagram

      2025

    • Preprint July 2025

      Check out our pre-print titled - KisMATH: Do LLMs Have Knowledge of Implicit Structures in Mathematical Reasoning?.

    • Dissertation accepted July 2025

      My dissertation titled “Domain Obedient Deep Learning” has been accepted by the PhD/DSC committee, ISI Kolkata.

    • Preprint May 2025

      Check out our pre-print titled - sudoLLM : On Multi-role Alignment of Language Models.

    • Presenting at NAACL 2025 April 2025

      [Poster] [Video].

    • OpenAI Grant February 2025

      Thanks to OpenAI for providing us with a computational credit grant under their Researcher Access Program.

    • Paper accepted to NAACL January 2025

      Our paper titled Language Models are Crossword Solvers was accepted to NAACL 2025 (main).

    • 2024

    • Preprint December 2024

      Check out our pre-print titled - Deep Learning Based Recalibration of SDSS and DESI BAO Alleviates Hubble and Clustering Tensions.

    • HIDA Helmholtz visiting researcher July 2024

      Visiting German Aerospace Institute (DLR), Cologne, Germany.

    • Attending AAAI 2024 February 2024

      Attending AAAI 2024, at Vancouver, BC, Canada. Presenting at the W3PHIAI workshop.

    • Email
    • GitHub
    © 2025 Soumadeep Saha. Powered by Jekyll & Minimal Mistakes.