About me

I'm a 3rd year Ph.D. student at Cornell Tech in NYC, advised by Prof. Volodymyr Kuleshov. My research focuses on Generative AI for text and images.

Projects

Select Papers

  • Subham S. Sahoo, Marianne Arriola, Yair Schiff, Aaron Gokaslan, Edgar Marroquin, Justin T Chiu, Alexander Rush, Volodymyr Kuleshov. Simple and Effective Masked Diffusion Language Models. Neural Information Processing Systems (NeurIPS - 2024), 2024. [paper, code, project]


    Subham S. Sahoo, Aaron Gokaslan, Chris De Sa, Volodymyr Kuleshov. Diffusion Models With Learned Adaptive Noise. Neural Information Processing Systems (NeurIPS - 2024, spotlight), 2024. [paper, code, project]


    Subham S. Sahoo*, Anselm Paulus*, Marin Vlastelica, Vit Musil, Volodymyr Kuleshov, Georg Martius. Backpropagation through Combinatorial Algorithms: Identity with Projection Works. International Conference on Learning Representations (ICLR - 2023), 2023. [paper, code]


    Subham S. Sahoo, Subhashini Venugopalan, Li Li, Rishabh Singh, Patrick Riley. Scaling Symbolic Methods using Gradients for Neural Model Explanation. International Conference on Learning Representations (ICLR - 2021), 2021. [paper, code]


    Subham S. Sahoo, Christoph H. Lampert, Georg Martius. Learning Equations for Extrapolation and Control. International Conference on Machine Learning (ICML - 2018), 2018. [paper, project, code]

Background

Education

  1. Cornell Tech, New York, USA.

    2022 — present

    Ph.D in Computer Science.
    Thesis: Diffusion Language Models.
    Committee: Prof. Volodymyr Kuleshov (chair), Prof. Noah Snavely, Prof. Bart Selman.

  2. Indian Institute of Technology - Kharagpur, India.

    2015 — 2019

    Bachelor's in Electrical Engineering.

Experience

  1. Cruise, San Francisco, USA.

    2023 (May - July)

    Research intern.
    Team: AV Behaviors.

  2. Max Planck Institute for Intelligent Systems, Tubingen, Germany.

    2021 (Aug - Dec)

    Visiting Researcher.
    Team: Autonomous Learning Group.

  3. Google Research, Mountain View, USA.

    2019 — 2021

    AI Resident.
    Teams: Accelerated Science, Operations Research.

Papers

  • Subham S. Sahoo, Marianne Arriola, Yair Schiff, Aaron Gokaslan, Edgar Marroquin, Justin T Chiu, Alexander Rush, Volodymyr Kuleshov. Simple and Effective Masked Diffusion Language Models. Neural Information Processing Systems (NeurIPS - 2024), 2024. [paper, code, project]


    Subham S. Sahoo, John X. Morris, Aaron Gokaslan, Vitaly Shamtikov, Volodymyr Kuleshov. Gradient-Free Classifier-Based Guidance for Diffusion Models. Under Review, 2024.


    Subham S. Sahoo, Aaron Gokaslan, Chris De Sa, Volodymyr Kuleshov. Diffusion Models With Learned Adaptive Noise. Neural Information Processing Systems (NeurIPS - 2024), 2024. [paper, code, project]

  • Subham S. Sahoo, Anselm Paulus, Marin Vlastelica, Vit Musil, Volodymyr Kuleshov, Georg Martius. Backpropagation through Combinatorial Algorithms: Identity with Projection Works. International Conference on Learning Representations (ICLR - 2023), 2023. [paper, code]


    Phillip Si, Zeyi Chen, Subham S. Sahoo, Subham S. Sahoo, Yair Schiff, Volodymyr Kuleshov. Semi-Autoregressive Energy Flows: Towards Determinant-Free Training of Normalizing Flows. International Conference on Machine Learning (ICML - 2023), 2023. [paper]

  • Subham S. Sahoo, Subhashini Venugopalan, Li Li, Rishabh Singh, Patrick Riley. Scaling Symbolic Methods using Gradients for Neural Model Explanation. International Conference on Learning Representations (ICLR - 2021), 2021. [paper, code]


    Subham S. Sahoo, Ross Anderson, Christian Tjandraatmadja. Local Search on TPUs. pre-print, 2021. [paper]

  • Subham S. Sahoo. Training Neual Networks using SAT solvers. pre-print, 2018. [paper]


    Subham S. Sahoo, Christoph H. Lampert, Georg Martius. Learning Equations for Extrapolation and Control. International Conference on Machine Learning (ICML - 2018), 2018. [paper, project, code]