Profile picture
Administrative data

Kaustubh Shripadraj Ponkshe

kaustubh.ponkshe@epfl.ch

EPFL › ETU › EDOC › EDIC

Current Work

I want to understand how language models learn. My research rethinks the training pipeline of foundation models — pretraining, midtraining, and post-training:

  • Knowledge-free & retrieval-grounded pretraining — training models that lean on external context instead of memorizing the world, to reduce parametric memorization and externalize memory.
  • Agentic pretraining — teaching models tool use and on-policy decision-making beyond passive next-token prediction.
  • Data & training dynamics — how midtraining and post-training data mixtures interact to shape reasoning and generalization.