Kowndinya Renduchintala
Research Associate II, Media and Data Science Research (MDSR) Lab, Adobe Inc.
Hey, there! I am a pre-doctoral researcher in the Media and Data Science Research (MDSR) lab at Adobe. Broadly, I’m interested in:
- Advancing our scientific understanding of Language Models, and leveraging the insights to train better models that are more robust, interpretable and generalizable.
- Advancing research on leveraging insights from human cognition to build better AI, as well as revealing more insights about human cognition using AI (e.g., using interpretability methods).
My past research explored diverse facets of language models including efficient pre-training, data mixing for instruction tuning, measuring prompt sensitivity, and more recently, alternative loss functions for instruction tuning.
Prior to joining Adobe, I graduated from the Indian Institute of Technology Bombay (IIT Bombay) in 2023, with a Bachelors degree (with Honors) in Computer Science and Engineering, where I was advised by Prof. Ganesh Ramakrishnan and Prof. Rishabh Iyer.
My long-term research goal is to develop Provably Beneficial Artificial Intelligence.
If you share an interest in these topics and would like to exchange research ideas, please do not hesitate to contact me via e-mail.
- also my first time visiting the US! 
