Jalend Bantupalli

I'm a CS Grad Student at UC San Diego, where my major focus is on LLM Reasoning and Program Synthesis. I was fortunate to work at Google and Microsoft before pursuing my Masters at UCSD.

At Google I've worked on Gemini Networking and at Microsoft on AI powered Dynamics 365 . I did my Bachelors from IIT Kharagpur where I was advised by Professor Animesh Mukherjee.

Email  /  CV  /  Linkedin  /  Scholar  /  Twitter  /  Github

profile photo
UCSD
Google
Microsoft
IIT KGP

Research

I'm interested in LLM Reasoning, Machine Learning methods for Program Synthesis, Reinforcement Learning and Computer Vision.

Flow of Reasoning:Training LLMs for Divergent Problem Solving with Minimal Examples
Contribution
project page / arXiv

Using a Flow of Reasoning approach, achieved state-of-the-art accuracy of 50.37% in 1D ARC program synthesis with LLMs.

Decoding Demographic un-fairness from Indian Names
Vahini Medidoddi, Jalend Bantupalli, Souvic Chakraborty, Animesh Mukherjee,
Soc Info 2022
code / arXiv

Introduced large-scale datasets of Indian names and trained transformer models to achieve state-of-the-art performance in gender and caste prediction.


Reference to source code.