Jyothir S V

I am researcher at the Center for Data Science, at New York University, where I work under supervision of Prof. Yann Lecun and Prof. Kyunghun Cho.

I am interested in self supervised learning, predictive control, tuning LLMs (RLHF), mathematical reasoning in LLMs.

I did my undergraduation at IIT Mandi, where I worked under supervision of Prof. Aditya Nigam.

Feel free to reach out to me : jyothir -at- nyu

Email  /  GitHub  /  LinkedIn

Research

Gradient-based Planning with World Models

Most model predictive control (MPC) algorithms designed for visual world models have traditionally explored gradient-free population-based optimization methods, such as Cross Entropy and Model Predictive Path Integral (MPPI) for planning. We present an exploration of a gradient-based alternative that fully leverages the differentiability of the world model.

A Massively Multi-System MultiReference Data Set for Dialog Metric Evaluation

Automatic metrics for dialogue evaluation should be robust proxies for human judgments; however, the verification of robustness is currently far from satisfactory. To quantify the robustness correlation and understand what is necessary in a test set, we create and release an 8-reference dialog datase. We then train 1750 systems and evaluate them and publicly available large models on our novel test set and the DailyDialog dataset.

Joint Embedding Predictive Architectures Focus on Slow Features

In this work, we analyze performance of JEPA trained with VICReg and SimCLR objectives in the fully offline setting without access to rewards, and compare the results to the performance of the generative architecture.

Design and source code from Jon Barron's website