Mar 5, 2025
Allen School Colloquium Series
Title: Discovering & Engineering the Computation Underlying Large Intelligent Agents
Speaker: Pratyusha Sharma (MIT)
Date: March 3, 2025
Abstract: The richness of language and intelligent behavior has often been attributed to latent compositional structure. Can we build tools for discovering how deep networks learn and represent this latent structure implicitly? And more importantly, can we use this knowledge to improve generalization in largely structure-less general purpose models or refine our understanding of the world they describe? In this talk, I present three perspectives to answer these questions. I will discuss experimental methods to functionally characterize the space of learnt solutions in LLMs and demonstrate how this understanding can be used to improve their empirical generalization in a gradient free manner, sometimes by as much as 30% points on language understanding benchmarks. Following that, I show how to decipher the structure of another (black box) language-like system, the naturally arising communication system of sperm whales in the wild, discovering for the first time a unique combinatorial communication system. Finally, I apply insights from these results to equip embodied agents with a latent language of thought–-hierarchical and compositional — and show how it can enable long-horizon reasoning and planning in these systems.
Bio: Pratyusha Sharma is a Ph.D. student in the Computer Science and Artificial Intelligence Lab at MIT, advised by Antonio Torralba and Jacob Andreas. She studies the interplay between language, sequential decision making and intelligence in natural and AI systems. Before this, she received her Bachelor’s degree from the Indian Institute of Technology, Delhi. Her research is published in interdisciplinary journals like Nature Communications, etc. and in academic conferences across machine learning, natural language processing, robotics, and marine biology. Her research has also been featured in articles in the New York Times, National Geographic Magazine, BBC, etc. She was recently a speaker at TED AI and was selected as a Rising Star in EECS, Data Science, and GenAI.
Her webpage, with links to her papers.
No comments:
Post a Comment