Foundation Models for Decision Making

08:15 am - 09:20 am: Poster session I

09:20 am - 09:30 am: Opening remark

09:30 am - 10:00 am: Percy Liang: Generative Agents

10:00 am - 10:30 am: Ruslan Salakhutdinov: Multimodal AI Agents

10:30 am - 11:00 am: Jürgen Schmidhuber: Reinforcement Learning Prompt Engineers and Natural Language-Based Societies of Mind

11:00 am - 12:00 pm: Oral presentations

• 11:00 am - 11:10 am: Universal Visual Decomposer: Long-Horizon Manipulation Made Easy 

• 11:10 am - 11:20 am: Goal Masked Diffusion Policies for Unified Navigation and Exploration 

• 11:20 am - 11:30 am: Motif: Intrinsic Motivation from Artificial Intelligence Feedback

• 11:30 am - 11:40 am: WebArena: A Realistic Web Environment for Building Autonomous Agents

• 11:40 am - 11:50 am: Free from Bellman Completeness: Trajectory Stitching via Model-based Return-conditioned Supervised Learning

• 11:50 am - 12:00 pm: Benchmarking Large Language Models as AI Research Agents

12:00 pm - 01:00 pm: Lunch & poster session II

01:00 pm - 01:30 pm: Phillip Isola: Quasimetric Reinforcement Learning

01:30 pm - 02:00 pm: Xinyun Chen: Self-Debugging with Large Language Models

02:00 pm - 02:30 pm: Chelsea Finn: Foundation Models for Robotic Manipulation

02:30 pm - 03:00 pm: Russ Tedrake: Towards Large Behavior Models for Dexterous Manipulation

03:00 pm - 03:30 pm: Panel Discussion: Percy Liang, Phillip Isola, Russ Tedrake, Chelsea Finn, Fei Xia

03:30 pm - 04:00 pm: Kristen Grauman: Ego-Exo4D: Learning about human skill from the first- and third-person perspectives

04:00 pm - 04:15 pm: Industry Demos: MultiOn AI

04:15 pm - 05:30 pm: Poster session III