The workshop will consist of 7 speaker talks, 1 spotlight session for papers that are awarded oral presentations, 1 poster session for all papers (which doubles as a coffee break), 2 debates (one on the role of humans in data collection, generation, and curation, and another on data collection in simulation vs. real world), and a coffee break.
Debate 1: Data generation in simulation vs. real world. Where should data collection take place, and what kinds of data should be collected? What are some opportunities and challenges for data collection in simulation and in the real world?
Debate 2: The role of humans in data collection, generation, and curation. Given the advent of powerful generative AI models, to what extent should humans be involved at different parts of the robotics data collection stack, and in what capacity?
09:00am - 09:10am: Introduction
09:10am - 09:30am: Invited Speaker 1: Caelan Garrett (NVIDIA)
09:30am - 09:50am: Invited Speaker 2: Mayank Mittal (ETH Zurich, NVIDIA)
10:00am - 10:30am: Coffee Break
10:30am - 10:50am: Invited Speaker 3: Katerina Fragkiadaki (CMU)
10:50am - 11:10am: Invited Speaker 4: Edward Johns (Imperial College London)
11:10am - 11:30am: Invited Speaker 5: Roberto Martin-Martin (UT Austin)
11:45am - 12:30pm: Debate 1: Data generation in simulation vs. real world
Panelists: Mayank Mittal, Katerina Fragkiadaki, Edward Johns, Roberto Martin-Martin
12:30pm - 02:00pm: Lunch
02:00pm - 02:20pm: Invited Speaker 6: Nathan Tsoi (Yale)
02:20pm - 03:00pm: Spotlights
03:00pm - 03:30pm: Poster Session 1
03:30pm - 04:00pm: Poster Session 2 and Coffee Break
04:00pm - 04:20pm: Invited Speaker 7: Chelsea Finn (Pi, Stanford)
04:20pm - 05:00pm: Debate 2: The role of humans in data collection, generation, and curation.
Panelists: Edward Johns, Roberto Martin-Martin, Nathan Tsoi, Chelsea Finn
ExploRLLM: Guiding Exploration in Reinforcement Learning with Large Language Models
Continuously Improving Mobile Manipulation with Autonomous Real-World RL
TAPAS: A Dataset for Task Assignment and Planning for Multi Agent Systems
CyberDemo: Augmenting Simulated Human Demonstration for Real-World Dexterous Manipulation
TeleMoMa: A Modular and Versatile Teleoperation System for Mobile Manipulation
TRANSIC: Sim-to-Real Policy Transfer by Learning from Online Correction
The COLOSSEUM: A Benchmark for Evaluating Generalization for Robotic Manipulation
RT-Sketch: Goal-Conditioned Imitation Learning from Hand-Drawn Sketches
RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots
Evaluating Real-World Robot Manipulation Policies in Simulation
DexCap: Scalable and Portable Mocap Data Collection System for Dexterous Manipulation
Evaluating a VR System for Collecting Safety-Critical Vehicle-Pedestrian Interactions
Manipulate-Anything: Automating Real-World Robots using Vision-Language Models
Grounding Language Plans in Demonstrations Through Counterfactual Perturbations
Efficient Data Collection for Robotic Manipulation via Compositional Generalization
RVT-2: Learning Precise Manipulation from Few Demonstrations
Scaling Robot-Learning by Crowdsourcing Simulation Environments
LaNMP: A Multifaceted Mobile Manipulation Benchmark for Robots
From Imitation to Refinement – Residual RL for Precise Visual Assembly
Safe & Accurate at Speed with Tendons: A Robot Arm for Exploring Dynamic Motion