5 papers
RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
Jenai Xuning Yang, Xuning Yang, Rishit Dagli +6
The pursuit of general-purpose robotics has yielded impressive foundation models, yet simulation-based benchmarking remains a bottleneck due to rapid performance saturation and a l…
Diversity You Can Actually Measure: A Fast, Model-Free Diversity Metric for Robotics Datasets
Sreevardhan Sirigiri, Nathan Samuel de Lara, Christopher Agia +2
Robotics datasets for imitation learning typically consist of long-horizon trajectories of different lengths over states, actions, and high-dimensional observations (e.g., RGB vide…
RoboArena: Distributed Real-World Evaluation of Generalist Robot Policies
Pranav Atreya, Karl Pertsch, Tony Lee +29
Comprehensive, unbiased, and comparable evaluation of modern generalist policies is uniquely challenging: existing approaches for robot benchmarking typically rely on heavy standar…
Robot Policy Evaluation for Sim-to-Real Transfer: A Benchmarking Perspective
Xuning Yang, Clemens Eppner, Jonathan Tremblay +3
Current vision-based robotics simulation benchmarks have significantly advanced robotic manipulation research. However, robotics is fundamentally a real-world problem, and evaluati…
GraspGen: A Diffusion-based Framework for 6-DOF Grasping with On-Generator Training
Adithyavairavan Murali, Balakumar Sundaralingam, Yu-Wei Chao +7
Grasping is a fundamental robot skill, yet despite significant research advancements, learning-based 6-DOF grasping approaches are still not turnkey and struggle to generalize acro…