A wet-lab simulation benchmark and success-checked augmentation loop for turning 30 demonstrations per task into policy training data, reported to lift SmolVLA from 44% to 75% simulated success.
Chemistry3D: Robotic Interaction Benchmark for Chemistry Experiments [J]
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.RO 2years
2026 2representative citing papers
A simulation-trained model predicts voxel occupancy from single RGB views for transparent object grasping and transfers to real robotic setups without fine-tuning.
citing papers explorer
-
Pipette: An Embodied Simulation Platform, Benchmark, and Data-Efficient Augmentation Framework for Wet-Lab Robotics
A wet-lab simulation benchmark and success-checked augmentation loop for turning 30 demonstrations per task into policy training data, reported to lift SmolVLA from 44% to 75% simulated success.
-
Trans2Occ: Voxel Occupancy Estimation and Grasp for Transparent Objects from Simulation to Reality
A simulation-trained model predicts voxel occupancy from single RGB views for transparent object grasping and transfers to real robotic setups without fine-tuning.