RoboTwin 2.0

'RoboTwin 2.0' is a dataset/benchmark that contains diverse robotic manipulation scenarios used to evaluate the performance of action generation models in predicting task-relevant interactions and dynamics.

Papers using RoboTwin 2.0 (2)

SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning2025 · 83 citesAIM: Intent-Aware Unified world action Modeling with Spatial Value Maps2026