RoboCurate: Harnessing Diversity with Action-Verified Synthetic Data for Robot Learning
RoboCurate: Harnessing Diversity with Action-Verified Synthetic Data for Robot Learning
This research presents RoboCurate, a framework designed to improve robot learning by validating synthetic data. It addresses inconsistencies between generated videos and action labels by executing predicted actions in a simulator and verifying motion agreement. The method also expands scene diversity using generative models with structured editing axes and employs vision-language models to pre-filter physically implausible videos. Results show substantial relative improvements in success rates across various manipulation scenarios compared to using only real data.
The paper introduces RoboCurate, a framework that verifies synthetic robot action data by executing predicted actions in a simulator and measuring motion agreement with generated videos, reporting relative success rate improvements on benchmarks including GR-1 Tabletop, DexMimicGen, ALLEX, and OpenArm.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Environment
- SimulationOpen source S3
- Data origin
- Synthetic dataOpen source S2
Reported numbers
demonstrations
300
View original evidence
red to using only real data, achieving +90.3% on GR-1 Tabletop (300 demos), +29.1% on DexMimicGen in the pre-training setup on simulation benchmarks, and +179.9% and +42.9% in the co-finetuning setup on challenging real-world ALLEX and OpenArm humanoid benchmarks, respectively.
Open source S5robots
1
View original evidence
red to using only real data, achieving +90.3% on GR-1 Tabletop (300 demos), +29.1% on DexMimicGen in the pre-training setup on simulation benchmarks, and +179.9% and +42.9% in the co-finetuning setup on challenging real-world ALLEX and OpenArm humanoid benchmarks, respectively.
Open source S5
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
What remains unknown
Dataset licensing and commercial-use rights are not established here.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
themselves. To tackle this issue, we introduce RoboCurate, a novel framework for action-verified synthetic robot data that addresses inconsistencies between generated videos and annotated actions. Specifically, RoboCurate executes the actions predicted by an inverse dynamics model (IDM) in a simulator and verifies action consistency by measuring the motion agreement between the simulator rollout and the generated vi
Open source S3
arXiv:2602.18742v2 Announce Type: replace Abstract: Synthetic data generated by video generative models has shown promise for robot learning as a scalable pipeline, but it often suffers from noisy action labels due to imperfect video generation or inaccurate action prediction. Recently, vision-language models (VLMs) have been leveraged to validate video quality, but they cannot directly evaluate the generated actions
Open source S2
red to using only real data, achieving +90.3% on GR-1 Tabletop (300 demos), +29.1% on DexMimicGen in the pre-training setup on simulation benchmarks, and +179.9% and +42.9% in the co-finetuning setup on challenging real-world ALLEX and OpenArm humanoid benchmarks, respectively.
Open source S5
Implications for data suppliers
RoboSignal interpretation and collection questions, not statements of buyer demand.
- Ask which physical-world evaluation and real-data validation the buyer requires for simulated data.
- Compare the reported units and scope before using these quantities in a budget. Recording hours, sensor-hours and trajectories are different measures.
- Validate demand and acceptance criteria with a buyer before scaling. Publication, popularity and a research result do not establish a purchase commitment.
Source：arXiv Robotics — research abstracts · arxiv.org