YZW00/RoboShackles

RoboShackles RoboShackles is a safety benchmark for evaluating whether Embodied Foundation Models can refuse unsafe robot instructions and anticipate human-injury risks before action execution. Paper: https://arxiv.org/abs/2606.18632 Dataset The full RoboShackles dataset contains 10,000 safety-critical robotic video clips. This public Hugging Face release contains only the test split: 1,200 videos, with 200 videos per category. We do not publicly release the… See the full description on the dataset page: https://huggingface.co/datasets/YZW00/RoboShackles.

<iframe src="https://silytics.si/embed/item/a84fba98-a9cf-4870-96d9-3c07d78e440b" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 16 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.