Downloads
266
HUGGING FACE DATASET
LIBERO-Para
HAI-Lab/LIBERO-Para — source facts cached from the per-dataset Hugging Face API.
DIRECT ANSWER
License: MIT; size: 1K<n<10K; creator: HAI-Lab; paper: arXiv 2603.28301.
Commercial use: Review the linked license and upstream rights.
Size category
1K<n<10K
Last modified
Apr 6, 2026
Access
Not marked as gated
SOURCE CARD
What the LIBERO-Para card says
Dataset-card excerpt
LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Models LIBERO-Para is a controlled benchmark for evaluating the paraphrase robustness of Vision-Language-Action (VLA) models. It independently varies action expressions and object references—the two core linguistic components of robotic manipulation instructions—enabling fine-grained analysis of how different types of linguistic variation affect VLA performance. 📄 Paper: arXiv:2603.28301 💻 Code:…
Captured tags
- Formats: Not provided in the captured tags
- Modalities: Not provided in the captured tags
Open the source dataset
Review the current card, files, access terms, and metadata.
ALTERNATIVES
Datasets related to LIBERO-Para
Robot training data marketplace
Robotics datasets
LIBERO-plus
Shared tags: Robotics, Language:En, Mit, Region:Us.
comet-1.5k
Shared tags: Robotics, Language:En, Mit, Size Categories:1K<N<10K.
computer-use-data-psai
Shared tags: Robotics, Language:En, Mit, Size Categories:1K<N<10K.
XVLA-Soft-Fold
Shared tags: Robotics, Language:En, Mit, Size Categories:1K<N<10K.
SandThink
Shared tags: Robotics, Language:En, Mit, Region:Us.
dynamics-probing
Shared tags: Robotics, Language:En, Mit, Size Categories:1K<N<10K.
TAGS
Metadata captured for LIBERO-Para
Sources
TRUELABEL ROUTING
Need data that matches a specific deployment?
Use the source facts and gaps from this profile to define a custom physical AI data request.