Datasets:

perpetual-research
/

shadowbench

Tasks:

Reinforcement Learning

Image-Text-to-Text

Modalities:

Formats:

Languages:

Size:

Tags:

spatial-reasoning

embodied-reasoning

Libraries:

License:

Dataset card Data Studio Files Files and versions

shadowbench / scripts

Commit History

Nov 5 10 days ago

12:37
replay.py: singular call count; runtime of --all

spratyey

Verified signature
795a17b
12:28
ShadowBench: environment, certified targets, episodes, transcripts and tasks (part 2)

spratyey

Verified signature
3c29d68