DILLO LIBERO Distillation Dataset This dataset contains LIBERO policy rollouts labeled for DILLO (DIstiLLed Language ActiOn World Model). Each example stores a chunked ACT policy rollout, boundary frame images, robot state traces, action chunks, and VLM generated descriptions/reasoning for distillation. Dataset Summary Total episodes: 1700 Total tasks: 130 Total files: 33079 Approximate size: 555.3 MiB Hugging Face repo: Sapienza/DILLO LIBERO dataset Layout: root/LIBERO {SUITE}/{TASK}/{EPISODE} Suites LIBERO 10 : 200 episodes across 10 tasks LIBERO 90 : 900 episodes across 90 tasks LIBERO GOAL : 200 episodes across 10 tasks LIBERO OBJECT : 200 episodes across 10 tasks LIBERO SPATIAL : 200 episodes across 10 tasks File Structure Each example is stored as: Root metadata files: metadata.jsonl : one row per episode with relative paths, shapes, task text, and success label. metadata.csv : CSV version of the same manifest. dataset summary.json : aggregate counts by suite/task and file extension. .gitattributes : LFS rules for binary arrays and images. File Types .jpeg : 19479 .npy : 8500 .txt : 5100 Example Rows LIBERO 10/KITCHEN SCENE3 turn on the stove and put the moka pot on it/000 :…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy