Today
-
Reconcile ordinal node scoring and refresh canonical data and paper assets
Yesterday
-
Publish complete VIBTARVE paper figures, tables, and downloadable bundle
jmberhold +1: Copilot Add the complete privacy-normalized publication bundle, preserving source figures, numeric results, development/evaluation separation, and all original dataset payloads.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> -
Add robotics and VLA/VLM auditing discovery tags
jmberhold +1: Copilot Add 64 relevant platform, task, evaluation, annotation, model, and modality tags. Preserve all dataset payloads.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> -
Document public VIBTARVE access and authentication-free loading
-
Publish VIBTARVE snapshot: 488 recordings, human annotations, frames, and videos
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Add files using upload-large-folder tool
-
Configure LFS for the VIBTARVE dataset snapshot
-
initial commit