COMPLETE RERUN OUTPUT
LIBERO ENV PREDICATE
Reset-to-success episodes
RELEASED CHECKPOINT
4-suite evaluation
0 episodes
AUDIT EVIDENCE
Sampled policy queries
Three synchronized observations per suite from the historical c1 run; this section is evidence sampling, not a complete episode video.
POLICY INPUT
Spatial
01/03
QUERY 001
INSTRUCTION
EPISODE OUTCOMES
Task-level reliability
0%0 / 0 successful
PUBLICATION ASSET
DOWNLOAD GIF