LIBERO-Industry
Harbor: pendingA robotic manipulation benchmark with 216 tasks evaluating generalist policies under industrial constraints like moving flows and tight tolerances.
- Samples
- 216
- Released
- 2026
- Scores
- 0
LIBERO-PeRM
LIBERO-PeRM: Benchmarking Personalized Robotic Manipulation
A benchmark for personalized robotic manipulation with 312 tasks across 5 suites, evaluating preference learning via automated metrics in simulation.
How to run
ltb · libero-perm
$ npm i -g lettokenburn-cli$ ltb bench info libero-perm
External harness. No Harbor adapter has been generated for this benchmark yet. `ltb bench info` prints the known links.
Runs happen on your machine with your own subscription; only the score summary is pushed. CLI docs
Scores
0 sourced entriesNo sourced scores recorded yet.
Every number links to where it was published. Low-confidence rows are muted and need per-cell verification.
Reproduced by the community
All runsLoading community runs…
About
LIBERO-PeRM comprises five task suites (312 tasks) with procedurally generated, human-reviewed demonstrations covering motion, skill, and user-profile preferences. It evaluates zero-shot following, learning efficiency, generalization, and conflict handling.
Scoring uses automated preference-satisfaction metrics alongside task success. Experiments reveal VLA policies struggle with zero-shot personalization, while SFT enables learning, though conflict handling remains difficult.