LIBERO-PeRM Unseen Layout - User Profile SA: leaderboard
Table3 starred row: same trained LoRA checkpoint, same task and preference but unseen layout. Distinct evaluation board, not a new trained model; fixed-checkpoint ten-run mean; success-gated SA. Metric: Success-gated preference satisfaction (%). Higher is better. Source: https://arxiv.org/html/2610.22372v1
Metric: Success-gated preference satisfaction (%). Source: arxiv.org. 1 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | LIBERO-PeRM pi0.5 LoRA | 20.58 |
Interactive version: theaggregate.ai/benchmark?slug=libero-perm-unseen-layout-user-profile-sa · How It Works · Data refreshed daily, snapshot 2026-10-17.