DIGDEEP / ROBOT USE LABSource ↗
AN INDEPENDENT EXPERIMENT · 16 NOV 2026

Same robot.
Different instruction.
Different winner.

General-purpose models are learning to act. Explore where PEAR's evidence supports that bet, and where specialist policies still lead.

Published-data analysis. No model is running in this browser.
REPORTED RESULTS

What changed?

PEAR LIBERO-only · xhigh
π0.5 Published baseline

Success rate (%). PEAR paper Tables 2–3; π0.5 values cited from prior work. Different training and compute prevent an architecture-only conclusion.


YOUR WORKLOAD CHANGES THE ANSWER

How often does the goal change?

A hypothetical mix of standard tasks and changed-goal tasks. Move the slider to weight the two published results.

All standardAll changed goals

Crossover ≈ 28.4% in this arithmetic model. This is not a measured deployment threshold.

THE THESIS

Robot use could become
a general model capability.

My bet is on general intelligence gaining physical experience. VLAs and WAMs already use foundation models and remain valuable approaches. This experiment examines the evidence without declaring the problem solved.

THE FOUR ORIGINAL LINKS
01 / Blog Method and demonstrations ↗02 / Paper Results, protocols and failures ↗03 / Code Official PEAR infrastructure ↗04 / Weights Three released checkpoints ↗