robotics
The Robot Remembered the Scene but Mishandled the Update
MEMOBench labels storage, update and compression separately across 4,200 checkpoints; its strongest memory baseline averaged 31.9 percent success.
Summary
MEMOBench labels storage, update and compression separately across 4,200 checkpoints; its strongest memory baseline averaged 31.9 percent success.
The suite contains 30 history-dependent manipulation tasks, 1,500 expert demonstrations and checkpoints from 84 templates. High storage scores often coexisted with weak updating and compression, separating memory errors from manipulation failure.
Why it matters
MEMOBench labels storage, update and compression separately across 4,200 checkpoints; its strongest memory baseline averaged 31.9 percent success.
Limits and context
No additional limitation was separately recorded.
Key claims
MEMOBench labels storage, update and compression separately across 4,200 checkpoints; its strongest memory baseline averaged 31.9 percent success.
Evidence: source-2026-09-09-020
Sources
- arXiv preprint 2609.07047arXiv · primary research
Corrections
No corrections have been recorded for this story.