MEMI Reverse-Engineers Robot Memory, Hits 97% on RoboMME (99.4% with Controller)
du_yilun · x · 2026-10-25
Liu Yuejiang's team introduces MEMI (Reverse-Engineering What Robots Should Remember), exploring what robots should remember and whether memory policies can be learned rather than hand-designed.
On the RoboMME benchmark, MEMI reaches 97.0% end-to-end and 99.4% with a low-level controller — the first time the benchmark has been nearly solved.
More from Embodied
- Nov 7 workshop: control a robot dog hands-free with an ESP32 — StewartalsopIII · 2026-10-26
- Tesla's Texas robotaxi fleet nears 1,000 vehicles, up 78% in October — mitchdeg · 2026-10-26
- Are we evaluating robot dexterity wrong? The case for superhuman tasks — DJiafei · 2026-10-26
- One prompt had AI optimize Gazebo: 68-79% CPU savings and up to 3.6x speedups, identical results — facontidavide · 2026-10-26
- Musk: Mine lunar silicon to build AI satellites — Earth launches can't scale to 500-1,000 TW a year — elonmusk · 2026-10-25
- Solo dev wires MCP into Meta Quest: AI edits appear as nodes around you in mixed reality — Potential-Art7696 · 2026-10-25