RoboMME
RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies
Yinpei Dai, Hongze Fu, Jayjun Lee, Yuejiang Liu, Haoran Zhang, Jianing Yang, et al. · 2026-05-26
- Modality
- Embodied & 3D
- Data
- Hybrid (real + synthetic/simulated)
- Task · ground truth
- Act. · State+Act.
- Scope
- Memory-specific benchmark
What it tests
- CECross-modal evidenceDirectly evaluated
- LELong-range evidenceDirectly evaluated
- SCState changesDirectly evaluated
- CFConflicting memoriesNot evaluated
- UQUnanswerable queriesNot evaluated
- MAMemory-grounded actionDirectly evaluated
- SMSpatial memoryDirectly evaluated
- OSObject-state trackingDirectly evaluated
- TSTask-state trackingDirectly evaluated
Survey notes
Temporal, spatial, referential and procedural manipulation tasks require earlier observations or demonstrations.