Memory-World
STAMP: Training Explicit Memory for Mobile GUI Agents in Controllable and Scalable Virtual Environments
Junyang Wang, Haiyang Xu, Xi Zhang, Zhaoqing Zhu, Ming Yan, Jieping Ye, et al. · 2026-05-28
- Modality
- GUI & computer use
- Data
- Simulated
- Task · ground truth
- Act. · State+Act.
- Scope
- Memory-specific benchmark
What it tests
- CECross-modal evidencePartially evaluated
- LELong-range evidenceDirectly evaluated
- SCState changesDirectly evaluated
- CFConflicting memoriesNot evaluated
- UQUnanswerable queriesNot evaluated
- MAMemory-grounded actionDirectly evaluated
- GSPersistent interface stateDirectly evaluated
- AOAction–outcome associationDirectly evaluated
- XRCross-session reusePartially evaluated
Survey notes
Transient interface information is deliberately made necessary for later actions; memory accuracy and task success are measured.