RoboGraph benchmark
Compiling and Benchmarking Task-State Horizons for Embodied Agents
Meiqi Wang, Shichao Li · 2026-08-08
- Modality
- Embodied & 3D
- Data
- Simulated
- Task · ground truth
- Plan/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
Task-state horizons explicitly vary retaining observed state, exploring hidden state and updating stale state; interventions and state-management metrics diagnose memory failures.