WorldLines (semantic states)
WorldLines: Benchmarking and Modeling Long-Horizon Stateful Embodied Agents
Yehang Zhang, Jianchong Su, Haojian Huang, Yifan Chang, Tianhao Zhou, Xinli Xu, et al. · 2026-08-21
- Modality
- Embodied & 3D
- Data
- Simulated
- Task · ground truth
- QA/Plan · Ans.+State+Act.
- Scope
- semantic state comparator
What it tests
- CECross-modal evidenceNot 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
Household event traces support questions and planning over temporally changing semantic states.