Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Long-horizon Task on Spatial and Temporal Simulation Tasks Episode-level

3.576Subtask Accuracy

Cortex (full harness)

2.184482.545742.9073.26826Jul 6, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.07
3.5763.8047.38
2026.07
3.0043.9226.925
2026.07
2.7743.876.644
2026.07
2.7434.1236.866
2026.07
2.7114.2856.996
2026.07
2.2743.7466.021
2026.07
2.2384.3626.6