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

Goal-conditioned visual planning on CrossTask T=3 88

47.47Success Rate (SR)

GeoWorld

-1.888410.925823.7436.5542Feb 26, 2026
Updated 15d ago

Evaluation Results

MethodLinks
2026.02
47.4773.6986.55
2026.02
46.2571.9585.44
2026.02
45.5872.7469.42
2026.02
45.3370.8484.7
2026.02
44.8471.6268.87
2026.02
44.0770.1868.32
2026.02
43.8968.9682.93
2026.02
43.3368.6367.84
2026.02
40.4567.1969.17
2026.02
38.9363.879.82
2026.02
38.1264.7467.15
2026.02
37.7965.3867.45
2026.02
37.264.6766.57
2026.02
30.5559.5976.86
2026.02
28.8558.1275.36
2026.02
26.453.0274.05
2026.02
23.3449.9673.89
2026.02
21.2749.4661.7
2026.02
17.0344.6658.08
2026.02
1636.1765.91
2026.02
14.652.668.38
2026.02
12.1831.2947.48
2026.02
8.0523.332.06
2026.02
2.8924.3931.56
2026.02
2.1520.2130.87
2026.02
1.8721.6431.7
2026.02
0.010.941.66