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

Affordance Grounding on HANDAL hard (test)

62.8gIoU

A-Harness

1.02417.06233.149.138May 1, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.05
62.861.7
2026.05
60.661
2026.05
59.459.1
2026.05
58.257.8
2026.05
57.956.5
2026.05
55.355.1
2026.05
55.249.7
2026.05
49.847.4
2026.05
4842.5
2026.05
40.737.9
2026.05
27.716.7
2026.05
12.38.1
2026.05
53.5
2026.05
3.43.1