Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Grounding on FigTalk-Gold All (reference set)
Loading...
74.3
Macro F1
MINARD
39.044
48.197
57.35
66.503
Jun 10, 2026
Macro F1
Macro Precision
Macro Recall
Updated 1mo ago
Evaluation Results
Method
Method
Links
Macro F1
Macro Precision
Macro Recall
MINARD
Backbone=Gemini
2026.06
74.3
78.3
78.7
MINARD
Backbone=Claude
2026.06
69.3
73.6
75.1
SAM Seg + BBox
Backbone=Claude
2026.06
67
79.7
66.8
MINARD
Backbone=GPT
2026.06
66.6
72.3
71.2
SAM Seg + BBox
Backbone=Gemini
2026.06
62.5
79.2
61.7
VLM-Grounding
Backbone=GPT
2026.06
59.3
60
70.8
VLM-Grounding
Backbone=Gemini
2026.06
55.4
62
57.9
SAM Seg + BBox
Backbone=GPT
2026.06
52.1
73.2
48.7
VLM-Grounding
Backbone=Claude
2026.06
40.4
52.8
38.6
Feedback
Search any
task
Search any
task