Grounding

Benchmarks

Dataset Name	SOTA Method	Metric
ScreenSpot Pro		Average Grounding Accuracy51.8	82	1mo ago
ScreenSpot V2	GUIrilla-See (7B)	Grounding Accuracy94.73	50	1mo ago
ScreenSpot v2 (test)	UI-Venus-Ground-72B	Overall Accuracy95.3	29	3mo ago
RefCOCO	TIES Merging	Score90.06	28	4mo ago
RefCOCO (val)	CogVLM-G	Accuracy92.8	23	17d ago
OSWorld	MGA w/ GPT-5	Overall Score64.7	22	3mo ago
OSworld G-R	UI-Venus-1.5-30B-A3B	Accuracy76.4	22	1mo ago
E.T. Bench	MeCo (ETChat 7B)	TVG F162.5	20	4mo ago
refCOCO (testB)	UNINEXT-H	Accuracy91.5	18	1mo ago
UI-Vision	UI-Venus-1.5-30B-A3B	Accuracy54.7	18	4mo ago
MMbench	MAI-UI-32B	Accuracy91.3	18	4mo ago
VenusBench GD	UI-Venus-1.5-30B-A3B	Accuracy75	18	4mo ago
OSWorld-G-Refine (test)	UI-Venus-1.5-30B-A3B	Overall Success Rate76.4	17	24d ago
UHR-Micro (test)	MAP-Agent	GD Score29.79	16	2mo ago
RGBX-Grounding	RGBX-R1	LasHeR36.46	15	4mo ago
RefCOCOg	TIES w/ DARE	Score86.77	14	4mo ago
ScreenSpot-Pro macOS	GUIrilla-See (7B)	Grounding Accuracy41.39	13	4mo ago
Breast Ultrasound same-center (val)	Echo-α-Grounding	F1 Score @ IoU=0.2552.96	12	2mo ago
Renal Ultrasound cross-center (test)	Echo-α-Grounding	F1 Score @ IoU=0.2569	12	2mo ago
FigTalk-Gold Hard (reference set)	MINARD	Macro F174.2	9	1mo ago
FigTalk-Gold Medium (reference set)	MINARD	Macro F170.4	9	1mo ago
FigTalk-Gold Easy (reference set)	MINARD	Macro F177.8	9	1mo ago
FigTalk-Gold All (reference set)	MINARD	Macro F174.3	9	1mo ago
T-SROIE dense text for robustness (test)	DocShield	mIOU9.1	8	3mo ago
T-IC13 adapted (test)	DocShield	mIoU44.5	8	3mo ago

Showing 25 of 48 rows