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

Visual Reasoning on LogicVista

52.13Accuracy

Vanilla

Updated 1mo ago

Evaluation Results

Method	Links
Vanilla 2026.04		52.13	-	100	-
Vanilla 2026.04		49.22	-	100	-
DSTP 2026.04		48.76	93.5	87.83	-
DSTP 2026.04		48.44	92.9	92.44	-
DSTP 2026.04		47.87	91.8	90.09	-
DSTP 2026.04		46.08	88.4	87	-
DSTP 2026.04		44.74	85.8	82.98	-
DSTP 2026.04		43.62	83.7	81.01	-
DSTP 2026.04		42.7	86.8	83.8	-
DSTP 2026.04		41.16	83.6	82.9	-
DSTP 2026.04		39.22	79.7	81.04	-
DSTP 2026.04		39.05	79.3	71.48	-
DSTP 2026.04		37.58	76.4	68.74	-
DSTP 2026.04		35.55	72.2	73.34	-
VisionZip (CVPR’25) 2026.04		34.45	66.1	59.54	-
DivPrune (CVPR’25) 2026.04		34	65.2	57.61	-
FastV (ECCV’24) 2026.04		33.08	63.5	56.66	-
DivPrune (CVPR’25) 2026.04		32.78	62.9	63.55	-
VisionZip (CVPR’25) 2026.04		31.76	60.9	51.04	-
FastV (ECCV’24) 2026.04		30.64	62.3	55.68	-
DivPrune (CVPR’25) 2026.04		30.64	62.3	55.24	-
FastV (ECCV’24) 2026.04		30.2	57.9	49.42	-
VisionZip (CVPR’25) 2026.04		29.53	60	60.42	-
DivPrune (CVPR’25) 2026.04		27.96	56.8	47.4	-
FastV (ECCV’24) 2026.04		26.96	54.8	45.18	-
VisionZip (CVPR’25) 2026.04		25.5	51.8	46.4	-
LLaVA-OneVision-7B 2025.10		-	-	-	32
InternLM-XComposer2.5 2025.10		-	-	-	34.7
InternVL3-8B 2025.10		-	-	-	43.6
InternVL2.5-8B 2025.10		-	-	-	36
InternVL2-8B 2025.10		-	-	-	33.6
QvQ-72B-Preview 2025.10		-	-	-	58.2
Kimi-VL-16B 2025.10		-	-	-	42.7
Qwen2-VL-7B 2025.10		-	-	-	33.3
Qwen2.5-VL-7B 2025.10		-	-	-	46.7
R1-Onevision-7B 2025.10		-	-	-	61.8
OpenVLThinker-7B 2025.10		-	-	-	44.5
SFT Baseline 2025.10		-	-	-	45.6
DRIFT 2025.10		-	-	-	47.2