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

Spatial and Temporal Understanding on VSTI-Bench

88.1Camera Movement Accuracy (MC)

Cog3DMap-8B

Updated 25d ago

Evaluation Results

Method	Links
Cog3DMap-8B 2026.03		88.1	67.5	40.9	47.1	90.9	70.6
Cog3DMap-8B 2026.03		88.1	67.5	40.9	47.1	90.9	70.6
VLM-3R-7B 2026.03		60.6	58.8	39.4	39.6	86.5	68.6
VLM-3R-7B 2026.03		60.6	58.8	39.4	39.6	86.5	68.6
LLaVA-NeXT-Video-7B 2026.03		49.8	40	28.2	1.8	64.7	55.6
LLaVA-NeXT-Video-7B 2026.03		49.8	40	28.2	1.8	64.7	55.6
LLaVA-NeXT-Video-72B 2026.03		48.1	44	32.3	10.5	78.3	50.9
InternVL2-8B 2026.03		48	43.5	32.9	13.5	68	55
InternVL2-8B 2026.03		48	43.5	32.9	13.5	68	55
LLaVA-OneVision-7B 2026.03		47.5	41.7	29.9	19.3	62.1	49.8
LLaVA-OneVision-7B 2026.03		47.5	41.7	29.9	19.3	62.1	49.8
LLaVA-OneVision-0.5B 2026.03		46.1	36.9	16.5	32.4	50.5	39
LongVA-7B 2026.03		43.7	32.3	13.5	5.1	57.9	41.2
LongVA-7B 2026.03		43.7	32.3	13.5	5.1	57.9	41.2
InternVL2-2B 2026.03		43	38.1	17.7	27.8	54.9	47.2
VILA-1.5-8B 2026.03		42.2	37.3	30.1	27.3	50.4	36.7
VILA-1.5-8B 2026.03		42.2	37.3	30.1	27.3	50.4	36.7
Frequency 2026.03		40.7	27.4	5.4	6.2	52.2	32.4
Frequency 2026.03		40.7	27.4	5.4	6.2	52.2	32.4
GPT-4o 2026.03		37.3	38.2	29.5	23.4	58.1	42.5
GPT-4o 2026.03		37.3	38.2	29.5	23.4	58.1	42.5
Random 2026.03		36.1	-	-	-	50	36.1
Random 2026.03		36.1	-	-	-	50	36.1
LongVILA-8B 2026.03		35.4	30.5	20	11.6	52.3	33.4
LongVILA-8B 2026.03		35.4	30.5	20	11.6	52.3	33.4
VILA-1.5-40B 2026.03		28.8	38.2	28.2	15.7	65.4	53
Gemini-1.5-Flash 2026.03		24.4	32.1	28.5	20.9	52.6	33.9
Gemini-1.5-Flash 2026.03		24.4	32.1	28.5	20.9	52.6	33.9