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

Temporal Spatial Reasoning on VSIBench

62.9Average Accuracy

G2F-RAG

Updated 12d ago

Evaluation Results

Method	Links
G2F-RAG 2026.04		62.9
InternVL3.5 2026.04		60.8
G2F-RAG 2026.04		60.1
Qwen3-VL 2026.04		59.4
G2F-RAG 2026.04		59
Qwen3-VL 2026.04		58.4
InternVL3.5 2026.04		56.3
InternVL3.5 2026.04		54.9
InternVL3 2026.04		48.9
Gemini1.5 Pro 2026.04		48.8
G2F-RAG 2026.04		44.8
G2F-RAG 2026.04		42.5
InternVL2.5 2026.04		41.6
InternVL3 2026.04		41.6
Qwen2.5-VL 2026.04		37.4
LLaVA-Video 2026.04		35.7
GPT-4o (2024-05-13) 2026.04		34
VILA-1.5 2026.04		28.9
Qwen2.5-VL 2026.04		27.9