List your company

Aerial Visual Question Answering on AerialVP

96.73AccuracyGPT-4.1
23.004442.144761.28580.4253Dec 8, 2025

Evaluation results

MethodLinks
GPT-4.1Semantic Enhancement=true, Spatial Relationship Enhancement=true2025.1296.73
InternVL3-14BSemantic Enhancement=true, Spatial Relationship Enhancement=true2025.1296.66
InternVL3-8BSemantic Enhancement=true, Spatial Relationship Enhancement=true2025.1295.4
Gemini-2.0Semantic Enhancement=true, Spatial Relationship Enhancement=true2025.1295.35
LLaVA-OneVision-7BSemantic Enhancement=true, Spatial Relationship Enhancement=true2025.1294.48
Qwen2-VL-7BSemantic Enhancement=true, Spatial Relationship Enhancement=true2025.1293.74
Claude-3.7Semantic Enhancement=true, Spatial Relationship Enhancement=true2025.1290.38
LLaVA-1.5-7BSemantic Enhancement=true, Spatial Relationship Enhancement=true2025.1290.06
GPT-4.1Semantic Enhancement=false, Spatial Relationship Enhancement=false2025.1282.94
SAILVL-8BSemantic Enhancement=true, Spatial Relationship Enhancement=true2025.1282.76
GPT-4oSemantic Enhancement=true, Spatial Relationship Enhancement=true2025.1280.53
Qwen2.5-MaxSemantic Enhancement=true, Spatial Relationship Enhancement=true2025.1277.54
InternVL3-14BSemantic Enhancement=false, Spatial Relationship Enhancement=false2025.1274.93
InternVL3-8BSemantic Enhancement=false, Spatial Relationship Enhancement=false2025.1274.91
Gemini-2.0Semantic Enhancement=false, Spatial Relationship Enhancement=false2025.1273.82
LLaVA-OneVision-7BSemantic Enhancement=false, Spatial Relationship Enhancement=false2025.1273.37
Claude-3.7Semantic Enhancement=false, Spatial Relationship Enhancement=false2025.1271.7
Qwen2-VL-7BSemantic Enhancement=false, Spatial Relationship Enhancement=false2025.1269.76
Qwen2.5-MaxSemantic Enhancement=false, Spatial Relationship Enhancement=false2025.1265.95
GPT-4oSemantic Enhancement=false, Spatial Relationship Enhancement=false2025.1260.03
LLaVA-1.5-7BSemantic Enhancement=false, Spatial Relationship Enhancement=false2025.1251.64
SAILVL-8BSemantic Enhancement=false, Spatial Relationship Enhancement=false2025.1225.84