| { | |
| "time": "241031154353", | |
| "results": { | |
| "GPT-4o (0513, detail-high)": { | |
| "META": { | |
| "Method": [ | |
| "GPT-4o (0513, detail-high)", | |
| "https://openai.com/index/hello-gpt-4o/" | |
| ], | |
| "Parameters": "", | |
| "Language Model": "", | |
| "Vision Model": "", | |
| "Org": "OpenAI", | |
| "Time": "2024/05/31", | |
| "Verified": "Yes", | |
| "OpenSource": "No", | |
| "key": 270, | |
| "dir_name": "GPT4o_HIGH" | |
| }, | |
| "Shopping Concept Understanding": { | |
| "Rec": 67.8, | |
| "Ocr": 76.8, | |
| "Know": 58.3, | |
| "Gen": 56.9, | |
| "Spat": 74.3, | |
| "Math": 76.2, | |
| "Overall": 69.1, | |
| "Overall (official)": "N/A" | |
| }, | |
| "Shopping Knowledge Reasoning": { | |
| "Overall": 61.3, | |
| "SCI": 64.8, | |
| "TQA": 70.3, | |
| "NUM": 44.4, | |
| "ARI": 58.4, | |
| "VQA": 47.5, | |
| "GEO": 61.5, | |
| "ALG": 62.3, | |
| "GPS": 60.1, | |
| "MWP": 69.9, | |
| "LOG": 43.2, | |
| "FQA": 60.2, | |
| "STA": 68.4 | |
| }, | |
| "User Behavior Alignment": { | |
| "Text Recognition": 199, | |
| "Scene Text-centric VQA": 181, | |
| "Doc-oriented VQA": 168, | |
| "Key Information Extraction": 170, | |
| "Handwritten Mathematical Expression Recognition": 18, | |
| "Overall": 736 | |
| }, | |
| "Multi-lingual Abilities": { | |
| "Overall": 63.9, | |
| "coarse perception": 73.6, | |
| "fine-grained perception": 54.8, | |
| "instance reasoning": 66.4, | |
| "logical reasoning": 72.0, | |
| "math": 66.4, | |
| "science & technology": 50.0 | |
| } | |
| } | |
| } | |
| } |