Compliance Classification on CompliVision Average across 3 Domains (test)
63.58Macro F1LLaVA-COT
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| LLaVA-COTPrompt Variant ID=E12026.05 | 63.58 | 67.93 | 100 | 12.1053 | |
| LLaVA-COTPrompt Variant ID=C42026.05 | 63.32 | 68.27 | 100 | 10.2467 | |
| LLaVA-COTPrompt Variant ID=E22026.05 | 58.46 | 60.73 | 100 | 11.146 | |
| LLaVAPrompt Variant ID=E12026.05 | 57.38 | 65.67 | 100 | 2.3567 | |
| LLaVA-NextPrompt Variant ID=E22026.05 | 55.51 | 65.67 | 100 | 3.59 | |
| LLaVAPrompt Variant ID=C42026.05 | 54.79 | 65.93 | 100 | 0.9647 | |
| LLaVA-COTPrompt Variant ID=C12026.05 | 53.66 | 55.2 | 100 | 7.7487 | |
| LLaVAPrompt Variant ID=E22026.05 | 49.67 | 56.2 | 100 | 2.0713 | |
| LLaVA-NextPrompt Variant ID=C12026.05 | 49.47 | 62.53 | 96.33 | 0.9473 | |
| LLaMA VisionPrompt Variant ID=E12026.05 | 43.41 | 36.27 | 82.33 | 5.8927 | |
| LLaMA VisionPrompt Variant ID=C42026.05 | 43.32 | 34.27 | 79.47 | 3.2493 | |
| LLaVAPrompt Variant ID=C12026.05 | 42.66 | 46.4 | 100 | 0.578 | |
| LLaVA-NextPrompt Variant ID=C42026.05 | 42.11 | 65.07 | 100 | 1.3473 | |
| LLaVA-NextPrompt Variant ID=E12026.05 | 42.11 | 65.13 | 100 | 3.8513 | |
| LLaMA VisionPrompt Variant ID=E22026.05 | 34.45 | 33 | 93.13 | 3.6293 | |
| LLaMA VisionPrompt Variant ID=C12026.05 | 31.61 | 32.67 | 98.8 | 0.7073 |