Visual Reasoning On Winogavil
Metrics
Jaccard Index
Results
Performance results of various models on this benchmark
Model Name | Jaccard Index | Paper Title | Repository |
|---|---|---|---|
| CLIP-ViL (Zero-Shot) | 15 | WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models | |
| CLIP-RN50x64/14 (Zero-Shot) | 38 | WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models | |
| CLIP-ViT-L/14 (Zero-Shot) | 40 | WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models | |
| CLIP-ViT-B/32 (Zero-Shot) | 41 | WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models | |
| Humans | 90 | WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models | |
| ViLT (Zero-Shot) | 52 | WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models | |
| X-VLM (Zero-Shot) | 46 | WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models | |
| CLIP-RN50 (Zero-Shot) | 35 | WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models |
0 of 8 row(s) selected.