| chart-question-answering-on-chartqa | PaLI-X (Single-task FT w/ OCR) | #5 | 1:1 Accuracy: 72.3 |
| chart-question-answering-on-chartqa | PaLI-X (Single-task FT) | #6 | 1:1 Accuracy: 70.9 |
| chart-question-answering-on-chartqa | PaLI-X (Multi-task FT) | #7 | 1:1 Accuracy: 70.6 |
| fine-grained-image-recognition-on-oven | PaLI-X | #1 | Accuracy: 23.1 |
| temporal-casual-qa-on-next-qa | PaLI-X | #1 | WUPS: 38.3 |
| visual-question-answering-on-docvqa-test | PaLI-X (Single-task FT w/ OCR) | #12 | ANLS: 0.868 |
| visual-question-answering-on-docvqa-test | PaLI-X (Multi-task FT) | #19 | ANLS: 0.809 |
| visual-question-answering-on-docvqa-test | PaLI-X (Single-task FT) | #20 | ANLS: 0.80 |
| visual-question-answering-on-ok-vqa | PaLI-X (Single-task FT) | #2 | Accuracy: 66.1 |
| visual-question-answering-vqa-on-infographicvqa | PaLI-X (Single-task FT w/ OCR) | #7 | ANLS: 54.8 |
| visual-question-answering-vqa-on-infographicvqa | PaLI-X (Multi-task FT) | #9 | ANLS: 50.7 |
| visual-question-answering-vqa-on-infographicvqa | PaLI-X (Single-task FT) | #10 | ANLS: 49.2 |
| visual-question-answering-vqa-on-infoseek | PaLI-X | #2 | Accuracy: 24 |