PaLI-3

Reported on 5 benchmarks across 3 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing4 results

Question AnsweringonNExT-QA
WUPS· uses extra data· 2023-10-13
37.7
best: 38.3 (PaLI-X)
PaLI-3 Vision Language Models: Smaller, Faster, Stronger arXiv:2310.09199
Visual Question Answering (VQA)onDocVQA test
ANLS· 2023-10-13
0.876
best: 0.9436 (Human)
PaLI-3 Vision Language Models: Smaller, Faster, Stronger arXiv:2310.09199
Visual Question Answering (VQA)onInfographicVQA
ANLS· 2023-10-13
57.8
best: 80.3 (Gemini Ultra (pixel only))
PaLI-3 Vision Language Models: Smaller, Faster, Stronger arXiv:2310.09199
Visual Question Answering (VQA)onChartQA
1:1 Accuracy· 2023-10-13
70
best: 81.3 (ChartPaLI-5B + PaLM 2-S)
PaLI-3 Vision Language Models: Smaller, Faster, Stronger arXiv:2310.09199

Computer Code1 result

Chart Question AnsweringonChartQA
1:1 Accuracy· 2023-10-13
70
best: 81.3 (ChartPaLI-5B + PaLM 2-S)
PaLI-3 Vision Language Models: Smaller, Faster, Stronger arXiv:2310.09199