Lyrics: Boosting Fine-grained Language-Vision Alignment and Comprehension via Semantic-aware Visual Objects

Open paper
Benchmark Model Rank Results
image-captioning-on-cocoLyrics–CIDEr: 121.1
image-captioning-on-nocaps-entireLyrics–CIDEr: 126.8
visual-question-answering-on-gqa-test-devLyrics–Accuracy: 62.4
visual-question-answering-on-ok-vqaLyrics–Accuracy: 58.2
visual-question-answering-on-vqa-v2-test-devLyrics–Accuracy: 81.2

Rank counts only results with a code link.