| Benchmark | Model | Rank | Results |
|---|---|---|---|
| object-hallucination-evaluation-on-pope | LLaVA-NeXT-7B (AnchorPrune, 160 tokens) | #12 | F1: 87.5 |
| video-question-answering-on-tempcompass-overall | LLaVA-Video-7B (AnchorPrune, 1,024 tokens) | #14 | Accuracy: 65.4 |
| video-question-answering-on-video-mme-with-subtitles | LLaVA-Video-7B (AnchorPrune, 1,024 tokens) | #30 | Accuracy: 60.89 |
| visual-question-answering-on-mm-vet | LLaVA-NeXT-7B (AnchorPrune, 320 tokens) | #121 | GPT-4 score: 37.2 |