Multimodal Performance Analysis on Aggregate Performance
100Average Normalized PerformanceVanilla
Evaluation results
| Method | Links | |
|---|---|---|
| VanillaBackbone=LLaVA-NeXT-7B, Retained Tokens=2880, Token Retention Ratio=100%2026.06 | 100 | |
| EvoCutBackbone=LLaVA-NeXT-7B, Retained Tokens=640, Token Retention Ratio=↓ 77.8%2026.06 | 98 | |
| ApETBackbone=LLaVA-NeXT-7B, Retained Tokens=640, Token Retention Ratio=↓ 77.8%2026.06 | 97.6 | |
| VisionZipBackbone=LLaVA-NeXT-7B, Retained Tokens=640, Token Retention Ratio=↓ 77.8%2026.06 | 97.1 | |
| PDropBackbone=LLaVA-NeXT-7B, Retained Tokens=640, Token Retention Ratio=↓ 77.8%2026.06 | 95.8 | |
| EvoCutBackbone=LLaVA-NeXT-7B, Retained Tokens=320, Token Retention Ratio=↓ 88.9%2026.06 | 94.9 | |
| ApETBackbone=LLaVA-NeXT-7B, Retained Tokens=320, Token Retention Ratio=↓ 88.9%2026.06 | 94.2 | |
| VisionZipBackbone=LLaVA-NeXT-7B, Retained Tokens=320, Token Retention Ratio=↓ 88.9%2026.06 | 93.3 | |
| EvoCutBackbone=LLaVA-NeXT-7B, Retained Tokens=160, Token Retention Ratio=↓ 94.4%2026.06 | 91.4 | |
| PDropBackbone=LLaVA-NeXT-7B, Retained Tokens=320, Token Retention Ratio=↓ 88.9%2026.06 | 90.4 | |
| ApETBackbone=LLaVA-NeXT-7B, Retained Tokens=160, Token Retention Ratio=↓ 94.4%2026.06 | 90.1 | |
| VisionZipBackbone=LLaVA-NeXT-7B, Retained Tokens=160, Token Retention Ratio=↓ 94.4%2026.06 | 88 | |
| PDropBackbone=LLaVA-NeXT-7B, Retained Tokens=160, Token Retention Ratio=↓ 94.4%2026.06 | 86.4 |