Cross-modal Alignment ((I+A)→T) on Image-Text-Audio
29.97CLIP ScoreUnifiedIO2-L
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| UnifiedIO2-LCategory=Generalists2026.06 | 29.97 | — | |
| LLaVA-NeXTCategory=Specialists2026.06 | 29.47 | — | |
| MUNI†Category=Generalists2026.06 | 29.18 | — | |
| FlowBindCategory=Generalists2026.06 | 29.14 | — | |
| MMVAECategory=Multimodal VAEs2026.06 | 29.14 | — | |
| MUNICategory=Generalists2026.06 | 29.05 | — | |
| MoPoECategory=Multimodal VAEs2026.06 | 29.05 | — | |
| MoPoECategory=Multimodal VAEs2026.06 | 27.81 | 36.02 | |
| MUNI†Category=Generalists, aggregation=Hellinger2026.06 | 27.6 | 36.61 | |
| FlowBindCategory=Generalists2026.06 | 27.54 | 39.56 | |
| OmniFlowCategory=Generalists2026.06 | 27.48 | — | |
| MMVAECategory=Multimodal VAEs2026.06 | 26.57 | 35.24 | |
| CoDiCategory=Generalists2026.06 | 26.05 | — | |
| MUNICategory=Generalists, aggregation=PoE2026.06 | 25.2 | 39.75 | |
| OmniFlowCategory=Generalists2026.06 | 24.73 | 36.26 | |
| CoDiCategory=Generalists2026.06 | 24.05 | 33.72 | |
| CoDiCategory=Generalists2026.06 | — | 32.72 | |
| FlowBindCategory=Generalists2026.06 | — | 41.8 | |
| MMVAECategory=Multimodal VAEs2026.06 | — | 38.3 | |
| MoPoECategory=Multimodal VAEs2026.06 | — | 38.98 | |
| MUNICategory=Generalists2026.06 | — | 39.38 | |
| MUNI†Category=Generalists2026.06 | — | 39.45 | |
| OmniFlowCategory=Generalists2026.06 | — | 44.74 | |
| Qwen2-AudioCategory=Specialists2026.06 | — | 45.45 | |
| UnifiedIO2-LCategory=Generalists2026.06 | — | 35.48 |