Visual World Modelling on WhatsUp
8.2GPT-4o ScoreUNIWORLD-V1
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| UNIWORLD-V12026.02 | 8.2 | 0.46 | 91 | |
| OMNIGEN2026.02 | 7.52 | 0.24 | 93 | |
| OMNIGEN22026.02 | 6.6 | 0.28 | 92 | |
| BAGEL-14B2026.02 | 5.4 | 0.22 | 95 | |
| LIQUID-SFTTest-time Verification (N)=82026.02 | 4.48 | 0.28 | 86 | |
| SWIRL (ITER. + SHARE)Method Strategy (SFT/Bootstrap/SWIRL)=SWIRL (ITER. + SHARE)2026.02 | 4.46 | 0.3 | 85 | |
| LIQUID-SFTTest-time Verification (N)=22026.02 | 4.38 | 0.36 | 86 | |
| LIQUID-SFTTest-time Verification (N)=42026.02 | 4.38 | 0.4 | 87 | |
| SWIRL (ITERATIVE)Method Strategy (SFT/Bootstrap/SWIRL)=SWIRL (ITERATIVE)2026.02 | 4.32 | 0.36 | 86 | |
| SWIRL (IDM → FWM)Method Strategy (SFT/Bootstrap/SWIRL)=SWIRL (IDM → FWM)2026.02 | 4.08 | 0.32 | 86 | |
| LIQUID-BOOTSTRAPMethod Strategy (SFT/Bootstrap/SWIRL)=Bootstrap2026.02 | 3.88 | 0.28 | 87 | |
| LIQUID-SFTMethod Strategy (SFT/Bootstrap/SWIRL)=SFT2026.02 | 3.6 | 0.32 | 85 | |
| BLIP3O-NEXT2026.02 | 3 | 0.4 | 78 | |
| GOT2026.02 | 1.58 | — | — | |
| CHAMELEON-BOOTSTRAP2026.02 | 0.98 | — | — | |
| CHAMELEON-SFT2026.02 | 0.88 | — | — | |
| SMARTEDIT2026.02 | 0.76 | — | — | |
| InstructPix2Pix2026.02 | 0 | — | — |