Visual World Modelling on Something-Something
7.37GPT-4o ScoreUNIWORLD-V1
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| UNIWORLD-V12026.02 | 7.37 | 0.25 | 77 | |
| OMNIGEN2026.02 | 7.11 | 0.46 | 80 | |
| BAGEL-14B2026.02 | 6.9 | 0.16 | 82 | |
| OMNIGEN22026.02 | 6 | 0.21 | 82 | |
| SWIRL (ITERATIVE)Method Strategy (SFT/Bootstrap/SWIRL)=SWIRL (ITERATIVE)2026.02 | 3.96 | 0.21 | 74 | |
| LIQUID-SFTTest-time Verification (N)=82026.02 | 3.74 | 0.21 | 74 | |
| SWIRL (ITER. + SHARE)Method Strategy (SFT/Bootstrap/SWIRL)=SWIRL (ITER. + SHARE)2026.02 | 3.73 | 0.31 | 74 | |
| LIQUID-SFTTest-time Verification (N)=42026.02 | 3.44 | 0.25 | 76 | |
| SWIRL (IDM → FWM)Method Strategy (SFT/Bootstrap/SWIRL)=SWIRL (IDM → FWM)2026.02 | 3.44 | 0.33 | 74 | |
| LIQUID-SFTTest-time Verification (N)=22026.02 | 3.4 | 0.21 | 76 | |
| CHAMELEON-SFT2026.02 | 3.11 | — | — | |
| CHAMELEON-BOOTSTRAP2026.02 | 3.11 | — | — | |
| LIQUID-SFTMethod Strategy (SFT/Bootstrap/SWIRL)=SFT2026.02 | 3 | 0.27 | 74 | |
| BLIP3O-NEXT2026.02 | 2.86 | 0.35 | 66 | |
| LIQUID-BOOTSTRAPMethod Strategy (SFT/Bootstrap/SWIRL)=Bootstrap2026.02 | 2.86 | 0.29 | 75 | |
| SMARTEDIT2026.02 | 2.81 | — | — | |
| GOT2026.02 | 2.62 | — | — | |
| InstructPix2Pix2026.02 | 0.96 | — | — |