Visual Reasoning on CLEVR C
52ScoreBaseline
Evaluation Results
| Method | Links | |
|---|---|---|
| BaselineRL Post-training Setup=Real RL, Data Selection Strategy=Equal2026.03 | 52 | |
| DATAPROPHETRL Post-training Setup=Real RL, Data Selection Strategy=D.P.2026.03 | 52 | |
| DATAPROPHETRL Post-training Setup=Synthetic RL, Data Selection Strategy=D.P.2026.03 | 52 | |
| BaselineRL Post-training Setup=Synthetic RL, Data Selection Strategy=Random2026.03 | 50 | |
| BaselineRL Post-training Setup=No RL, Data Selection Strategy=Baseline2026.03 | 49 |