Video Question Answering on CLEVRER (test)
96.46Descriptive AccuracyCRCG_VRDP + IOD + SPS
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| CRCG_VRDP + IOD + SPSBase Model=VRDP, Modules=IOD + SPS2025.06 | 96.46 | 99.32 | 98.8 | 96.11 | 92.28 | 96.61 | 90.72 | — | — | — | |
| ODDN-Aloe2025.06 | 95.8 | 98.9 | 97 | 95.7 | 91.8 | 93 | 80.1 | — | — | — | |
| CRCG_NS-DR^approx + IOD + SPSBase Model=NS-DR, Modules=IOD + SPS2025.06 | 95.55 | 99.94 | 99.81 | 88.12 | 76.64 | 90.73 | 78.31 | — | — | — | |
| Aloe2025.06 | 94 | 98.47 | 96 | 93.5 | 87.5 | 91.42 | 75.61 | — | — | — | |
| VRDP2025.06 | 93.4 | 96.3 | 91.9 | 95.7 | 91.4 | 94.8 | 84.3 | — | — | — | |
| DCL2025.06 | 90.7 | 89.6 | 82.8 | 90.5 | 82 | 80.4 | 46.5 | — | — | — | |
| NS-DR2025.06 | 88.1 | 87.6 | 79.6 | 82.9 | 68.7 | 74.1 | 42.2 | — | — | — | |
| Averaged FrameInput Data=Image (T = 1)2026.05 | — | — | — | — | — | — | — | — | 62.2 | 1.6 | |
| ChronoSCInput Data=Stacked Img2026.05 | — | — | — | — | — | — | — | — | 76.2 | 1.6 | |
| Full Video JSCCInput Data=Video (T = 8)2026.05 | — | — | — | — | — | — | — | — | 80.1 | 1.3 | |
| H.264 + LDPCInput Data=Video (T = 8)2026.05 | — | — | — | — | — | — | — | — | 35.2 | 1.6 | |
| Single FrameInput Data=Image (T = 1)2026.05 | — | — | — | — | — | — | — | — | 43.4 | 1.6 | |
| SlotContrast + Aloenumber of slots=72025.08 | — | — | — | — | — | — | — | 97.2 | 95.6 | — | |
| SmoothSAv + Aloenumber of slots=72025.08 | — | — | — | — | — | — | — | 98.7 | 96.9 | — |