Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Vision-Language Understanding and Reasoning on LLaVA Multimodal Evaluation Suite 1.5 (test/val)

0.619GQA

Vanilla

0.469240.508120.5470.58588Jan 30, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.01
0.6190.6471,8620.8590.6950.7850.5820.5860.3110.5011
2026.01
0.610.6441,7980.8680.7080.7840.5840.5830.3320.5131.006
2026.01
0.6060.6391,8060.8620.6860.7780.577--0.5040.989
2026.01
0.6030.641,7880.8660.6970.7820.5820.5750.3270.5140.999
2026.01
0.5980.631,7920.8610.6890.7710.573--0.5170.986
2026.01
0.5930.631,7830.8530.6890.7680.5730.5640.3170.5050.983
2026.01
0.5880.631,7800.8620.710.7680.5680.5480.3220.5230.988
2026.01
0.5830.6211,6980.850.6910.7540.556--0.5180.968
2026.01
0.5760.6251,7210.8360.6910.7560.5610.5580.3020.50.964
2026.01
0.5760.621,7620.8320.6890.7560.5680.5490.3260.50.972
2026.01
0.5730.6331,7970.8480.6920.7640.5650.5720.3080.50.976
2026.01
0.5710.6161,7610.8260.6840.760.5660.5620.310.4960.965
2026.01
0.560.61,6960.8050.6710.7380.5490.5340.30.510.942
2026.01
0.5510.6011,6900.770.690.7240.5550.5220.3170.5190.944
2026.01
0.5270.5621,5050.7510.6220.6820.5180.5110.2330.4960.867
2026.01
0.4750.5881,5610.7620.690.7330.5060.530.3050.5020.908