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

Multimodal Question Answering on ScienceQA v1.3 (test)

0.9019NAT Score

Full Precision (Baseline)

0.52230.620850.71940.81795Oct 10, 2024
Updated 3d ago

Evaluation Results

MethodLinks
2024.10
0.90190.93140.87090.89390.87060.89830.9
2024.10
0.89540.93180.8650.88120.87010.88850.897
2024.10
0.89430.95730.840.88710.87510.87250.8934
2024.10
0.89390.96060.85640.88710.87650.8850.8981
2024.10
0.88870.92890.85640.87590.86560.87530.8887
2024.10
0.88450.94710.84450.87630.86070.87870.8843
2024.10
0.88030.9260.840.86020.85180.86410.8757
2024.10
0.85390.92010.83270.8480.83540.85990.8623
2024.10
0.82550.73320.83180.81030.70820.86740.8078
2024.10
0.80860.75930.80730.80010.72480.8390.7979
2024.10
0.80710.70610.78490.79460.70760.81820.7791
2024.10
0.79620.71430.82450.78250.68420.8530.7864
2024.10
0.77530.75480.79180.76640.7070.81950.7753
2024.10
0.74330.72220.74820.73410.67130.77980.7402
2024.10
0.64140.58680.62850.6280.55230.66690.6246
2024.10
0.64010.58570.6330.6280.54780.66970.6268
2024.10
0.62980.57780.61850.61870.53990.65370.616
2024.10
0.59830.57780.61480.58940.52950.64250.5983
2024.10
0.55060.51940.56270.54570.48030.58340.5472
2024.10
0.54480.49690.55550.5320.4730.57580.5324
2024.10
0.53690.49580.54270.52520.47810.57650.5298