List your company

Multi-modal Visual Question Answering on VQA-T

64.84Accuracy16-bit Baseline
49.458453.451757.44561.4383Nov 15, 2024

Evaluation results

MethodLinks
16-bit BaselineData Format=16-bit, Backbone=LLaVA1.6-7B2024.1164.84
AMXFP4Data Format=AMXFP4, Backbone=LLaVA1.6-7B2024.1159.13
MXFP4Data Format=MXFP4, Backbone=LLaVA1.6-7B2024.1157.88
MXFP4-PoTData Format=MXFP4-PoT, Backbone=LLaVA1.6-7B2024.1150.05