Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Contribute
Light
Dark
VQA-X
Report a problem with this record
anchor
Findings
IC-1166
MPLUG-Owl's VQA accuracy on VQA-X increases from 68.30% to 74.48% when prompted with progressively higher-quality rationales generated by RAPPER
[eval]
IC-1408
OpenFlamingo and Idefics models generate low-quality explanations in zero-shot, but ICL and model scale significantly improve explanation CIDEr
[eval]