List your company

Overall Vision-Language Performance on Vision-Language Tasks Aggregate

86.8Targeted ASRCroPA
20.531237.735654.9472.1444Jun 28, 2025

Evaluation results

MethodLinks
CroPATarget Prompt=metaphor, VLM Backbone=Flamingo2025.0686.8
CroPATarget Prompt=I am sorry, VLM Backbone=Flamingo2025.0679.85
Multi-PTarget Prompt=metaphor, VLM Backbone=Flamingo2025.0679.33
CroPATarget Prompt=unknown, VLM Backbone=Flamingo2025.0677.08
Multi-PTarget Prompt=I am sorry, VLM Backbone=Flamingo2025.0673.18
CroPATarget Prompt=very good, VLM Backbone=Flamingo2025.0670.2
CroPATarget Prompt=too late, VLM Backbone=Flamingo2025.0669.2
Multi-PTarget Prompt=very good, VLM Backbone=Flamingo2025.0661.75
Multi-PTarget Prompt=unknown, VLM Backbone=Flamingo2025.0660.95
Multi-PTarget Prompt=too late, VLM Backbone=Flamingo2025.0659.18
CroPATarget Prompt=not sure, VLM Backbone=Flamingo2025.0650.33
Single-PTarget Prompt=metaphor, VLM Backbone=Flamingo2025.0639.73
Multi-PTarget Prompt=not sure, VLM Backbone=Flamingo2025.0639.3
Single-PTarget Prompt=I am sorry, VLM Backbone=Flamingo2025.0636.4
Single-PTarget Prompt=very good, VLM Backbone=Flamingo2025.0631
Single-PTarget Prompt=unknown, VLM Backbone=Flamingo2025.0628.65
Single-PTarget Prompt=too late, VLM Backbone=Flamingo2025.0624.55
Single-PTarget Prompt=not sure, VLM Backbone=Flamingo2025.0623.08