Image Captioning on Vision-Language Tasks Captioning
68.9Targeted ASRCroPA
Evaluation Results
| Method | Links | |
|---|---|---|
| CroPATarget Prompt=I am sorry, VLM Backbone=Flamingo2025.06 | 68.9 | |
| Multi-PTarget Prompt=I am sorry, VLM Backbone=Flamingo2025.06 | 61.4 | |
| CroPATarget Prompt=metaphor, VLM Backbone=Flamingo2025.06 | 58.4 | |
| Multi-PTarget Prompt=metaphor, VLM Backbone=Flamingo2025.06 | 49.1 | |
| CroPATarget Prompt=unknown, VLM Backbone=Flamingo2025.06 | 42 | |
| Multi-PTarget Prompt=unknown, VLM Backbone=Flamingo2025.06 | 28.5 | |
| CroPATarget Prompt=very good, VLM Backbone=Flamingo2025.06 | 25.4 | |
| Single-PTarget Prompt=I am sorry, VLM Backbone=Flamingo2025.06 | 25.3 | |
| Multi-PTarget Prompt=very good, VLM Backbone=Flamingo2025.06 | 20.8 | |
| CroPATarget Prompt=too late, VLM Backbone=Flamingo2025.06 | 17.9 | |
| Multi-PTarget Prompt=too late, VLM Backbone=Flamingo2025.06 | 16.3 | |
| Single-PTarget Prompt=unknown, VLM Backbone=Flamingo2025.06 | 14 | |
| Single-PTarget Prompt=metaphor, VLM Backbone=Flamingo2025.06 | 10.4 | |
| Single-PTarget Prompt=very good, VLM Backbone=Flamingo2025.06 | 4.8 | |
| Single-PTarget Prompt=too late, VLM Backbone=Flamingo2025.06 | 2.2 | |
| Multi-PTarget Prompt=not sure, VLM Backbone=Flamingo2025.06 | 0.3 | |
| CroPATarget Prompt=not sure, VLM Backbone=Flamingo2025.06 | 0.1 | |
| Single-PTarget Prompt=not sure, VLM Backbone=Flamingo2025.06 | 0 |