OFA large (TLC-A)

Reported on 3 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Reasoning3 results

Visual ReasoningonWinoground
Group Score· 2023-05-11
17.5
best: 58.75 (GPT-4V (CoT, pick b/w two options))
Simple Token-Level Confidence Improves Caption Correctness arXiv:2305.07021
Visual ReasoningonWinoground
Image Score· 2023-05-11
27
best: 68.75 (GPT-4V (CoT, pick b/w two options))
Simple Token-Level Confidence Improves Caption Correctness arXiv:2305.07021
Visual ReasoningonWinoground
Text Score· 2023-05-11
29.25
best: 75.5 (GPT-4o + CA)
Simple Token-Level Confidence Improves Caption Correctness arXiv:2305.07021