KeyComp* (GPT-3.5)

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· 2024-01-20
17.4
best: 58.75 (GPT-4V (CoT, pick b/w two options))
Prompting Large Vision-Language Models for Compositional Reasoning arXiv:2401.11337
Visual ReasoningonWinoground
Image Score· 2024-01-20
27.8
best: 68.75 (GPT-4V (CoT, pick b/w two options))
Prompting Large Vision-Language Models for Compositional Reasoning arXiv:2401.11337
Visual ReasoningonWinoground
Text Score· 2024-01-20
42.7
best: 75.5 (GPT-4o + CA)
Prompting Large Vision-Language Models for Compositional Reasoning arXiv:2401.11337