Visual Question Answering on GQA (dev)
63.8AccuracyRADIOv2.5-H
Evaluation Results
| Method | Links | |
|---|---|---|
| RADIOv2.5-HResolution=768^2, Tokens/im=196, LLM=MN-Minitron-8B, Data mixture=LLaVA1.62024.12 | 63.8 | |
| RADIOv2.1-HResolution=512^2, Tokens/im=196, LLM=MN-Minitron-8B, Data mixture=LLaVA1.62024.12 | 63.4 | |
| RADIOv2.5-LResolution=768^2, Tokens/im=196, LLM=MN-Minitron-8B, Data mixture=LLaVA1.62024.12 | 63.4 | |
| RADIOv2.5-gResolution=672^2, Tokens/im=196, LLM=MN-Minitron-8B, Data mixture=LLaVA1.62024.12 | 63.4 | |
| SigLIP-SO400MResolution=384^2, Tokens/im=196, LLM=MN-Minitron-8B, Data mixture=LLaVA1.62024.12 | 63.1 | |
| RADIOv2.1-HResolution=768^2, Tokens/im=196, LLM=MN-Minitron-8B, Data mixture=LLaVA1.62024.12 | 62.7 | |
| OpenAI-CLIPResolution=336^2, Tokens/im=196, LLM=MN-Minitron-8B, Data mixture=LLaVA1.62024.12 | 62.4 |