Visual Question Answering on VQA v2 (Accuracy Only)
78.7AccuracyGroundingGPT
Evaluation Results
| Method | Links | |
|---|---|---|
| GroundingGPTLLM Size=7B2024.01 | 78.7 | |
| LLaVA-1.5LLM Size=7B2024.01 | 78.5 | |
| ShikraLLM Size=13B2024.01 | 77.4 | |
| BLIP-2LLM Size=13B2024.01 | 41 |
| Method | Links | |
|---|---|---|
| GroundingGPTLLM Size=7B2024.01 | 78.7 | |
| LLaVA-1.5LLM Size=7B2024.01 | 78.5 | |
| ShikraLLM Size=13B2024.01 | 77.4 | |
| BLIP-2LLM Size=13B2024.01 | 41 |