Knowledge-based Visual Question Answering on INFOSEEK (Overall)
33.2AccuracyMMhops-R1
Evaluation results
| Method | Links | |
|---|---|---|
| MMhops-R1Model Size=7B2025.12 | 33.2 | |
| EchoSight2025.12 | 31.3 | |
| Wiki-LLaVA2025.12 | 28.9 | |
| CLIP-FiD2025.12 | 20.9 | |
| CLIP-PaLM2025.12 | 20.4 |
| Method | Links | |
|---|---|---|
| MMhops-R1Model Size=7B2025.12 | 33.2 | |
| EchoSight2025.12 | 31.3 | |
| Wiki-LLaVA2025.12 | 28.9 | |
| CLIP-FiD2025.12 | 20.9 | |
| CLIP-PaLM2025.12 | 20.4 |