Multimodal Instruction Following on MIA-Bench
8.86ScoreGPT-4o
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GPT-4o2024.10 | 8.86 | — | |
| ARIA2024.10 | 8.76 | — | |
| PixtralParameters=12B2024.10 | 8.43 | — | |
| Qwen2-VLParameters=7B2024.10 | 8.07 | — | |
| Claude-3-Sonnet2024.10 | 7.94 | — | |
| Reka-Core2024.10 | 7.7 | — | |
| MiniCPM-VVersion=2.52024.10 | 7.63 | — | |
| Phi-3 Vision2024.10 | 7.6 | — | |
| LLaVA-NeXTParameters=34B2024.10 | 7.56 | — | |
| InternVL-ChatVersion=v1.52024.10 | 7.54 | — | |
| Gemini-1.0-Pro2024.10 | 7.06 | — | |
| Idefics22024.10 | 5.14 | — | |
| CIA2025.10 | — | 34.07 | |
| EWC2025.10 | — | 29.79 | |
| Full-FTLearning Strategy=Fine-Tuning2025.10 | — | 25.25 | |
| KORErank=2352025.10 | — | 38.54 | |
| KORErank=2562025.10 | — | 60.02 | |
| LLaVA-v1.5Backbone Parameters=7B2025.10 | — | 66.33 | |
| LoRA2025.10 | — | 29.66 | |
| LwF2025.10 | — | 36.19 | |
| MoELoRA2025.10 | — | 35.03 | |
| O-LoRA2025.10 | — | 34.85 | |
| Replay2025.10 | — | 62.33 | |
| SEFE2025.10 | — | 42.9 |