Artifact Explanation on LOKI (test)
0.169ROUGEQwen2.5-VL-7B + ArtiAgent
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen2.5-VL-7B + ArtiAgentFine-tuning=100K training set generated by ArtiAgent2026.02 | 0.169 | 0.454 | |
| InternVL3.5-8B + ArtiAgentFine-tuning=100K training set generated by ArtiAgent2026.02 | 0.137 | 0.401 | |
| LEGIONTraining Split=SynthScars training dataset2026.02 | 0.133 | 0.314 | |
| GPT-52026.02 | 0.121 | 0.382 | |
| GPT-4o2026.02 | 0.107 | 0.266 | |
| Qwen2.5-VL-7BFine-tuning=Vanilla2026.02 | 0.106 | 0.267 | |
| Gemini-2.5-Pro2026.02 | 0.097 | 0.358 | |
| InternVL3.5-8BFine-tuning=Vanilla2026.02 | 0.081 | 0.189 |