Document Visual Question Answering on DocVQA (Score)
90.4ScoreSPOT-E
Evaluation Results
| Method | Links | |
|---|---|---|
| SPOT-EBase Model=Qwen3-VL-32B2026.06 | 90.4 | |
| SPOT-EBase Model=Qwen2.5-VL-32B2026.06 | 89.8 | |
| Qwen3-VL-32BBase Model=Qwen3-VL-32B2026.06 | 89.8 | |
| Qwen2.5-VL-32BBase Model=Qwen2.5-VL-32B2026.06 | 89.1 | |
| SPOT-EBase Model=InternVL2.5-26B2026.06 | 86 | |
| SPOT-EBase Model=LLaVA-OV-72B2026.06 | 86 | |
| InternVL2.5-26BBase Model=InternVL2.5-26B2026.06 | 85 | |
| LLaVA-OV-72BBase Model=LLaVA-OV-72B2026.06 | 84.9 | |
| Full-FinetuneTraining Budget=100%, Backbone=LLaVA-1.5-13B2026.05 | 28.9 | |
| COINCIDETraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 27.9 | |
| ICONSTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 27.9 | |
| RDSTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 27.5 | |
| Self-SupTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 27.4 | |
| MAGICTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 27.2 | |
| CLIP-ScoreTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 27.1 | |
| PerplexityTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 27.1 | |
| Self-FilterTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 27.1 | |
| EL2NTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 26.9 | |
| D2-PruningTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 26.5 | |
| SemDeDupTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 26.4 | |
| RandomTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 25.2 |