Visual Question Answering on ChartQA (Score Metric)
90.9ScoreSPOT-E
Evaluation Results
| Method | Links | |
|---|---|---|
| SPOT-EBase Model=Qwen3-VL-32B2026.06 | 90.9 | |
| SPOT-EBase Model=Qwen2.5-VL-32B2026.06 | 90.6 | |
| Qwen3-VL-32BBase Model=Qwen3-VL-32B2026.06 | 90.2 | |
| Qwen2.5-VL-32BBase Model=Qwen2.5-VL-32B2026.06 | 89.8 | |
| SPOT-EBase Model=InternVL2.5-26B2026.06 | 88.8 | |
| InternVL2.5-26BBase Model=InternVL2.5-26B2026.06 | 87.5 | |
| SPOT-EBase Model=LLaVA-OV-72B2026.06 | 87.1 | |
| LLaVA-OV-72BBase Model=LLaVA-OV-72B2026.06 | 85.6 | |
| Pelican-UnifiedModel Category=Our Unified Model2026.05 | 81.5 | |
| Qwen3-VL-4B-InstructModel Category=Vision-Language Models2026.05 | 81.1 | |
| MolmoActModel Category=Vision-Language-Action Models2026.05 | 55.9 | |
| Gemma3-4B-ITModel Category=Vision-Language Models2026.05 | 50.3 | |
| Decoupled DiLoCoModel Scale=Gemma 9B, Method=DiLoCo2026.04 | 23 | |
| DPModel Scale=Gemma 9B, Method=DP2026.04 | 21.6 | |
| DPModel Scale=Gemma 5B, Method=DP2026.04 | 19 | |
| EL2NTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 17.9 | |
| Full-FinetuneTraining Budget=100%, Backbone=LLaVA-1.5-13B2026.05 | 17.5 | |
| Decoupled DiLoCoModel Scale=Gemma 5B, Method=DiLoCo2026.04 | 17.3 | |
| D2-PruningTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 17.3 | |
| ICONSTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 17.1 | |
| MAGICTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 17.1 | |
| COINCIDETraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 17 | |
| RDSTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 16.8 | |
| Self-SupTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 16.6 | |
| CLIP-ScoreTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 16.4 | |
| PerplexityTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 16.4 | |
| SemDeDupTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 16.4 | |
| Self-FilterTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 16 | |
| RandomTraining Budget=20%, Backbone=LLaVA-1.5-13B2026.05 | 15.1 | |
| Decoupled DiLoCoModel Scale=Gemma 2B, Method=DiLoCo2026.04 | 13.3 | |
| DPModel Scale=Gemma 2B, Method=DP2026.04 | 12.2 | |
| π0.5Model Category=Vision-Language-Action Models2026.05 | 5.1 |