Yes/No Question Answering on Quilt-VQA
75.8AccuracyPatho-AgenticRAG
Evaluation Results
| Method | Links | |
|---|---|---|
| Patho-AgenticRAG2025.08 | 75.8 | |
| Llama3.2-3B-vlModel scale=Small, Thinking ability=Without thinking ability2025.05 | 69.39 | |
| LLaVA-Med-v1.5-Mistral-7BModel scale=Large, Thinking ability=Without thinking ability2025.05 | 67.35 | |
| Patho-R1-7BModel scale=Large, Thinking ability=With thinking ability2025.05 | 64.72 | |
| Patho-R1-7B2025.08 | 64.72 | |
| Patho-R1-3BModel scale=Small, Thinking ability=With thinking ability2025.05 | 64.14 | |
| InternVL2-8BModel scale=Large, Thinking ability=With thinking ability2025.05 | 63.56 | |
| Llama-3.2-11B-Vision-InstructModel scale=Large, Thinking ability=With thinking ability2025.05 | 63.27 | |
| Llama-3.2-11B-VI2025.08 | 63.27 | |
| InternVL2-8B2025.08 | 60.56 | |
| InternVL2.5-8BModel scale=Large, Thinking ability=With thinking ability2025.05 | 60.06 | |
| InternVL2.5-8B2025.08 | 60.06 | |
| HuatuoGPT-Vision-7BModel scale=Large, Thinking ability=Without thinking ability2025.05 | 59.18 | |
| Llama-3.2V-11B-cotModel scale=Large, Thinking ability=With thinking ability2025.05 | 54.81 | |
| Llama-3.2V-11B-cotReasoning=Chain-of-Thought2025.08 | 54.81 | |
| Qwen2.5VL-7BModel scale=Large, Thinking ability=With thinking ability2025.05 | 52.19 | |
| Qwen2.5VL-7B2025.08 | 52.19 | |
| PathGen-LLaVA-13BModel scale=Large, Thinking ability=Without thinking ability2025.05 | 47.52 | |
| VILA1.5-3BModel scale=Small, Thinking ability=Without thinking ability2025.05 | 46.65 | |
| DeepSeek-VL2-TinyModel scale=Small, Thinking ability=With thinking ability2025.05 | 44.31 | |
| Qwen2.5VL-3BModel scale=Small, Thinking ability=With thinking ability2025.05 | 40.52 | |
| InternVL3-8BModel scale=Large, Thinking ability=With thinking ability2025.05 | 33.82 | |
| InternVL3-8B2025.08 | 33.82 | |
| Paligemma2-3b-pt-224Model scale=Small, Thinking ability=Without thinking ability2025.05 | 32.94 | |
| LLaVA-Onevision-7BModel scale=Large, Thinking ability=With thinking ability2025.05 | 24.2 | |
| LLaVA-Onevision-7B2025.08 | 24.2 | |
| Quilt-LLaVA-v1.5-7BModel scale=Large, Thinking ability=Without thinking ability2025.05 | 21.28 |