Reasoning and Synthesis on Gaucher Disease (GD)
99.71FaithfulnessSARG
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SARGBackbone=Qwen2.5-7B-Instruct2025.06 | 99.71 | 73.58 | 91.6 | |
| SARGBackbone=GPT-4o-mini2025.06 | 99.7 | 83.5 | 96.1 | |
| RAGBackbone=Qwen2.5-7B-Instruct2025.06 | 98.7 | 74.03 | 85.78 | |
| RAGBackbone=DeepSeek-R1-Distill-LLaMA-8B2025.06 | 98.7 | 68.87 | 87.98 | |
| HippoRAG2Backbone=GPT-4o-mini2025.06 | 98.49 | 72.52 | 93.55 | |
| StructRAGBackbone=GPT-4o-mini2025.06 | 98.22 | 70.09 | 91 | |
| DirectBackbone=Qwen2.5-7B-Instruct2025.06 | 98 | 65.85 | 85.02 | |
| SARGBackbone=LLaMA-3.1-8B-Instruct2025.06 | 97.61 | 64.32 | 88.24 | |
| HyperGraphRAGBackbone=GPT-4o-mini2025.06 | 97.51 | 70 | 93.34 | |
| GraphRAGBackbone=GPT-4o-mini2025.06 | 97.46 | 74.31 | 92.78 | |
| RAGBackbone=LLaMA-3.1-8B-Instruct2025.06 | 97.22 | 67.26 | 66.23 | |
| DirectBackbone=LLaMA-3.1-8B-Instruct2025.06 | 97.21 | 56.11 | 70.73 | |
| DirectBackbone=DeepSeek-R1-Distill-LLaMA-8B2025.06 | 96.53 | 69.16 | 86.55 | |
| CoT-RAGBackbone=Qwen2.5-7B-Instruct2025.06 | 96.03 | 66.26 | 74.01 | |
| CoT-RAGBackbone=DeepSeek-R1-Distill-LLaMA-8B2025.06 | 95.83 | 67.28 | 85.91 | |
| SARGBackbone=DeepSeek-R1-Distill-LLaMA-8B2025.06 | 95.24 | 58.36 | 85.03 | |
| CoT-RAGBackbone=LLaMA-3.1-8B-Instruct2025.06 | 94.17 | 62.24 | 61.71 |