Yes/No Question Answering on BioASQ Task B
93.51AccuracyParallaxRAG + Qwen3-Plus
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| ParallaxRAG + Qwen3-PlusGenerator=Qwen3-Plus, Retriever=ParallaxRAG2025.10 | 93.51 | 95.24 | 89.8 | 92.52 | |
| GPT-4oGenerator=GPT-4o, Retriever=None2025.10 | 91.4 | 93.47 | 86.7 | 90.09 | |
| GPT4-TurboGenerator=GPT4-Turbo, Retriever=None2025.10 | 91.29 | 93.57 | 86.16 | 89.86 | |
| Single-View + Qwen3-PlusGenerator=Qwen3-Plus, Retriever=Single-View2025.10 | 91.12 | 92.67 | 83.74 | 88.21 | |
| Deepseek-R1-8BGenerator=Deepseek-R1-8B, Retriever=None2025.10 | 90.77 | 93.12 | 85.58 | 89.35 | |
| Qwen3-PlusGenerator=Qwen3-Plus, Retriever=None2025.10 | 90.12 | 92.55 | 84.02 | 88.29 |