Knowledge Acquisition on Wikipedia OOD
100AccuracyOracle RAG
Evaluation Results
| Method | Links | |
|---|---|---|
| Oracle RAGBackbone=Qwen2.5-7B-Instruct2026.01 | 100 | |
| SDFTBackbone=Qwen2.5-7B-Instruct2026.01 | 98 | |
| SFTBackbone=Qwen2.5-7B-Instruct2026.01 | 80 | |
| CPTBackbone=Qwen2.5-7B-Instruct2026.01 | 7 | |
| BaseBackbone=Qwen2.5-7B-Instruct2026.01 | 0 |