Evidence Retrieval on FAKEWIKI Clean
78.1Recall@10SCORINGMODEL
Evaluation Results
| Method | Links | |
|---|---|---|
| SCORINGMODELTarget Model=Qwen3-8B2026.05 | 78.1 | |
| SCORINGMODELTarget Model=Llama-3.1-8B2026.05 | 76.7 | |
| STEERFUSETarget Model=Qwen3-8B2026.05 | 76.4 | |
| STEERFUSETarget Model=Llama-3.1-8B2026.05 | 57.8 | |
| BGE-baseTarget Model=Qwen3-8B, Query Type=QA2026.05 | 43.1 | |
| SBERT-MiniLMTarget Model=Qwen3-8B, Query Type=QA2026.05 | 39.7 | |
| SBERT-MPNetTarget Model=Qwen3-8B, Query Type=QA2026.05 | 38.1 | |
| BGE-baseTarget Model=Llama-3.1-8B, Query Type=QA2026.05 | 33.3 | |
| ContrieverTarget Model=Qwen3-8B, Query Type=answer2026.05 | 32.4 | |
| BGE-baseTarget Model=Qwen3-8B, Query Type=answer2026.05 | 30.8 | |
| ContrieverTarget Model=Qwen3-8B, Query Type=QA2026.05 | 30.1 | |
| SBERT-MPNetTarget Model=Llama-3.1-8B, Query Type=QA2026.05 | 29.1 | |
| SBERT-MiniLMTarget Model=Llama-3.1-8B, Query Type=QA2026.05 | 28.9 | |
| SBERT-MiniLMTarget Model=Qwen3-8B, Query Type=answer2026.05 | 28.4 | |
| SBERT-MPNetTarget Model=Qwen3-8B, Query Type=answer2026.05 | 25.5 | |
| Finetuned EmbedSimTarget Model=Qwen3-8B2026.05 | 17.4 | |
| Finetuned EmbedSimTarget Model=Llama-3.1-8B2026.05 | 16.9 | |
| ContrieverTarget Model=Llama-3.1-8B, Query Type=QA2026.05 | 5.1 | |
| BGE-baseTarget Model=Llama-3.1-8B, Query Type=answer2026.05 | 4.3 | |
| SBERT-MiniLMTarget Model=Llama-3.1-8B, Query Type=answer2026.05 | 3.7 | |
| SBERT-MPNetTarget Model=Llama-3.1-8B, Query Type=answer2026.05 | 3.6 | |
| ContrieverTarget Model=Llama-3.1-8B, Query Type=answer2026.05 | 3.6 | |
| MinHashTarget Model=Qwen3-8B, Query Type=answer2026.05 | 0.1 | |
| MinHashTarget Model=Qwen3-8B, Query Type=QA2026.05 | 0.1 | |
| MinHashTarget Model=Llama-3.1-8B, Query Type=answer2026.05 | 0 | |
| MinHashTarget Model=Llama-3.1-8B, Query Type=QA2026.05 | 0 |