Question Answering on ConflictBank (EM, ROUGE-L)
55EMMACR
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MACRModel=Qwen2.5-7B2026.06 | 55 | 75 | |
| MACRModel=Llama3.1-8B2026.06 | 54.9 | 67.8 | |
| CK-PLUGModel=Qwen2.5-7B2026.06 | 32.6 | 48.5 | |
| InstructRAGICLModel=Llama3.1-8B2026.06 | 31.2 | 50.6 | |
| ICLModel=Qwen2.5-7B2026.06 | 30.2 | 45.6 | |
| TruthfulRAGModel=Llama3.1-8B2026.06 | 30.1 | 41.6 | |
| CK-PLUGModel=Llama3.1-8B2026.06 | 28.8 | 44.9 | |
| TruthfulRAGModel=Qwen2.5-7B2026.06 | 27.6 | 39.5 | |
| InstructRAGICLModel=Qwen2.5-7B2026.06 | 23.1 | 40.8 | |
| ICLModel=Llama3.1-8B2026.06 | 19.2 | 39.8 | |
| DirectModel=Llama3.1-8B2026.06 | 4.3 | 14 | |
| DirectModel=Qwen2.5-7B2026.06 | 2.2 | 9.2 |