Long-form Factuality on FActScore
59.3FActScoreKLCF-zero
Evaluation Results
| Method | Links | |
|---|---|---|
| KLCF-zeroBackbone=Qwen2.5-32B, Setting=zero-shot2025.09 | 59.3 | |
| KARLAModel=Qwen 4B, Setup=KARLA2026.06 | 58.9 | |
| 1-hop graph RAGModel=Qwen 8B, Setup=1-hop graph RAG2026.06 | 58.2 | |
| KARLAModel=Qwen 8B, Setup=KARLA2026.06 | 57.3 | |
| 1-hop graph RAGModel=Qwen 4B, Setup=1-hop graph RAG2026.06 | 56.8 | |
| 1-hop graph RAGModel=Qwen 1.7B, Setup=1-hop graph RAG2026.06 | 55.5 | |
| KLCF-zeroBackbone=Qwen2.5-7B, Setting=zero-shot2025.09 | 55 | |
| KARLAModel=Qwen 1.7B, Setup=KARLA2026.06 | 53.7 | |
| 1-hop graph RAGModel=Qwen 0.6B, Setup=1-hop graph RAG2026.06 | 53.1 | |
| KARLAModel=Qwen 0.6B, Setup=KARLA2026.06 | 53 | |
| CoVeBackbone=Qwen2.5-32B2025.09 | 45.9 | |
| CoVeBackbone=Qwen2.5-7B2025.09 | 40.1 | |
| BaseBackbone=Qwen2.5-32B, Setting=10-shot prompting2025.09 | 40 | |
| KARLA-no-descModel=Qwen 4B, Setup=KARLA-no-desc2026.06 | 39.07 | |
| KARLA-no-descModel=Qwen 8B, Setup=KARLA-no-desc2026.06 | 38.72 | |
| BaseBackbone=Qwen2.5-7B, Setting=10-shot prompting2025.09 | 36.6 | |
| KARLA-empty-KBModel=Qwen 8B, Setup=KARLA-empty-KB2026.06 | 35.9 | |
| KARLA-empty-KBModel=Qwen 4B, Setup=KARLA-empty-KB2026.06 | 33.8 | |
| KARLA-no-descModel=Qwen 1.7B, Setup=KARLA-no-desc2026.06 | 33.75 | |
| Tool-schema promptModel=Qwen 8B, Setup=Tool-schema prompt2026.06 | 32.2 | |
| LMLMModel=GPT2-774M, Setup=LMLM2026.06 | 31.9 | |
| Tool-schema promptModel=Qwen 4B, Setup=Tool-schema prompt2026.06 | 30 | |
| KARLA-empty-KBModel=Qwen 1.7B, Setup=KARLA-empty-KB2026.06 | 29.1 | |
| Tool-schema promptModel=Qwen 1.7B, Setup=Tool-schema prompt2026.06 | 29 | |
| Base LMModel=Qwen 8B, Setup=Base LM2026.06 | 26.4 | |
| KARLA-no-descModel=Qwen 0.6B, Setup=KARLA-no-desc2026.06 | 24.54 | |
| Tool-schema promptModel=Qwen 0.6B, Setup=Tool-schema prompt2026.06 | 24.4 | |
| Base LMModel=Qwen 4B, Setup=Base LM2026.06 | 24.16 | |
| LMLMModel=LLAMA2-382M, Setup=LMLM2026.06 | 23.9 | |
| Base LMModel=Qwen 1.7B, Setup=Base LM2026.06 | 23.71 | |
| KARLA-empty-KBModel=Qwen 0.6B, Setup=KARLA-empty-KB2026.06 | 23.2 | |
| Base LMModel=Qwen 0.6B, Setup=Base LM2026.06 | 22.78 |