Question Answering on FactKG
85.8EM@1CLAUSE
Evaluation Results
| Method | Links | |
|---|---|---|
| CLAUSEFamily=Ours, Reader Model=GPT-OSS-120B2025.09 | 85.8 | |
| CLAUSEFamily=Ours, LLM Reader=LLaMA3.3-70B2025.09 | 85 | |
| CLAUSEFamily=Ours2025.09 | 84.2 | |
| KG-AgentFamily=Agent-based, Reader Model=GPT-OSS-120B2025.09 | 84.2 | |
| KG-AgentFamily=Agent-based, LLM Reader=LLaMA3.3-70B2025.09 | 83.3 | |
| Hybrid RAGFamily=RAG-based, Reader Model=GPT-OSS-120B2025.09 | 82.9 | |
| KG-AgentFamily=Agent-based (Qwen3-32B)2025.09 | 82.1 | |
| Hybrid RAGFamily=RAG-based, LLM Reader=LLaMA3.3-70B2025.09 | 82 | |
| ReActFamily=Agent-based, Reader Model=GPT-OSS-120B2025.09 | 80.6 | |
| Hybrid RAGFamily=RAG-based (Qwen3-32B)2025.09 | 80.2 | |
| ReActFamily=Agent-based, LLM Reader=LLaMA3.3-70B2025.09 | 79.5 | |
| Vanilla RAGFamily=RAG-based, Reader Model=GPT-OSS-120B2025.09 | 79.2 | |
| Vanilla RAGFamily=RAG-based, LLM Reader=LLaMA3.3-70B2025.09 | 78.3 | |
| AutoGenFamily=Agent-based, Reader Model=GPT-OSS-120B2025.09 | 78.3 | |
| ReActFamily=Agent-based (Qwen3-32B)2025.09 | 78.2 | |
| AutoGenFamily=Agent-based, LLM Reader=LLaMA3.3-70B2025.09 | 77.3 | |
| Vanilla RAGFamily=RAG-based (Qwen3-32B)2025.09 | 77 | |
| AutoGenFamily=Agent-based (Qwen3-32B)2025.09 | 76.5 | |
| Graph-of-ThoughtsFamily=Agent-based, Reader Model=GPT-OSS-120B2025.09 | 76 | |
| Graph-of-ThoughtsFamily=Agent-based, LLM Reader=LLaMA3.3-70B2025.09 | 75.2 | |
| GraphRAGFamily=RAG-based, Reader Model=GPT-OSS-120B2025.09 | 74.3 | |
| Graph-of-ThoughtsFamily=Agent-based (Qwen3-32B)2025.09 | 74 | |
| GraphRAGFamily=RAG-based, LLM Reader=LLaMA3.3-70B2025.09 | 73.4 | |
| GraphRAGFamily=RAG-based (Qwen3-32B)2025.09 | 72 | |
| LightRAGFamily=RAG-based, Reader Model=GPT-OSS-120B2025.09 | 69.2 | |
| LightRAGFamily=RAG-based, LLM Reader=LLaMA3.3-70B2025.09 | 68.2 | |
| GPT-OSS-120BFamily=Pretrained-LLMs2025.09 | 68 | |
| LLaMA3.3-70BFamily=Pretrained-LLMs2025.09 | 66.7 | |
| LightRAGFamily=RAG-based (Qwen3-32B)2025.09 | 64.5 | |
| Qwen3-32BFamily=Pretrained-LLMs2025.09 | 60.1 |