Aggregation Query Execution on AGGBench-Core
0.463NACE (mean)Naive RAG
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Naive RAGCategory=Single-Step RAG, Base LLM=GPT-4o-mini2026.02 | 0.463 | 0.5 | 3.522 | 4.2 | |
| HippoRAGCategory=Single-Step RAG, Base LLM=GPT-4o-mini2026.02 | 0.503 | 0.5 | 3.511 | 4.1 | |
| DFACategory=Ours, Base LLM=GPT-4o-mini2026.02 | 0.544 | 0.464 | 2.67 | 62.3 | |
| DeepNoteCategory=Multi-Hop RAG, Base LLM=GPT-4o-mini2026.02 | 0.621 | 0.667 | 4.552 | 1.5 | |
| ReActCategory=Agentic Framework, Base LLM=GPT-4o-mini2026.02 | 0.77 | 0.5 | 3.773 | 5.4 | |
| Multi-AgentCategory=Agentic Framework, Base LLM=GPT-4o-mini2026.02 | 0.77 | 0.571 | 6.412 | 2.6 | |
| RATCategory=Multi-Hop RAG, Base LLM=GPT-4o-mini2026.02 | 0.787 | 0.5 | 4.076 | 3.8 | |
| LightRAGCategory=Single-Step RAG, Base LLM=GPT-4o-mini2026.02 | 0.902 | 0.5 | 3.461 | 9.6 | |
| smolagentsCategory=Agentic Framework, Base LLM=GPT-4o-mini2026.02 | 1.079 | 0.5 | 6.82 | 5 | |
| LLM-onlyCategory=LLM-only, Base LLM=GPT-4o-mini2026.02 | 1.083 | 0.5 | 3.908 | 0 | |
| Single AgentCategory=Agentic Framework, Base LLM=GPT-4o-mini2026.02 | 1.522 | 0.5 | 5.939 | 3.1 |