Retrieval on Airline Services SOP Benchmark
0.76MRRSOPRAG
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| SOPRAGBackbone=GPT-4o2026.02 | 0.76 | 0.64 | 0.89 | 0.93 | |
| OpenAI Embedmodel=text-embedding-3-small2026.02 | 0.75 | 0.67 | 0.83 | 0.86 | |
| BM252026.02 | 0.69 | 0.58 | 0.79 | 0.83 | |
| HippoRAG 2Backbone=GPT-4o2026.02 | 0.63 | 0.52 | 0.7 | 0.86 | |
| LightRAGBackbone=GPT-4o2026.02 | 0.43 | 0.29 | 0.54 | 0.71 | |
| GraphRAGBackbone=GPT-4o2026.02 | 0.4 | 0.26 | 0.49 | 0.67 |