Question Answering on HotpotQA (test) (Comprehensive Metrics)
0.822F1HGN
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| HGNModel Category=Specialized Models2022.01 | 0.822 | 0.692 | — | — | — | — | — | — | — | |
| POET-SQL_RoBERTaModel Category=Language Models, Model Size=Large2022.01 | 0.816 | 0.687 | — | — | — | — | — | — | — | |
| C2F ReaderModel Category=Specialized Models2022.01 | 0.812 | 0.68 | — | — | — | — | — | — | — | |
| SpanBERTModel Category=Language Models, Model Size=Large2022.01 | 0.812 | 0.674 | — | — | — | — | — | — | — | |
| SAEModel Category=Specialized Models2022.01 | 0.808 | 0.677 | — | — | — | — | — | — | — | |
| POET-SQL_BARTModel Category=Language Models, Model Size=Large2022.01 | 0.797 | 0.665 | — | — | — | — | — | — | — | |
| ReasonBERTModel Category=Language Models, Model Size=not Large2022.01 | 0.792 | 0.648 | — | — | — | — | — | — | — | |
| BERTModel Category=Language Models, Model Size=Large2022.01 | 0.734 | 0.591 | — | — | — | — | — | — | — | |
| DFGNModel Category=Specialized Models2022.01 | 0.693 | 0.557 | — | — | — | — | — | — | — | |
| IntegrationBackbone=Llama 3.1-70B2025.02 | 0.665 | — | 0.6922 | 0.667 | — | — | — | — | — | |
| IntegrationBackbone=Llama 3.1-8B2025.02 | 0.6476 | — | 0.6721 | 0.6509 | — | — | — | — | — | |
| GraphRAGBackbone=Llama 3.1-70B2025.02 | 0.646 | — | 0.672 | 0.6489 | — | — | — | — | — | |
| RAGBackbone=Llama 3.1-70B2025.02 | 0.6388 | — | 0.6634 | 0.6399 | — | — | — | — | — | |
| GraphRAGBackbone=Llama 3.1-8B2025.02 | 0.6166 | — | 0.6414 | 0.6208 | — | — | — | — | — | |
| RAGBackbone=Llama 3.1-8B2025.02 | 0.6004 | — | 0.6232 | 0.6047 | — | — | — | — | — | |
| DioRLLM=Qwen2.5-7B, Retrieval=BM252025.04 | 0.2548 | 0.163 | 0.2605 | 0.3052 | 2.524 | 2.553 | 1.263 | 194.395 | 12.273 | |
| RaDIOLLM=Qwen2.5-7B, Retrieval=BM252025.04 | 0.2236 | 0.124 | 0.2261 | 0.3023 | 3.024 | 3.985 | 1.765 | 200.398 | 13.562 | |
| BaseLLM=Qwen2.5-7B, Retrieval=BM252025.04 | 0.1951 | 0.111 | 0.1931 | 0.3003 | 2.904 | 4.386 | 2.094 | 302.894 | 19.146 |