Question Answering on RealtimeQA
86.7AccuracyKRE
Evaluation Results
| Method | Links | |
|---|---|---|
| KREBackbone LLM=qwen2.5-7b-instruct2025.06 | 86.7 | |
| FaithfulRAGBackbone LLM=qwen2.5-7b-instruct2025.06 | 84.1 | |
| ATTRBackbone LLM=qwen2.5-7b-instruct2025.06 | 83 | |
| Opin(Instr)Backbone LLM=qwen2.5-7b-instruct2025.06 | 81.4 | |
| FaithfulRAGBackbone LLM=llama3.1-8b-instruct2025.06 | 81.4 | |
| COIECDBackbone LLM=qwen2.5-7b-instruct2025.06 | 78.8 | |
| Origin model with full contextBackbone LLM=qwen2.5-7b-instruct2025.06 | 78.7 | |
| CADBackbone LLM=qwen2.5-7b-instruct2025.06 | 77 | |
| FaithfulRAGBackbone LLM=mistral-7b-instruct2025.06 | 77 | |
| ATTRBackbone LLM=llama3.1-8b-instruct2025.06 | 76.9 | |
| KREBackbone LLM=mistral-7b-instruct2025.06 | 76.9 | |
| Opin(Instr)Backbone LLM=llama3.1-8b-instruct2025.06 | 75.2 | |
| Origin model with full contextBackbone LLM=llama3.1-8b-instruct2025.06 | 67.3 | |
| COIECDBackbone LLM=llama3.1-8b-instruct2025.06 | 62.8 | |
| CADBackbone LLM=llama3.1-8b-instruct2025.06 | 61.9 | |
| COIECDBackbone LLM=mistral-7b-instruct2025.06 | 58.4 | |
| ChatQA-2.0Backbone LLM=Llama3.1-8B2025.06 | 57.5 | |
| ChatQA-1.5Backbone LLM=Llama3.1-8B2025.06 | 56.7 | |
| Self-RAGBackbone LLM=Llama2-7B2025.06 | 55.8 | |
| CADBackbone LLM=mistral-7b-instruct2025.06 | 55.8 | |
| Origin model with full contextBackbone LLM=mistral-7b-instruct2025.06 | 52.2 | |
| ATTRBackbone LLM=mistral-7b-instruct2025.06 | 52.2 | |
| Opin(Instr)Backbone LLM=mistral-7b-instruct2025.06 | 51.3 | |
| KREBackbone LLM=llama3.1-8b-instruct2025.06 | 48.6 | |
| Origin model without contextBackbone LLM=qwen2.5-7b-instruct2025.06 | 40.7 | |
| Origin model without contextBackbone LLM=mistral-7b-instruct2025.06 | 29.2 | |
| Origin model without contextBackbone LLM=llama3.1-8b-instruct2025.06 | 28.3 |