Reasoning and Synthesis on Bitcoin Price Fluctuations (BP)
99.06FaithfulnessDirect
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| DirectBackbone=LLaMA-3.1-8B-Instruct2025.06 | 99.06 | 24.42 | 61.65 | |
| DirectBackbone=DeepSeek-R1-Distill-LLaMA-8B2025.06 | 98.75 | 41.77 | 71.64 | |
| DirectBackbone=Qwen2.5-7B-Instruct2025.06 | 97.62 | 38.1 | 70.68 | |
| RAGBackbone=DeepSeek-R1-Distill-LLaMA-8B2025.06 | 97.2 | 43.78 | 76.06 | |
| CoT-RAGBackbone=DeepSeek-R1-Distill-LLaMA-8B2025.06 | 94.98 | 47.15 | 79.89 | |
| SARGBackbone=Qwen2.5-7B-Instruct2025.06 | 94.83 | 65 | 86.72 | |
| RAGBackbone=Qwen2.5-7B-Instruct2025.06 | 92.34 | 42.4 | 76.11 | |
| StructRAGBackbone=GPT-4o-mini2025.06 | 92.26 | 45 | 86.12 | |
| RAGBackbone=LLaMA-3.1-8B-Instruct2025.06 | 92.04 | 29.32 | 55.24 | |
| GraphRAGBackbone=GPT-4o-mini2025.06 | 91.91 | 52.73 | 82.33 | |
| SARGBackbone=GPT-4o-mini2025.06 | 91.9 | 70.4 | 87.9 | |
| HyperGraphRAGBackbone=GPT-4o-mini2025.06 | 91.83 | 53.65 | 84.36 | |
| HippoRAG2Backbone=GPT-4o-mini2025.06 | 91.54 | 64.29 | 84.7 | |
| CoT-RAGBackbone=Qwen2.5-7B-Instruct2025.06 | 89.42 | 38.88 | 78.7 | |
| SARGBackbone=DeepSeek-R1-Distill-LLaMA-8B2025.06 | 86.32 | 47.04 | 77.91 | |
| SARGBackbone=LLaMA-3.1-8B-Instruct2025.06 | 82.34 | 42.2 | 56.57 | |
| CoT-RAGBackbone=LLaMA-3.1-8B-Instruct2025.06 | 80.74 | 27.91 | 55.65 |