Question Answering on AbsR
91.82AccuracyKALE
Evaluation Results
| Method | Links | |
|---|---|---|
| KALEBackbone=Qwen2.5 32B, Category=Augmented-based2026.01 | 91.82 | |
| KALEBackbone=LlaMA3 8B, Category=Augmented-based2026.01 | 83.62 | |
| GPT3MixBackbone=Qwen2.5 32B, Category=Augmented-based2026.01 | 80.1 | |
| KG-SFTBackbone=Qwen2.5 32B, Category=SFT-based2026.01 | 78.91 | |
| AugGPTBackbone=Qwen2.5 32B, Category=Augmented-based2026.01 | 78.91 | |
| KG-SFTBackbone=LlaMA3 8B, Category=SFT-based2026.01 | 78.2 | |
| KALEBackbone=Mistral 7B, Category=Augmented-based2026.01 | 76.9 | |
| SDFTBackbone=LlaMA3 8B, Category=SFT-based2026.01 | 76.15 | |
| GraphRAGBackbone=LlaMA3 8B, Category=Retrieval-based2026.01 | 75.83 | |
| GraphRAGBackbone=Qwen2.5 32B, Category=Retrieval-based2026.01 | 75.24 | |
| DMTBackbone=Qwen2.5 32B, Category=SFT-based2026.01 | 75.24 | |
| TOGBackbone=Qwen2.5 32B, Category=Retrieval-based2026.01 | 74.64 | |
| DMTBackbone=LlaMA3 8B, Category=SFT-based2026.01 | 74.57 | |
| StructGPTBackbone=Qwen2.5 32B, Category=Retrieval-based2026.01 | 74.17 | |
| SDFTBackbone=Mistral 7B, Category=SFT-based2026.01 | 73.82 | |
| SDFTBackbone=Qwen2.5 32B, Category=SFT-based2026.01 | 73.34 | |
| DMTBackbone=Mistral 7B, Category=SFT-based2026.01 | 73.22 | |
| STaRBackbone=Qwen2.5 32B, Category=Augmented-based2026.01 | 72.99 | |
| KG-SFTBackbone=Mistral 7B, Category=SFT-based2026.01 | 72.39 | |
| SFTBackbone=Qwen2.5 32B, Category=SFT-based2026.01 | 72.03 | |
| MeanLearnBackbone=LlaMA3 8B, Category=SFT-based2026.01 | 71.09 | |
| MeanLearnBackbone=Qwen2.5 32B, Category=SFT-based2026.01 | 71.09 | |
| MeanLearnBackbone=Mistral 7B, Category=SFT-based2026.01 | 70.97 | |
| STaRBackbone=Mistral 7B, Category=Augmented-based2026.01 | 70.02 | |
| STaRBackbone=LlaMA3 8B, Category=Augmented-based2026.01 | 69.95 | |
| CoTBackbone=Qwen2.5 32B, Category=Prompt-based2026.01 | 68.72 | |
| SFTBackbone=Mistral 7B, Category=SFT-based2026.01 | 68.48 | |
| GPT3MixBackbone=LlaMA3 8B, Category=Augmented-based2026.01 | 68.27 | |
| GraphRAGBackbone=Mistral 7B, Category=Retrieval-based2026.01 | 68.26 | |
| SFTBackbone=LlaMA3 8B, Category=SFT-based2026.01 | 67.77 | |
| CoTBackbone=Mistral 7B, Category=Prompt-based2026.01 | 67.18 | |
| VanillaBackbone=Qwen2.5 32B, Category=Prompt-based2026.01 | 66.35 | |
| TOGBackbone=LlaMA3 8B, Category=Retrieval-based2026.01 | 65.98 | |
| StructGPTBackbone=LlaMA3 8B, Category=Retrieval-based2026.01 | 65.35 | |
| AugGPTBackbone=Mistral 7B, Category=Augmented-based2026.01 | 65.28 | |
| StructGPTBackbone=Mistral 7B, Category=Retrieval-based2026.01 | 65.17 | |
| TOGBackbone=Mistral 7B, Category=Retrieval-based2026.01 | 64.6 | |
| AugGPTBackbone=LlaMA3 8B, Category=Augmented-based2026.01 | 64.45 | |
| CoTBackbone=LlaMA3 8B, Category=Prompt-based2026.01 | 63.15 | |
| VanillaBackbone=LlaMA3 8B, Category=Prompt-based2026.01 | 62.68 | |
| VanillaBackbone=Mistral 7B, Category=Prompt-based2026.01 | 62.35 | |
| GPT3MixBackbone=Mistral 7B, Category=Augmented-based2026.01 | 59.72 |