List your company

Question Answering on AbsR

91.82AccuracyKALE
58.43667.10375.7784.437Jan 12, 2026

Evaluation results

MethodLinks
KALEBackbone=Qwen2.5 32B, Category=Augmented-based2026.0191.82
KALEBackbone=LlaMA3 8B, Category=Augmented-based2026.0183.62
GPT3MixBackbone=Qwen2.5 32B, Category=Augmented-based2026.0180.1
KG-SFTBackbone=Qwen2.5 32B, Category=SFT-based2026.0178.91
AugGPTBackbone=Qwen2.5 32B, Category=Augmented-based2026.0178.91
KG-SFTBackbone=LlaMA3 8B, Category=SFT-based2026.0178.2
KALEBackbone=Mistral 7B, Category=Augmented-based2026.0176.9
SDFTBackbone=LlaMA3 8B, Category=SFT-based2026.0176.15
GraphRAGBackbone=LlaMA3 8B, Category=Retrieval-based2026.0175.83
GraphRAGBackbone=Qwen2.5 32B, Category=Retrieval-based2026.0175.24
DMTBackbone=Qwen2.5 32B, Category=SFT-based2026.0175.24
TOGBackbone=Qwen2.5 32B, Category=Retrieval-based2026.0174.64
DMTBackbone=LlaMA3 8B, Category=SFT-based2026.0174.57
StructGPTBackbone=Qwen2.5 32B, Category=Retrieval-based2026.0174.17
SDFTBackbone=Mistral 7B, Category=SFT-based2026.0173.82
SDFTBackbone=Qwen2.5 32B, Category=SFT-based2026.0173.34
DMTBackbone=Mistral 7B, Category=SFT-based2026.0173.22
STaRBackbone=Qwen2.5 32B, Category=Augmented-based2026.0172.99
KG-SFTBackbone=Mistral 7B, Category=SFT-based2026.0172.39
SFTBackbone=Qwen2.5 32B, Category=SFT-based2026.0172.03
MeanLearnBackbone=LlaMA3 8B, Category=SFT-based2026.0171.09
MeanLearnBackbone=Qwen2.5 32B, Category=SFT-based2026.0171.09
MeanLearnBackbone=Mistral 7B, Category=SFT-based2026.0170.97
STaRBackbone=Mistral 7B, Category=Augmented-based2026.0170.02
STaRBackbone=LlaMA3 8B, Category=Augmented-based2026.0169.95
CoTBackbone=Qwen2.5 32B, Category=Prompt-based2026.0168.72
SFTBackbone=Mistral 7B, Category=SFT-based2026.0168.48
GPT3MixBackbone=LlaMA3 8B, Category=Augmented-based2026.0168.27
GraphRAGBackbone=Mistral 7B, Category=Retrieval-based2026.0168.26
SFTBackbone=LlaMA3 8B, Category=SFT-based2026.0167.77
CoTBackbone=Mistral 7B, Category=Prompt-based2026.0167.18
VanillaBackbone=Qwen2.5 32B, Category=Prompt-based2026.0166.35
TOGBackbone=LlaMA3 8B, Category=Retrieval-based2026.0165.98
StructGPTBackbone=LlaMA3 8B, Category=Retrieval-based2026.0165.35
AugGPTBackbone=Mistral 7B, Category=Augmented-based2026.0165.28
StructGPTBackbone=Mistral 7B, Category=Retrieval-based2026.0165.17
TOGBackbone=Mistral 7B, Category=Retrieval-based2026.0164.6
AugGPTBackbone=LlaMA3 8B, Category=Augmented-based2026.0164.45
CoTBackbone=LlaMA3 8B, Category=Prompt-based2026.0163.15
VanillaBackbone=LlaMA3 8B, Category=Prompt-based2026.0162.68
VanillaBackbone=Mistral 7B, Category=Prompt-based2026.0162.35
GPT3MixBackbone=Mistral 7B, Category=Augmented-based2026.0159.72