List your company
ResearchFormerly WizWand

Knowledge Base Question Answering on T-REx WikiData (test)

83.38Hits@1KG-R1
14.7432.5650.3868.2Sep 30, 2025Oct 28, 2025Nov 26, 2025Dec 25, 2025Jan 23, 2026Feb 21, 2026Mar 22, 2026
Updated 5mo ago

Evaluation results

MethodLinks
KG-R1Backbone Model=Qwen2.5 3B, Training Dataset=WebQSP2025.0983.38
KG-R1Backbone Model=Qwen2.5 3B, Training Dataset=CWQ2025.0980.1
PoGBackbone Model=Qwen 2.5-72B2025.0979.52
PoGBackbone Model=Qwen 3-235B2025.0978.75
ReknoSBackbone Model=Qwen 2.5-7B2025.0978.17
ReknoSBackbone Model=Qwen 2.5-72B2025.0977.11
ReknoSBackbone Model=Qwen 3-235B2025.0977.05
ToGBackbone Model=Qwen 3-235B2025.0977
PoGBackbone Model=Qwen 2.5-7B2025.0975.69
KG-Hopper w/Qwen-2.5-7BSize=7B, Strategy=KG-Augmented with RL Fine-Tuning2026.0372.14
DeepSeek-R1-Distill-Llama-70B + KGSize=70B, Strategy=KG-Augmented without Fine-Tuning2026.0372.04
LLaMA-3.1-8B (SFT) + KGSize=8B, Strategy=KG-Augmented with Supervised Fine-Tuning2026.0370.23
GPT-4o-mini + KGStrategy=KG-Augmented without Fine-Tuning2026.0369.7
Qwen-2.5-7B (SFT) + KGSize=7B, Strategy=KG-Augmented with Supervised Fine-Tuning2026.0368.77
ToGBackbone Model=Qwen 2.5-72B2025.0967.93
LLaMA-3.1-8B + KGSize=8B, Strategy=KG-Augmented without Fine-Tuning2026.0365.8
LLaMA-3.3-70B + KGSize=70B, Strategy=KG-Augmented without Fine-Tuning2026.0365.42
Qwen-2.5-7B + KGSize=7B, Strategy=KG-Augmented without Fine-Tuning2026.0364.21
ToGBackbone Model=Qwen 2.5-7B2025.0958.55
GPT-4oStrategy=LLM Prompting Only2026.0344.46
SCBackbone Model=Qwen 3-235B2025.0938.79
CoTBackbone Model=Qwen 3-235B2025.0937.72
RoGBackbone Model=Qwen 3-235B2025.0937.31
IOBackbone Model=Qwen 3-235B2025.0934.61
DeepSeek-R1-Distill-Llama-70BSize=70B, Strategy=LLM Prompting Only2026.0334.21
RoGBackbone Model=Qwen 2.5-72B2025.0932.5
Qwen-2.5-7BSize=7B, Strategy=LLM Prompting Only2026.0331.15
GPT-4o-miniStrategy=LLM Prompting Only2026.0326.9
IOBackbone Model=Qwen 2.5-72B2025.0926.16
SCBackbone Model=Qwen 2.5-72B2025.0926.01
CoTBackbone Model=Qwen 2.5-72B2025.0924.48
LLaMA-3.1-8BSize=8B, Strategy=LLM Prompting Only2026.0323
IOBackbone Model=Qwen 2.5-7B2025.0922.79
RoGBackbone Model=Qwen 2.5-7B2025.0922.37
LLaMA-3.3-70BSize=70B, Strategy=LLM Prompting Only2026.0320.12
SCBackbone Model=Qwen 2.5-7B2025.0920.01
CoTBackbone Model=Qwen 2.5-7B2025.0917.38