Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Knowledge Base Question Answering on Creak WikiData (test)
Loading...
91.82
Hits@1
KG-Hopper w/Qwen-2.5-7B
72.5176
77.5288
82.54
87.5512
Mar 22, 2026
Hits@1
Updated 25d ago
Evaluation Results
Method
Method
Links
Hits@1
KG-Hopper w/Qwen-2.5-7B
Size=7B, Strategy=KG-A...
2026.03
91.82
DeepSeek-R1-Distill-Llama-70B + KG
Size=70B, Strategy=KG-...
2026.03
91.2
GPT-4o
Strategy=LLM Prompting...
2026.03
90.7
GPT-4o-mini + KG
Strategy=KG-Augmented...
2026.03
90.2
LLaMA-3.3-70B + KG
Size=70B, Strategy=KG-...
2026.03
87.04
LLaMA-3.1-8B (SFT) + KG
Size=8B, Strategy=KG-A...
2026.03
84.47
LLaMA-3.3-70B
Size=70B, Strategy=LLM...
2026.03
83.72
GPT-4o-mini
Strategy=LLM Prompting...
2026.03
83.72
Qwen-2.5-7B (SFT) + KG
Size=7B, Strategy=KG-A...
2026.03
83.43
LLaMA-3.1-8B + KG
Size=8B, Strategy=KG-A...
2026.03
80.5
DeepSeek-R1-Distill-Llama-70B
Size=70B, Strategy=LLM...
2026.03
79.1
Qwen-2.5-7B + KG
Size=7B, Strategy=KG-A...
2026.03
79.04
LLaMA-3.1-8B
Size=8B, Strategy=LLM...
2026.03
75.8
Qwen-2.5-7B
Size=7B, Strategy=LLM...
2026.03
73.26
Feedback
Search any
task
Search any
task