Knowledge Base Question Answering on Creak WikiData (test)
91.82Hits@1KG-Hopper w/Qwen-2.5-7B
Evaluation Results
| Method | Links | |
|---|---|---|
| KG-Hopper w/Qwen-2.5-7BSize=7B, Strategy=KG-Augmented with RL Fine-Tuning2026.03 | 91.82 | |
| DeepSeek-R1-Distill-Llama-70B + KGSize=70B, Strategy=KG-Augmented without Fine-Tuning2026.03 | 91.2 | |
| GPT-4oStrategy=LLM Prompting Only2026.03 | 90.7 | |
| GPT-4o-mini + KGStrategy=KG-Augmented without Fine-Tuning2026.03 | 90.2 | |
| LLaMA-3.3-70B + KGSize=70B, Strategy=KG-Augmented without Fine-Tuning2026.03 | 87.04 | |
| LLaMA-3.1-8B (SFT) + KGSize=8B, Strategy=KG-Augmented with Supervised Fine-Tuning2026.03 | 84.47 | |
| LLaMA-3.3-70BSize=70B, Strategy=LLM Prompting Only2026.03 | 83.72 | |
| GPT-4o-miniStrategy=LLM Prompting Only2026.03 | 83.72 | |
| Qwen-2.5-7B (SFT) + KGSize=7B, Strategy=KG-Augmented with Supervised Fine-Tuning2026.03 | 83.43 | |
| LLaMA-3.1-8B + KGSize=8B, Strategy=KG-Augmented without Fine-Tuning2026.03 | 80.5 | |
| DeepSeek-R1-Distill-Llama-70BSize=70B, Strategy=LLM Prompting Only2026.03 | 79.1 | |
| Qwen-2.5-7B + KGSize=7B, Strategy=KG-Augmented without Fine-Tuning2026.03 | 79.04 | |
| LLaMA-3.1-8BSize=8B, Strategy=LLM Prompting Only2026.03 | 75.8 | |
| Qwen-2.5-7BSize=7B, Strategy=LLM Prompting Only2026.03 | 73.26 |