Knowledge Base Question Answering on WebQuestion Freebase (test)
81.02Hits@1 AccuracyGPT-4o-mini + KG
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4o-mini + KGStrategy=KG-Augmented without Fine-Tuning2026.03 | 81.02 | |
| DeepSeek-R1-Distill-Llama-70B + KGSize=70B, Strategy=KG-Augmented without Fine-Tuning2026.03 | 75.8 | |
| LLaMA-3.3-70B + KGSize=70B, Strategy=KG-Augmented without Fine-Tuning2026.03 | 72.6 | |
| DeepSeek-R1-Distill-Llama-70BSize=70B, Strategy=LLM Prompting Only2026.03 | 68.84 | |
| KG-Hopper w/Qwen-2.5-7BSize=7B, Strategy=KG-Augmented with RL Fine-Tuning2026.03 | 66.9 | |
| KG-CoT w/GPT 3.5-TurboStrategy=KG-Augmented without Fine-Tuning2026.03 | 66.5 | |
| GPT-4oStrategy=LLM Prompting Only2026.03 | 64.79 | |
| Qwen-2.5-7B (SFT) + KGSize=7B, Strategy=KG-Augmented with Supervised Fine-Tuning2026.03 | 61.42 | |
| LLaMA-3.1-8B (SFT) + KGSize=8B, Strategy=KG-Augmented with Supervised Fine-Tuning2026.03 | 60 | |
| LLaMA-3.3-70BSize=70B, Strategy=LLM Prompting Only2026.03 | 59.73 | |
| GPT-4o-miniStrategy=LLM Prompting Only2026.03 | 57.26 | |
| LLaMA-3.1-8B + KGSize=8B, Strategy=KG-Augmented without Fine-Tuning2026.03 | 57.05 | |
| Qwen-2.5-7B + KGSize=7B, Strategy=KG-Augmented without Fine-Tuning2026.03 | 56.33 | |
| LLaMA-3.1-8BSize=8B, Strategy=LLM Prompting Only2026.03 | 45.88 | |
| Qwen-2.5-7BSize=7B, Strategy=LLM Prompting Only2026.03 | 44.23 |