Knowledge Base Question Answering on GrailQA (dev)
89.5Overall F1GAPD
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| GAPDLLM=Llama-3.1-8B, Method Category=Fine-tune-based Methods2026.05 | 89.5 | — | — | — | — | — | — | 87.8 | |
| GAPDLLM=Llama-3.1-8B, Method Category=Fine-tune-based Methods2026.05 | 88.7 | — | — | — | — | — | — | 87.8 | |
| KBQA-R1LLM=Llama-3.1-8B, Method Category=Fine-tune-based Methods2026.05 | 86.1 | — | — | — | — | — | — | 83.9 | |
| KBQA-R1LLM=Llama-3.1-8B, Method Category=Fine-tune-based Methods2026.05 | 85.2 | — | — | — | — | — | — | 83.6 | |
| KnowCoder-A1LLM=Qwen2.5-Coder-7B, Method Category=Fine-tune-based Methods2026.05 | 84.1 | — | — | — | — | — | — | — | |
| TIARALLM=T5-large, Method Type=Fine-tune-based Methods2025.12 | 81.9 | 88.4 | 91.2 | 66.4 | 74.8 | 73.3 | 80.7 | 75.3 | |
| KBQA-R1 HarnessLLM=Claude 4.6 Sonnet, Method Category=Prompting Methods2026.05 | 81.9 | — | — | — | — | — | — | 79.1 | |
| TIARALLM=T5-large, Method Category=Fine-tune-based Methods2026.05 | 81.9 | — | — | — | — | — | — | 75.3 | |
| DecAFLLM=T5-large, Method Type=Fine-tune-based Methods2025.12 | 81.4 | 88.7 | 92.4 | 71.5 | 79.8 | 65.9 | 77.3 | 72.5 | |
| DecAFLLM=T5-large, Method Category=Fine-tune-based Methods2026.05 | 81.4 | — | — | — | — | — | — | 72.5 | |
| TIARALLM=T5-large, Method Category=Fine-tune-based Methods2026.05 | 80.7 | — | — | — | — | — | — | 73.3 | |
| KnowCoder-A1LLM=Qwen2.5-Coder-7B, Method Category=Fine-tune-based Methods2026.05 | 80.5 | — | — | — | — | — | — | — | |
| KBQA-R1 HarnessLLM=gpt-5.3-codex, Method Category=Prompting Methods2026.05 | 79.6 | — | — | — | — | — | — | 77.4 | |
| KBQA-o1LLM=Llama-3.1-8B, Method Category=Fine-tune-based Methods2026.05 | 78.5 | — | — | — | — | — | — | 71.9 | |
| KBQA-R1 HarnessLLM=Kimi-K2.5, Method Category=Prompting Methods2026.05 | 78 | — | — | — | — | — | — | 73.9 | |
| DecAFLLM=T5-large, Method Category=Fine-tune-based Methods2026.05 | 77.3 | — | — | — | — | — | — | 65.9 | |
| RnG-KBQALLM=T5-large, Method Type=Fine-tune-based Methods2025.12 | 76.9 | 86.7 | 89 | 61.7 | 68.9 | 68.8 | 74.7 | 69.5 | |
| RnG-KBQALLM=T5-large, Method Category=Fine-tune-based Methods2026.05 | 76.9 | — | — | — | — | — | — | 69.5 | |
| KBQA-R1 HarnessLLM=Claude 4.6 Sonnet, Method Category=Prompting Methods2026.05 | 76.3 | — | — | — | — | — | — | 72 | |
| KBQA-o1LLM=Llama-3.1-8B, Method Category=Fine-tune-based Methods2026.05 | 76.1 | — | — | — | — | — | — | 68.1 | |
| KBQA-R1 HarnessLLM=GLM-5, Method Category=Prompting Methods2026.05 | 75.5 | — | — | — | — | — | — | 68.7 | |
| RnG-KBQALLM=T5-large, Method Category=Fine-tune-based Methods2026.05 | 74.7 | — | — | — | — | — | — | 68.8 | |
| KBQA-R1 HarnessLLM=Kimi-K2.5, Method Category=Prompting Methods2026.05 | 73.9 | — | — | — | — | — | — | 68.5 | |
| KBQA-R1 HarnessLLM=gpt-5.3-codex, Method Category=Prompting Methods2026.05 | 73.5 | — | — | — | — | — | — | 68.6 | |
| KBQA-R1 HarnessLLM=GLM-5, Method Category=Prompting Methods2026.05 | 70.2 | — | — | — | — | — | — | 60 | |
| ARG-KBQALLM=GPT-3.5-turbo, Method Category=Prompting Methods2026.05 | 52.1 | — | — | — | — | — | — | 46.6 | |
| ARG-KBQALLM=GPT-3.5-turbo, Method Type=Prompting Methods2025.12 | 48.5 | 46.6 | 51.5 | 36.4 | 41.8 | 46.6 | 52.1 | 43.8 | |
| ARG-KBQALLM=GPT-3.5-turbo, Method Category=Prompting Methods2026.05 | 48.5 | — | — | — | — | — | — | 43.8 | |
| KB-CoderLLM=GPT-3.5-turbo, Method Category=Prompting Methods2026.05 | 47.3 | — | — | — | — | — | — | 42.2 | |
| KB-CoderLLM=GPT-3.5-turbo, Method Type=Prompting Methods2025.12 | 44.9 | 40.6 | 45.5 | 34.5 | 38.6 | 42.2 | 47.3 | 40.1 | |
| KB-CoderLLM=GPT-3.5-turbo, Method Category=Prompting Methods2026.05 | 44.9 | — | — | — | — | — | — | 40.1 | |
| KB-BINDERLLM=Codex-davinci-002, Method Category=Prompting Methods2026.05 | 44 | — | — | — | — | — | — | 40.1 | |
| KB-BINDERLLM=Codex-davinci-002, Method Type=Prompting Methods2025.12 | 42.2 | 40 | 43.3 | 33.9 | 36.6 | 40.1 | 44 | 38.7 | |
| KB-BINDERLLM=Codex-davinci-002, Method Category=Prompting Methods2026.05 | 42.2 | — | — | — | — | — | — | 38.7 |