Open-World Knowledge Graph Question Answering on GLOW-Bench RH 2
54Exact Match AccuracyGLOW-G
Evaluation results
| Method | Links | |
|---|---|---|
| GLOW-GLLM-Model=DeepSeek-V32026.04 | 54 | |
| GLOW-GLLM-Model=GPT-4o-mini2026.04 | 53 | |
| GLOW-GNLLM-Model=DeepSeek-V32026.04 | 53 | |
| GLOW-GNLLM-Model=GPT-4o-mini2026.04 | 49 | |
| GoGLLM-Model=DeepSeek-V32026.04 | 47 | |
| GLOW-GLLM-Model=Qwen3-8B2026.04 | 44 | |
| GLOW-GNLLM-Model=Qwen3-8B2026.04 | 42 | |
| GLOW-GNLLM-Model=Granite-3.3-8B-Instruct2026.04 | 42 | |
| GLOW-GNLLM-Model=DeepSeek-R1-Distill-Qwen-7B2026.04 | 39 | |
| GLOW-GLLM-Model=Granite-3.3-8B-Instruct2026.04 | 39 | |
| GLOW-NLLM-Model=Qwen3-8B2026.04 | 37 | |
| AskGNNLLM-Model=Granite-3.3-8B-Instruct2026.04 | 35 | |
| GLOW-NLLM-Model=GPT-4o-mini2026.04 | 35 | |
| AskGNNLLM-Model=Qwen3-8B2026.04 | 34 | |
| GLOW-NLLM-Model=DeepSeek-V32026.04 | 34 | |
| GLOW-NLLM-Model=DeepSeek-R1-Distill-Qwen-7B2026.04 | 32 | |
| GLOW-NLLM-Model=Granite-3.3-8B-Instruct2026.04 | 32 | |
| AskGNNLLM-Model=DeepSeek-R1-Distill-Qwen-7B2026.04 | 31 | |
| GoGLLM-Model=GPT-4o-mini2026.04 | 31 | |
| LLM-OnlyLLM-Model=DeepSeek-V32026.04 | 30 | |
| GoGLLM-Model=Qwen3-8B2026.04 | 29 | |
| LLM-OnlyLLM-Model=GPT-4o-mini2026.04 | 28 | |
| GLOW-GLLM-Model=DeepSeek-R1-Distill-Qwen-7B2026.04 | 27 | |
| GoGLLM-Model=Granite-3.3-8B-Instruct2026.04 | 27 | |
| LLM-OnlyLLM-Model=Qwen3-8B2026.04 | 19 | |
| GCRLLM-Model=Granite-3.3-8B-Instruct2026.04 | 19 | |
| GCRLLM-Model=DeepSeek-R1-Distill-Qwen-7B2026.04 | 16 | |
| GCRLLM-Model=Qwen3-8B2026.04 | 15 | |
| LLM-OnlyLLM-Model=Granite-3.3-8B-Instruct2026.04 | 14 | |
| GoGLLM-Model=DeepSeek-R1-Distill-Qwen-7B2026.04 | 13 | |
| LLM-OnlyLLM-Model=DeepSeek-R1-Distill-Qwen-7B2026.04 | 12 |