Location Prediction on Location answer-based questions
90.4Location AccuracyZEPHYRUS-REFLECTIVE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ZEPHYRUS-REFLECTIVELLM=gpt-5.22025.10 | 90.4 | 1,452 | |
| ZEPHYRUS-DIRECTLLM=gpt-5.22025.10 | 89.9 | 1,277 | |
| ZEPHYRUS-DIRECTLLM=gpt-5-mini2025.10 | 88.4 | 1,353 | |
| ZEPHYRUS-REFLECTIVELLM=gpt-5-mini2025.10 | 87.9 | 1,486 | |
| ZEPHYRUS-REFLECTIVELLM=gpt-oss-120b2025.10 | 78.3 | 1,472 | |
| ZEPHYRUS-REFLECTIVELLM=gemini-2.5-flash2025.10 | 76.8 | 1,651 | |
| ZEPHYRUS-DIRECTLLM=gemini-2.5-flash2025.10 | 76.3 | 1,890 | |
| ZEPHYRUS-DIRECTLLM=gpt-oss-120b2025.10 | 67.2 | 1,870 | |
| ZEPHYRUS-REFLECTIVELLM=qwen3-30b2025.10 | 67.2 | 1,597 | |
| ZEPHYRUS-DIRECTLLM=qwen3-30b2025.10 | 60.6 | 1,370 | |
| Text Only LLMLLM=gpt-5.22025.10 | 18.2 | 5,445 | |
| Text Only LLMLLM=gpt-5-mini2025.10 | 18.2 | 5,772 | |
| Text Only LLMLLM=gemini-2.5-flash2025.10 | 16.2 | 3,750 | |
| Text Only LLMLLM=qwen3-30b2025.10 | 13.1 | 3,899 | |
| Text Only LLMLLM=gpt-oss-120b2025.10 | 9.09 | 2,823 |