Mathematical Reasoning and Formula Selection on UAV-Math-Bench
75AccuracyOpenAI o1
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| OpenAI o1RAG=true2025.06 | 75 | 95 | 15 | 21.05 | |
| OpenAI o1-previewRAG=true2025.06 | 70 | 85 | 10 | 11.76 | |
| ChatGPT-4-TurboRAG=true2025.06 | 60 | 75 | 20 | 26.67 | |
| DeepSeek R1RAG=true2025.06 | 50 | 90 | 35 | 38.89 | |
| LLaMA3.3 70BRAG=true2025.06 | 35 | 65 | 30 | 46.15 | |
| ChatGPT TurboRAG=false2025.06 | 30 | 75 | 45 | 60 | |
| LLaMA3.2 3BRAG=true2025.06 | 10 | 60 | 50 | 83.33 | |
| Mistral 7BRAG=true2025.06 | 5 | 70 | 55 | 78.57 |