Prompt Optimization on Dataset with human annotations (test)
69AccuracyLinGO (RAG)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LinGO (RAG)Model=Gemini 2.5 Flash-Lite2026.02 | 69 | 69.9 | |
| FTModel=Qwen3-4B-Instruct-25072026.02 | 59 | 53.5 | |
| FTModel=DeepSeek-V2-Lite-Chat2026.02 | 30.2 | 34 | |
| FTModel=Mistral-7B2026.02 | 21.5 | 25 |