Open-ended QA on Tele-Eval SAGIN domain
0.7685ScoreARMOR
Evaluation Results
| Method | Links | |
|---|---|---|
| ARMORretrieval=top-16, judge=GPT-5.22026.06 | 0.7685 | |
| InfoNCE QE FTretrieval=top-16, judge=GPT-5.22026.06 | 0.7662 | |
| Base RAGretrieval=top-16, judge=GPT-5.22026.06 | 0.766 | |
| Mix QE FTretrieval=top-16, judge=GPT-5.22026.06 | 0.7591 | |
| RAG QE FTretrieval=top-16, judge=GPT-5.22026.06 | 0.7573 | |
| Base Genjudge=GPT-5.22026.06 | 0.3017 |