Task-oriented Plan Retrieval on To Do List Single-turn
84Pass RateSemantic XPath
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Semantic XPathModel=GPT-5 mini, Scorer=Entail2026.03 | 84 | 5,184 | |
| In-contextModel=GPT-5 mini, Scorer=–2026.03 | 80 | 5,444 | |
| In-contextModel=Gemini 3 Flash, Scorer=–2026.03 | 72 | 10,465 | |
| Semantic XPathModel=Gemini 3 Flash, Scorer=Entail2026.03 | 72 | 5,605 | |
| Semantic XPathModel=Gemini 3 Flash, Scorer=Sem2026.03 | 68 | 5,710 | |
| Semantic XPathModel=GPT-5 mini, Scorer=Sem2026.03 | 64 | 5,937 | |
| Flat RAGModel=Gemini 3 Flash, Scorer=–2026.03 | 48 | 2,934 | |
| Flat RAGModel=GPT-5 mini, Scorer=–2026.03 | 40 | 2,722 |