Decision Making on HotelBookings
100Accuracy (vs Bayes-optimal)SFT with chain-of-thought reasoning
Evaluation Results
| Method | Links | |
|---|---|---|
| SFT with chain-of-thought reasoningFraming=Original, Trained=Yes2026.03 | 100 | |
| SFT with chain-of-thought reasoningFraming=Wording, Trained=Yes2026.03 | 100 | |
| SFT with chain-of-thought reasoningFraming=Dollar, Trained=Yes2026.03 | 98.3 |