Reward Performance on Car Negotiation Low-data
40.92Test-Time RewardMC
Evaluation Results
| Method | Links | |
|---|---|---|
| MCtraining_regime=BOOST2026.05 | 40.92 | |
| MCtraining_regime=Bilevel Only2026.05 | 36.28 | |
| ILQLtraining_regime=BOOST2026.05 | 34.91 | |
| MCtraining_regime=Vanilla+Syn2026.05 | 31.08 | |
| ILQLtraining_regime=Bilevel Only2026.05 | 29.08 | |
| ILQLtraining_regime=Vanilla2026.05 | 27.22 | |
| MCtraining_regime=Vanilla2026.05 | 26.73 | |
| ILQLtraining_regime=Vanilla+Syn2026.05 | 25.94 |