Repeated Negotiation on Repeated Negotiation vs Qwen (test)
5.43Buyer Average RewardBoN-oppo
Evaluation results
| Method | Links | ||
|---|---|---|---|
| BoN-oppoBase Model=Gemini, Opponent=Qwen2026.02 | 5.43 | 4.07 | |
| BoN-simulationBase Model=Gemini, Opponent=Qwen2026.02 | 2.17 | 2.63 | |
| BoN-evalBase Model=Gemini, Opponent=Qwen2026.02 | 1.5 | -1.61 | |
| BoN-oppo (iid)Base Model=Gemini, Opponent=Qwen2026.02 | 0.6 | 0.1 | |
| Baseline w. thinkingBase Model=Gemini, Opponent=Qwen2026.02 | 0.3 | -0.57 |