Repeated Negotiation on Repeated Negotiation vs Llama (test)
5.56Buyer Average RewardBoN-oppo
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| BoN-oppoBase Model=Gemini, Opponent=Llama2026.02 | 5.56 | 5.23 | |
| BoN-simulationBase Model=Gemini, Opponent=Llama2026.02 | 5.41 | 2.53 | |
| BoN-evalBase Model=Gemini, Opponent=Llama2026.02 | 2.18 | -0.81 | |
| BoN-oppo (iid)Base Model=Gemini, Opponent=Llama2026.02 | 0.76 | 3.5 | |
| Baseline w. thinkingBase Model=Gemini, Opponent=Llama2026.02 | -2.19 | 1.73 |