Agentic decision making on Weather
58.3ScoreAMC
Evaluation Results
| Method | Links | |
|---|---|---|
| AMCPolicy Model=GPT-5.1, Search Budget=15, Value Model=Llama-3.2-11B2026.06 | 58.3 | |
| AMCPolicy Model=GPT-5.1, Search Budget=5, Value Model=Llama-3.2-11B2026.06 | 56.7 | |
| Best-of-15Policy Model=Llama-3.2-11B2026.06 | 55 | |
| SMC (Zero-shot)Policy Model=Llama-3.2-11B, Search Budget=15, Value Model=Llama-3.2-11B2026.06 | 55 | |
| AMCPolicy Model=Llama-3.2-11B, Search Budget=15, Value Model=Llama-3.2-11B2026.06 | 55 | |
| AMCPolicy Model=GPT-4.1-mini, Search Budget=5, Value Model=Llama-3.2-11B2026.06 | 55 | |
| SMC (Zero-shot)Policy Model=GPT-4.1-mini, Search Budget=15, Value Model=Llama-3.2-11B2026.06 | 55 | |
| AMCPolicy Model=GPT-4.1-mini, Search Budget=15, Value Model=Llama-3.2-11B2026.06 | 55 | |
| ReActPolicy Model=GPT-5.12026.06 | 55 | |
| Best-of-5Policy Model=GPT-5.12026.06 | 55 | |
| SMC (Zero-shot)Policy Model=GPT-5.1, Search Budget=5, Value Model=Llama-3.2-11B2026.06 | 55 | |
| Best-of-15Policy Model=GPT-5.12026.06 | 55 | |
| SMC (Zero-shot)Policy Model=GPT-5.1, Search Budget=15, Value Model=Llama-3.2-11B2026.06 | 55 | |
| AMCPolicy Model=Llama-3.2-11B, Search Budget=5, Value Model=Llama-3.2-11B2026.06 | 53.3 | |
| Best-of-15Policy Model=GPT-4.1-mini2026.06 | 53.3 | |
| Best-of-5Policy Model=Llama-3.2-11B2026.06 | 51.7 | |
| SMC (Zero-shot)Policy Model=GPT-4.1-mini, Search Budget=5, Value Model=Llama-3.2-11B2026.06 | 51.7 | |
| SMC (Zero-shot)Policy Model=Llama-3.2-11B, Search Budget=5, Value Model=Llama-3.2-11B2026.06 | 48.3 | |
| Best-of-5Policy Model=GPT-4.1-mini2026.06 | 48.3 | |
| ReActPolicy Model=GPT-4.1-mini2026.06 | 45 | |
| ReActPolicy Model=Llama-3.2-11B2026.06 | 43.3 |