Persona-driven decision-making on LifeChoice
85.53AccuracyGPT-5
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-52026.01 | 85.53 | |
| Claude Sonnet 4.52026.01 | 85.49 | |
| HumanLLM-32BParameters=32B2026.01 | 50.64 | |
| Qwen3-32BParameters=32B2026.01 | 47.71 | |
| HumanLLM-8BParameters=8B2026.01 | 47.19 | |
| Qwen3-8B (OT+CoSER)Parameters=8B, Training Data=OpenThoughts + CoSER2026.01 | 46.99 | |
| Qwen3-8BParameters=8B2026.01 | 44.51 |