User Simulation on ConvApparel Domain Expert Academic Verbose V2 (held-out agent)
12.1Average Words per TurnDynamic, Aware
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Dynamic, AwareZero-shot=true, Simulator Paradigm=Dynamic, Aware (Ours)2026.05 | 12.1 | 38.5 | 13.6 | 31.2 | |
| Ground Truth DataZero-shot=true, Simulator Paradigm=Ground Truth Data2026.05 | 12 | 38.7 | 13.4 | 31.7 | |
| Dynamic, AgnosticZero-shot=true, Simulator Paradigm=Dynamic, Agnostic2026.05 | 11.2 | 35.1 | 14.5 | 22.3 | |
| Static, AwareZero-shot=true, Simulator Paradigm=Static, Aware2026.05 | 10.5 | 32 | 15.1 | 25 | |
| Static, AgnosticZero-shot=true, Simulator Paradigm=Static, Agnostic (Standard)2026.05 | 8.8 | 28 | 18 | 12.5 |