Abstract
Existing evaluations often test persona expression rather than whether persona information persists in extended decision-making. We introduce THEATRON, a controlled benchmark that uses financial replay to evaluate persona persistence in long-horizon sequential decision-making.
In these environments, models conditioned on fixed behavioral-finance personas observe shared daily market evidence and trajectory-specific account-state feedback and make sequential decisions over multi-month trajectories.
Experimental Setup
We evaluate Zero-Shot, One-Shot, and Chain-of-Thought persona injection across four LLMs, using cumulative-return and action trajectories to diagnose within-persona stability, cross-persona separability, and action collapse.
Key Findings
- Persona conditioning produces systematic behavioral shifts beyond a no-persona baseline.
- Persona-level separation generally exceeds the variation induced by semantically equivalent descriptions.
- However, these shifts do not reliably form coherent persona-specific policies.
- Apparently stable separation can arise from stereotyped action collapse.
- Diverse policies can exhibit noisy trajectory switching.
These findings expose a gap between observable persona-induced behavior and persistent, context-sensitive persona-conditioned decision policies.
Availability
Code and benchmark data are publicly available at:
https://github.com/yuyan12138/theatron