Tag
Proposes Online Agent-as-a-Judge, an evaluation framework that uses an in-world evaluator agent to actively generate situations for testing interactive social agents, improving coverage and reliability over passive methods.
PersonaArena is a dynamic simulation framework that uses a large corpus of social content and a multi-agent debating judge to evaluate and improve LLMs' ability to maintain coherent and authentic persona-level role-playing in realistic social scenarios.
ALSO introduces a framework for online strategy optimization in multi-agent social simulation, formulating multi-turn interaction as an adversarial bandit problem and using a neural surrogate for reward prediction. Experiments on the Sotopia benchmark show it outperforms static baselines and existing optimization methods.
This paper proposes Dual-Scale Evolutionary Policy Training (DEPT) to address the evolution impasse in social language agents, using asymmetric advantage reshaping to restore gradient signals during self-play.