Short-video platforms rely on personalized recommendation, raising concerns about filter bubbles that narrow content exposure. Auditing such phenomena at scale is challenging because real user studies are costly and privacy-sensitive, and existing simulators fail to reproduce realistic behaviors due to their reliance on textual signals and weak personalization. We propose PersonaAct, a framework for simulating short-video users with persona-conditioned multimodal agents trained on real behavioral traces for auditing filter bubbles in breadth and depth. PersonaAct synthesizes interpretable personas through automated interviews combining behavioral analysis with structured questioning, then trains agents on multimodal observations using supervised fine-tuning and reinforcement learning. We deploy trained agents for filter bubble auditing and evaluate bubble breadth via content diversity and bubble depth via escape potential. The evaluation demonstrates substantial improvements in fidelity over generic LLM baselines, enabling realistic behavior reproduction. Results reveal significant content narrowing over interaction. However, we find that Bilibili demonstrates the strongest escape potential. We release the first open multimodal short-video dataset and code to support reproducible auditing of recommender systems.

翻译：短视频平台依赖个性化推荐系统，引发了关于过滤气泡导致内容接触面收窄的担忧。对此现象进行大规模审计具有挑战性：真实用户研究成本高昂且涉及隐私敏感问题，而现有模拟器因依赖文本信号和弱个性化能力，难以复现真实用户行为。本文提出PersonaAct框架，该框架通过基于真实行为轨迹训练、以用户画像为条件的多模态智能体来模拟短视频用户，从而对过滤气泡的广度和深度进行审计。PersonaAct通过结合行为分析与结构化提问的自动化访谈来合成可解释的用户画像，随后利用监督微调与强化学习，基于多模态观察数据训练智能体。我们部署训练后的智能体进行过滤气泡审计，并通过内容多样性评估气泡广度，通过逃脱潜力评估气泡深度。评估结果表明，相较于通用的LLM基线模型，本方法在保真度上取得显著提升，能够实现真实的行为复现。结果显示，随着交互进行，内容窄化现象显著。然而，我们发现Bilibili平台展现出最强的逃脱潜力。我们发布了首个开源的多模态短视频数据集及代码，以支持可复现的推荐系统审计。