This study explores how LLM-driven simulations improve public engagement with oral history, revealing complex narratives from the Cultural Revolution.
Recent advances in generative artificial intelligence offer unprecedented opportunities to reimagine archival sources and deepen public engagement with contested histories. This study investigates how large‐language‐model (LLM) multi‐agent simulations can transform oral‐history interviews from China's Cultural Revolution in Memories: The CR/10 Project into dynamic dialogues among artificial agents. Using the open‐source tool TinyTroupe, we construct personas grounded in authentic testimony and embed them in themed conversation spaces focused on memory, reconciliation, intergenerational understanding, and historical responsibility. The resulting simulations enable audiences to observe how divergent personal narratives collide, converge, and evolve across generations, providing a richer appreciation of this complex past. Preliminary findings indicate that LLM‐driven agents can convey multifaceted historical experience in a more immersive and interactive form than traditional interfaces.
No takes yet. Share an insight, caveat, or question.
Zhuang et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: