Search papers, labs, and topics across Lattice.
The paper introduces Synonymix, a pipeline for constructing unified group personas from multiple individual life story personas using graph-based abstraction and merging. This approach enables meso-level simulations that bridge the gap between individual agent interactions and population-level analysis. The evaluation shows that synthetic agents generated from Synonymix preserve behavioral signals beyond demographic baselines while providing privacy guarantees.
Unlock richer, more realistic agent simulations by moving beyond individual personas to unified group representations that capture collective behavior.
Generative agent simulations operate at two scales: individual personas for character interaction, and population models for collective behavior analysis and intervention testing. We propose a third scale: meso-level simulation - interaction with group-level representations that retain grounding in rich individual experience. To enable this, we present Synonymix, a pipeline that constructs a "unigraph" from multiple life story personas via graph-based abstraction and merging, producing a queryable collective representation that can be explored for sensemaking or sampled for synthetic persona generation. Evaluating synthetic agents on General Social Survey items, we demonstrate behavioral signal preservation beyond demographic baselines (p<0.001, r=0.59) with demonstrable privacy guarantee (max source contribution <13%). We invite discussion on interaction modalities enabled by meso-level simulations, and whether "high-fidelity" personas can ever capture the texture of lived experience.
Stanford HAI