SalesSim

SalesSim is a benchmark developed by arXiv researchers that evaluates multimodal large language models on retail persona alignment, with top models scoring below 79%. It specifically tests a model's ability to simulate realistic customer personas in

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.