WeSearch

Grounding AI shopping agents using personas learned from raw clickstream data

·3 min read · 0 reactions · 0 comments · 24 views
#artificial intelligence#e-commerce#buyer personas
Grounding AI shopping agents using personas learned from raw clickstream data
TL;DR · WeSearch summary

A new framework called SimPersona has been introduced to enhance e-commerce agents by learning discrete buyer personas from raw clickstream data. This approach aims to overcome the limitations of existing personalization methods that rely on average buyer profiles. SimPersona has shown promising results, achieving a high conversion-rate alignment with real buyers while maintaining the diversity of buyer behavior.

Key facts
Original article
arXiv.org
Read full at arXiv.org →
Opening excerpt (first ~120 words) tap to expand

Computer Science > Artificial Intelligence arXiv:2605.14205 (cs) [Submitted on 14 May 2026] Title:SimPersona: Learning Discrete Buyer Personas from Raw Clickstreams for Grounded E-Commerce Agents Authors:Zahra Zanjani Foumani, Alberto Castelo, Shuang Xie, Ted Chaiwachirasak, Han Li, Lingyun Wang View a PDF of the paper titled SimPersona: Learning Discrete Buyer Personas from Raw Clickstreams for Grounded E-Commerce Agents, by Zahra Zanjani Foumani and 5 other authors View PDF HTML (experimental) Abstract:LLM-based web agents can navigate live storefronts, yet they often collapse to a single "average buyer" policy, failing to capture the heterogeneous and distributional nature of real buyer populations.

Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv.org.

Anonymous · no account needed
Share 𝕏 Facebook Reddit LinkedIn Threads WhatsApp Bluesky Mastodon Email

Discussion

0 comments

More from arXiv.org