CARD: Controlled Agentic Reddit Discussions for Credit Card Simulation
2026-08-10 • Artificial Intelligence
Artificial IntelligenceMultiagent SystemsSocial and Information Networks
AI summaryⓘ
The authors created a system called CARD to mimic how people talk about credit cards in online discussions. Instead of just making random comments, CARD plans out the conversation by considering how replies are structured and what tone or stance people take. It then tweaks the generated discussion to make it more like real conversations seen on Reddit. Their tests show that CARD makes more realistic simulated conversations than other methods. This suggests that planning and careful adjustments help create believable online credit card discussions.
credit card discussionsonline forumslanguage modelsconversation simulationstancetonereply structureRedditcalibration looplexical and semantic metrics
Authors
Yaoning Yu, Kai-Min Chang, Ye Yu, Yi-Chia Wang, Haojing Luo, Haohan Wang
Abstract
Online credit card discussions provide a natural setting for studying how consumers communicate about financial products. Simulating these discussions requires more than just generating individual comments, the generated threads should also match how real users express themselves and interact with others. We introduce CARD, a framework for generating realistic credit card discussion threads. Given a credit card post and its matched real thread, CARD uses non-verbatim guidance on reply structure, comment function, stance, tone, and conversational variation. A planner organizes these controls, a writer generates the discussion, and a calibration loop updates comments' populations that contribute to differences between the generated and real thread distributions. We evaluate CARD on real Reddit credit card discussions using lexical, semantic, behavioral, and structural metrics. CARD matches the distributions of real credit card discussions better than simulation baselines across multiple LLMs and also demonstrates smaller effect sizes and distribution distances across metrics. These results show that structured planning and targeted revision can generate the realism of simulated credit card discussions.