User Modeling

Follow this topic Ask about this topic

Momentum

12 papers in the last four weeks, up 100% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 101

All topics

Hierarchical Multi-Persona Induction from User Behavioral Logs: Learning Evidence-Grounded and Truthful Personas
Apr 28, 2026Nayoung Choi, Haeyu Jeong, Changbong Kim +2Personality Modeling in Language Models User Modeling
Department of Computer Science, Emory University, Atlanta, GA, USA · Naver Corporation, South Korea
A Survey on LLM-based Conversational User Simulation
Apr 27, 2026Bo Ni, Leyao Wang, Yu Wang +27Human Behavior Simulation User Modeling
Vanderbilt University · Yale University · University of Oregon +8
EgoSelf: From Memory to Personalized Egocentric Assistant
Apr 21, 2026Yanshuo Wang, Yuan Xu, Xuesong Li +4Agent Memory User Modeling
Eastern Institute of Technology, Ningbo · Hong Kong Polytechnic University · Peking University +2
Assessing Capabilities of Large Language Models in Social Media Analytics: A Multi-task Quest
Apr 21, 2026Ramtin Davoudi, Kartik Thakkar, Nazanin Donyapour +2Social Media Analysis LLM Evaluation
Utah State University · Independent Researcher · Vanderbilt University
HORIZON: A Benchmark for In-the-wild User Behaviour Modeling
Apr 19, 2026Arnav Goel, Pranjal A Chitale, Bhawna Paliwal +2Sequential Recommendation User Modeling
Carnegie Mellon University · Microsoft Research India · University of California, Berkeley
CogEvolution: A Human-like Generative Educational Agent to Simulate Student's Cognitive Evolution
Apr 16, 2026Wei Zhang, Yihang Cheng, Zhirong Ye +1Cognitive Modeling Human Behavior Simulation
Faculty of Artificial Intelligence in Education, Central China Normal University, Wuhan, Hubei, China · National Engineering Research Center for Educational Big Data, Central China Normal University, Wuhan, Hubei, China · Central China Normal University Wollongong Joint Institute, Central China Normal University, Wuhan, Hubei, China
DeepTutor: Towards Agentic Personalized Tutoring
Apr 10, 2026Bingxi Zhao, Jiahao Zhang, Xubin Ren +4LLM Personalization Intelligent Tutoring Systems
The University of Hong Kong
Can LLMs Model Incorrect Student Reasoning? A Case Study on Distractor Generation
Mar 16, 2026Yanick Zengaffinen, Andreas Opedal, Donya Rooein +3LLM Prompting Generative AI in Education
1ETH Zürich · 2Bocconi University · University of Central Florida
Modeling Distinct Human Interaction in Web Agents
Feb 19, 2026Faria Huq, Zora Zhiruo Wang, Zhanqiu Guo +6User Modeling Human-AI Collaboration
Carnegie Mellon University · Duke University
Modeling Story Expectations: A Generative Framework using LLMs
Dec 13, 2024Hortense Fong, George Gui, Bo YangUser Modeling
Columbia Business School
"What Are You Really Trying to Do?": Co-Creating Life Goals from Everyday Computer Use
Date pendingShardul Sapkota, Matthew Jörke, Zane Sabbagh +3User Modeling Human-AI Co-Creation
Stanford University Stanford, USA

Apr 28, 2026, cs.AIJ/K move · Enter open · S save

Hierarchical Multi-Persona Induction from User Behavioral Logs: Learning Evidence-Grounded and Truthful Personas

Nayoung Choi, Haeyu Jeong, Changbong Kim, Hongjun Lim+1

Department of Computer Science, Emory University, Atlanta, GA, USA · Naver Corporation, South Korea

Personality Modeling in Language Models User Modeling

Open paper AskPDF arXiv

Behavioral logs provide rich signals for user modeling, but are noisy and interleaved across diverse intents. Recent work uses LLMs to generate interpretable natural-language personas from user logs, yet evaluation often emphasizes downstream utility, providing limited assurance of persona quality itself. We propose a hierarchical framework that aggregates user actions into intent memories and induces multiple evidence-grounded personas by clustering and labeling these memories. We formulate persona induction as an optimization problem over persona quality-captured by cluster cohesion, persona-evidence alignment, and persona truthfulness-and train the persona model using a groupwise extension of Direct Preference Optimization (DPO). Experiments on a large-scale service log and two public datasets show that our method induces more coherent, evidence-grounded, and trustworthy personas, while also improving future interaction prediction.

Momentum

Related topics

Latest papers 101