I am Rongxin Chen, currently a Master’s student in Computer Science and Technology at the Institute of Computing Technology, Chinese Academy of Sciences (ICT, CAS). I received my Bachelor’s degree from the Software College of Northeastern University (NEU). My research interests focus on LLM Agent, Personalization, Alignment, RolePlaying, and Social Simulation. I expect to graduate in 2027 and am currently on the job market! If you are interested in me, please feel free to contact me.🔥
🔥 News
- 2026.04: 🎉🎉 Our paper “HAG: Hierarchical Demographic Tree-based Agent Generation for Topic-Adaptive Simulation” was accepted in ACL 2026 (Main)!
- 2026.01: 🎉🎉 Honored to receive the Hua Luogeng Master Award (The only recipient in the institute) at ICT, CAS!
- 2025.11: 🥳 Presented a poster at the 13th National Conference on Social Media Processing (SMP 2025) in Wuhan, China.
- 2025.11: 🎉🎉 Gave an Oral presentation on “Multi-Personality Generation of LLMs at Decoding-time” at The 1st Workshop on LLM Agents for Social Simulation (CIKM 2025) in Seoul, Korea, and received the Outstanding Paper award!
- 2025.10: 🎉🎉 Our paper “Multi-Personality Generation of LLMs at Decoding-time” was accepted in WSDM 2026!
📝 Publications
HAG: Hierarchical Demographic Tree-based Agent Generation for Topic-Adaptive Simulation
Rongxin Chen, Tianyu Wu, Bingbing Xu*, Jiatang Luo, Xiucheng Xu, Huawei Shen
ACL 2026 (Main)
- Proposed HAG to decompose and model the conditional joint distribution of demographic attributes, enabling topic-adaptive population distribution trees. We also introduced the PACE evaluation framework for statistical-semantic dual-level assessment, achieving significant reduction in demographic errors.
Multi-Personality Generation of LLMs at Decoding-time
Rongxin Chen, Yunfan Li, Yige Yuan, Bingbing Xu*, Huawei Shen
WSDM 2026 & LASS Workshop of CIKM 2025 (Outstanding Paper)
- Proposed MPG, a flexible and robust decoding-time generation framework for multi-personality LLMs without requiring extra training or external reward models. We also designed the SCR algorithm for efficient generation, achieving significant improvements in MBTI simulation and role-playing experiments.
🎖 Honors and Awards
- 2026 Hua Luogeng Master Award (The only recipient in the institute), ICT CAS.
- 2025 E Fund FinTech Freshman Award, ICT CAS.
- 2023 National Scholarship for Undergraduates.
- 2023 National Encouragement Scholarship, Ministry of Education.
- 2023 Finalist Winner in the American Region Mathematical Contest in Modeling.
- 2023 National Gold Award in the 13th “Challenge Cup” National College Student Entrepreneurship Plan Competition.
📖 Educations
- 2024.09 - 2027.07 (Expected), Master in Computer Science and Technology, UCAS - ICT CAS.
- 2020.09 - 2024.07, Bachelor in Software Engineering (Artificial Intelligence), NEU.
💬 Presentations
- 2026.02, Poster presentation at WSDM 2026, Boise, USA.
- 2025.11, Poster at the 13th National Conference on Social Media Processing, Wuhan, China.
- 2025.11, Oral presentation at LASS Workshop of CIKM 2025, Seoul, South Korea.