The hallucination neuroimaging study was accepted to ICML 2026.
research.log LLMs -> people
杜邦得
Bangde Du
Ph.D. Student, Department of Computer Science and Technology, Tsinghua University
I study how large language models can empower, understand, and simulate people.
Research Focus
Understanding people through language models.
My research focuses on large language models, agents, personalization, and digital twins, with an emphasis on how AI systems can better model human values, behaviors, and communication patterns.
- LLM Evaluation
- Agents
- Personalization
- Digital Twins
- Persona Simulation
- User Behavior Analysis
News
Recent Updates
The public opinion simulation paper was accepted to KDD 2026.
TwinVoice was accepted to ACL 2026 Findings.
SimVBG was accepted to EMNLP 2025.
The EEG study on polarized short-video content was published in Scientific Reports.
How do Humans Process AI-generated Hallucination Contents: a Neuroimaging Study
ICML 2026
Parametric Social Identity Injection and Diversification in Public Opinion Simulation
KDD 2026
TwinVoice: A Multi-dimensional Benchmark Towards Digital Twins via LLM Persona Simulation
ACL 2026 Findings
SimVBG: Simulating Individual Values by Backstory Generation
EMNLP 2025 Main Conference
Understanding the Effect of Opinion Polarization in Short Video Browsing
SIGIR 2025
EEG Reveals the Cognitive Impact of Polarized Content in Short Video Scenarios
Scientific Reports, 2025
Education
Academic Background
Tsinghua University
Ph.D. Student, Department of Computer Science and Technology
Awards
Honors & Scholarships
Tsinghua Friends - Bosch Talent Scholarship, First Class
2026
Tsinghua University Scholarship
2022
Second Prize, 40th Tsinghua University Challenge Cup
First author
Contact