5 papers
The Bureaucratization of the Internet: Analyzing the Diffusion of Governance Regimes on Reddit (2011--2023)
Katherine Van Koevering, Yuanhao Liu, Jon Kleinberg
As online platforms scale, distributed communities must develop complex governance structures to maintain order. Reddit represents a unique experiment in this process: millions of…
MIDSim: Simulating Multi-Channel Information Diffusion in Social Media with LLM-Powered Multi-Agent System
Lexi Liu, Qi Cao, Yuanhao Liu +2
Information diffusion in social media shapes public opinion and collective behavior, making its modeling and simulation an important research problem. Existing studies have investi…
Agent System Operations: Categorization, Challenges, and Future Directions
Zexin Wang, Changhua Pei, Yuanhao Liu +10
As the reasoning capabilities of Large Language Models (LLMs) continue to advance, LLM-based agent systems offer advantages in flexibility and interpretability over traditional sys…
Why Do Aligned LLMs Remain Jailbreakable: Refusal-Escape Directions, Operator-Level Sources, and Safety-Utility Trade-off
Yu Chen, Yuanhao Liu, Qi Cao
Aligned large language models (LLMs) remain vulnerable to jailbreak attacks. Recent mechanistic studies have identified latent features and representation shifts associated with ja…
A Survey on AgentOps: Categorization, Challenges, and Future Directions
Zexin Wang, Jingjing Li, Quan Zhou +7
As the reasoning capabilities of Large Language Models (LLMs) continue to advance, LLM-based agent systems offer advantages in flexibility and interpretability over traditional sys…