Fanqi Kong

Aegis: Automated Error Generation and Attribution for Multi-Agent Systems

SIV-Bench: A Video Benchmark for Social Interaction Understanding and Reasoning

Adaptive Punishment for Cooperation in Mixed-Motive Games

ToMPO: Training LLM Strategic Decision Making from a Multi-Agent Perspective

Automating Environments For Measuring Agentic Learning

Learning to Balance Altruism and Self-interest Based on Empathy in Mixed-Motive Games

Planning with Theory of Mind for Few-Shot Adaptation in Sequential Social Dilemmas