Submitted by
Yuanzhi Wang
AI & ML interests
The official organization of Tencent Hunyuan team
Recent Activity
View all activity
Papers
Harness Self-Evolution for Long-Horizon Terminal Agents
Tree-Aware Posterior Distillation for Long-Horizon Agentic Reinforcement Learning
