Zhenghai Xue

发表论文 7 篇 · 总被引 455 次 · h-index 4

代表论文

• Group-in-Group Policy Optimization for LLM Agent Training (2025 · Neural Information Processing Systems · 被引 413)
• SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning (2025 · arXiv.org · 被引 147)
• AgentStudio: A Toolkit for Building General Virtual Agents (2024 · International Conference on Learning Representations · 被引 47)
• S2AC: Energy-Based Reinforcement Learning with Stein Soft Actor Critic (2024 · International Conference on Learning Representations · 被引 22)
• The Optimal Token Baseline: Variance Reduction for Long-Horizon LLM-RL (2026 · arXiv.org · 被引 5)
• Policy Optimization under Imperfect Human Interactions with Agent-Gated Shared Autonomy (2025 · International Conference on Learning Representations)