Tu, Songjun

4 publications

NeurIPS 2025 AlphaDecay: Module-Wise Weight Decay for Heavy-Tailed Balancing in LLMs Di He, Songjun Tu, Ajay Jaiswal, Li Shen, Ganzhao Yuan, Shiwei Liu, Lu Yin

AAAI 2025 In-Dataset Trajectory Return Regularization for Offline Preference-Based Reinforcement Learning Songjun Tu, Jingbo Sun, Qichao Zhang, Yaocheng Zhang, Jia Liu, Ke Chen, Dongbin Zhao

NeurIPS 2025 Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RL Songjun Tu, Jiahao Lin, Qichao Zhang, Xiangyu Tian, Linjing Li, Xiangyuan Lan, Dongbin Zhao

ICLR 2025 Unsupervised Zero-Shot Reinforcement Learning via Dual-Value Forward-Backward Representation Jingbo Sun, Songjun Tu, Qichao Zhang, Haoran Li, Xin Liu, Yaran Chen, Ke Chen, Dongbin Zhao