Songjun Tu

Venues

Frequent coauthors

Papers

1. AutoSearch: Adaptive Search Depth for Efficient Agentic RAG via Reinforcement Learning
   ACL 2026 ·

   Jingbo Sun

   DBLP profile ↗ORCID search ↗
2. AlphaDecay: Module-wise Weight Decay for Heavy-Tailed Balancing in LLMs
   NeurIPS 2025 ·

   Di He

   DBLP profile ↗ORCID search ↗
3. In-Dataset Trajectory Return Regularization for Offline Preference-based Reinforcement Learning
   AAAI 2025 · Songjun Tu
4. Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RL
   NeurIPS 2025 · Songjun Tu
5. Offline Goal-Conditioned Reinforcement Learning with Elastic-Subgoal Diffused Policy Learning
   AAMAS 2025 ·

   Yaocheng Zhang

   DBLP profile ↗ORCID search ↗
6. Online Preference-based Reinforcement Learning with Self-augmented Feedback from Large Language Model
   AAMAS 2025 · Songjun Tu
7. Salience-Invariant Consistent Policy Learning for Generalization in Visual Reinforcement Learning
   AAMAS 2025 ·

   Jingbo Sun

   DBLP profile ↗ORCID search ↗
8. Unsupervised Zero-Shot Reinforcement Learning via Dual-Value Forward-Backward Representation
   ICLR 2025 ·

   Jingbo Sun

   DBLP profile ↗ORCID search ↗