Lu Pan
Identifiers
No identifiers captured yet.
Papers (3)
- From $\log \pi$ to $\pi$: Taming Divergence in Soft Clipping via Bilateral Decoupled Decay of Probability Gradient Weight cs.LG · 2026 · author #8
- How to Allocate, How to Learn? Dynamic Rollout Allocation and Advantage Modulation for Policy Optimization cs.LG · 2026 · author #7
- MASPO: Unifying Gradient Utilization, Probability Mass, and Signal Reliability for Robust and Sample-Efficient LLM Reasoning cs.LG · 2026 · author #8
Mentions
No mention provenance yet.
Frequent Coauthors
- Chaowen Hu 3 shared papers
- Cong Qin 3 shared papers
- Jiaye Lin 3 shared papers
- Ke Zeng 3 shared papers
- Xiaoliang Fu 3 shared papers
- Yangyi Fang 3 shared papers
- Binbin Zheng 2 shared papers
- Xunliang Cai 2 shared papers
- Zekai Shao 2 shared papers
- Haolin Shi 1 shared papers