Author

Shuowei Jin

Observed name: Jin, Shuowei

Institutions: Georgia Institute of Technology, Amazon (United States), University of California, Los Angeles, University of Michigan, New York University

Observed institutions: Georgia Institute of Technology, Amazon (United States), University of California, Los Angeles, University of Michigan, New York University

OpenAlex ID: A5135296145

Rethinking Importance Sampling in LLM Policy Optimization: A Cumulative Token Perspective

2026 · Zhang, Yuheng, Ye, Chenlu, et al. · arXiv (Cornell University) · 13

CoMem: Context Management with A Decoupled Long-Context Model

2026 · Zhang, Yuwei, Dong, Chengyu, et al. · arXiv (Cornell University) · 7

Learning with Rare Success but Rich Feedback via Reflection-Enhanced Self-Distillation

2026 · Zhang, Yuwei, Li, Sha, et al. · arXiv (Cornell University) · 7

T$^2$PO: Uncertainty-Guided Exploration Control for Stable Multi-Turn Agentic Reinforcement Learning

2026 · Wang, Haixin, Cui, Hejie, et al. · arXiv (Cornell University) · 6

Harness-Delta Attribution: Dissecting Gains From Harness Evolution

2026 · Ding, Wenxuan, Lu, Qin, et al. · arXiv (Cornell University)

Works count: 5

Citation count: 33

H-index: 4

I10-index: 1