paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达

更新于

返回新星榜

ZL

Ziniu Li

ICML 2026×7强化学习与后训练效率与系统安全可解释性多模态

新星分新星

77

一作贡献30

核心/参与30

方向广度10

合作网络12

1 一作 · 7 篇 · ICML 2026

获取完整研究档案

首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。

联系方式

实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。

2026 顶会论文

7

作者位次 #4/6ICML 2026Poster· Reinforcement Learning / Inverse

Non-Adversarial Imitation Learning Provably Free of Compounding Errors: The Value Flow Mechanism

Tian Xu · Chenyang Wang · Xiaochen Zhai · Ziniu Li · Yi-Chen Li · Yang Yu

安全可解释性强化学习与后训练OpenReview

作者位次 #4/19ICML 2026Poster· Reinforcement Learning / Everything Else

TreePO: Enhancing Policy Efficacy and Inference Efficiency with Tree Modeling

Yizhi Li · Qingshui Gu · Zhoufutu Wen · Ziniu Li · Ruibin Yuan · Tianshun Xing · Shuyue Guo · Tuney Zheng · 周欣 · Xingwei Qu · Wangchunshu Zhou · Zheng Zhang · Wei Shen · Wei Xue · Qian Liu · Chenghua Lin · Jian Yang · Ge Zhang · Wenhao Huang

效率与系统强化学习与后训练OpenReview

作者位次 #4/7ICML 2026Poster· Reinforcement Learning / Everything Else

Beyond Token-level Supervision: Unlocking the Potential of Decoding-based Regression via Reinforcement Learning

Ming Chen · Sheng Tang · Rong-Xi Tan · Ziniu Li · Jiacheng Chen · Ke Xue · Chao Qian

多模态强化学习与后训练代码OpenReview

作者位次 #3/13ICML 2026Poster· Deep Learning / Large Language Models

The Optimal Token Baseline: Variance Reduction for Long-Horizon LLM-RL

Yingru Li · Jiawei Xu · Ziniu Li · Jiacai Liu · Wei Liu · Yuxuan Tong · Longtao Zheng · Zhenghai Xue · Yaxiang Zhang · Tianle Cai · Ge Zhang · Qian Liu · Baoxiang Wang

强化学习与后训练OpenReview

一作ICML 2026Poster· Deep Learning / Large Language Models

Knapsack RL: Compute-Efficient Reinforcement Learning via Heterogeneous Rollout Allocation

Ziniu Li · Congliang Chen · Tianyun Yang · Tian Ding · Ruoyu Sun · Ge Zhang · Wenhao Huang · Zhiquan Luo

强化学习与后训练效率与系统OpenReview

作者位次 #5/6ICML 2026Poster· Deep Learning / Large Language Models

Trust Region Masking for Long-Horizon LLM Reinforcement Learning

Yingru Li · Jiacai Liu · Jiawei Xu · Yuxuan Tong · Ziniu Li · Baoxiang Wang

强化学习与后训练OpenReview

作者位次 #3/4ICML 2026Poster· Deep Learning / Large Language Models

OnePO: Direct One-stage Policy Optimization for SFT-free Domain Adaptation

Junying Chen · Xinyuan Xie · Ziniu Li · Benyou Wang

强化学习与后训练OpenReview

常见合作者

GZGe Zhang×3 QLQian Liu×2 WHWenhao Huang×2 YLYingru Li×2 JXJiawei Xu×2 JLJiacai Liu×2 YTYuxuan Tong×2 BWBaoxiang Wang×2 TXTian Xu CWChenyang Wang XZXiaochen Zhai YLYi-Chen Li