Researcher — paper.dinq.me

作者位次 #4/9NeurIPS 2026Poster· Reinforcement Learning / Everything Else

Towards Optimal Pre-training Data Mixtures for Empowering Reinforcement Learning in LLMs

Xue Han · Qian Hu · Yitong Wang · WenChun Gao · Lianlian Zhang · Qing Wang · Lijun Mei · Zhaoxuefeng · Junlan Feng

数据与预训练强化学习与后训练OpenReview

作者位次 #3/8ACL 2026Poster· Natural Language Processing

Soft Orthogonal Low-Rank Adaptation for Knowledge Sharing in Large Language Model Continual Learning

Yitong Wang · Xue Han · WenChun Gao · Qian Hu · Jiahui Wang · Ziqing Wang · Lijun Mei · Junlan Feng

摘要

When large language models are used in real-world scenarios, continual learning (CL) becomes a non-trivial problem. In particular, continual learning with modern LLMs is challenged both by the substantial computational costs induced by their massive parameter scale, and by the limitations of curr...

效率与系统评测与基准检索与知识OpenReview 项目页

WenChun Gao

获取完整研究档案

联系方式

2026 顶会论文

Towards Optimal Pre-training Data Mixtures for Empowering Reinforcement Learning in LLMs

Soft Orthogonal Low-Rank Adaptation for Knowledge Sharing in Large Language Model Continual Learning

常见合作者