Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus
混合线性注意力大语言模型中的大规模激活:注意力前峰值与峰间平台期
Zunhai Su, Bohan Sun, Xialie Zhuang, Shuibai Zhang, He Xiao, Jing Xiong, Hengyuan Zhang, Zhongzhu Zhou, Tiantian Zhang, Ngai Wong, Chuan-Wei Kuo
机构
*
Startlux(星创公司)
;
Tsinghua University(清华大学)
;
University of Chinese Academy of Sciences(中国科学院大学)
;
The University of Hong Kong(香港大学)
;
University of Sydney(悉尼大学)
;
Columbia University(哥伦比亚大学)
Surfsvr: 2D Surface Priors as 3D Geometric Regularizers for Sparse Voxel Reconstruction
Surfsvr:将2D表面先验作为稀疏体素重建的3D几何正则化项
Yan Di, Chengxi Li, Yaoxing Wang, Mengge Liu, Zhigang Li, Ruida Zhang, Mingyang Li, Pengyuan Wang, Shan Gao, Xiangyang Ji
机构
*
HIT (Shenzhen)(哈尔滨工业大学(深圳))
;
Tsinghua University(清华大学)
;
NWPU(西北工业大学)
;
Beijing Institute of Control and Engineering(北京控制与工程研究所)
;
Technical University of Munich(慕尼黑工业大学)
机构
*
School of Artificial Intelligence, Beijing Normal University(北京师范大学人工智能学院)
;
Peking University(北京大学)
;
National Center for Applied Mathematics Shenzhen (NCAMS), Southern University of Science and Technology(南方科技大学深圳应用数学国家中心)
;
Department of Automation, Tsinghua University(清华大学自动化系)
;
National Research Council of Italy, Institute of Integrated Methodologies for Earth Observation (CNR-IMIOT)(意大利国家研究委员会综合地球观测方法研究所)
Drift and Dependence: Layer-wise Information-Theoretic Bounds for Replay-Based Continual Learning
漂移与依赖:基于重放的持续学习的分层信息论边界
Tieliang Gong, Zhongbo Zhang, Wen Wen, Yong-Jin Liu
机构
*
School of Computer Science and Technology, Xi’an Jiaotong University(西安交通大学计算机科学与技术学院)
;
Department of Computer Science and Technology, Tsinghua University(清华大学计算机科学与技术系)
ComBodied Agents: a New Paradigm of Human-Centric Agentic AI
具身融合智能体:以人为中心的智能体人工智能新范式
Qianggang Ding, Xingyao Wang, Rui Feng, Zhibin Wang, Feixiang Yao, Kelong Mao, Hao Sun, Zhiyao Luo, Jiankai Tang, Lei Li, Jiadong Guo, Minheng Ni, Weicong Lin, Chenxi Yang, Hongxiang Gao, Zhenghua Chen, Yang Bai, Min Wu, Jun Cheng, Huazhu Fu, Dacheng Tao, Bang Liu
机构
*
Université de Montréal(蒙特利尔大学)
;
Mila – Quebec Artificial Intelligence Institute(米拉-魁北克人工智能研究所)
;
Institute of Advanced Intelligence and Computing (IAIC), A*STAR(新加坡科技研究局高级智能与计算研究所)
;
Nanjing Medical University(南京医科大学)
;
Nanjing University(南京大学)
;
Renmin University of China(中国人民大学)
;
University of Cambridge(剑桥大学)
;
University of Oxford(牛津大学)
;
Tsinghua University(清华大学)
;
National University of Singapore(新加坡国立大学)
;
The Hong Kong University of Science and Technology(香港科技大学)
;
The Hong Kong Polytechnic University(香港理工大学)
;
Southern University of Science and Technology(南方科技大学)
;
Southeast University(东南大学)
;
University of Glasgow(格拉斯哥大学)
;
Nanyang Technological University(南洋理工大学)
AI总结
该研究提出以人为中心的 Combodied Agents 新范式,整合多类智能体能力形成闭环,聚焦人类状态轨迹建模,推动智能体 AI 从任务完成转向人类持续福祉。
机构
*
School of Computer Science, Peking University(北京大学计算机科学学院)
;
Microsoft Research Asia(微软亚洲研究院)
;
Princeton University(普林斯顿大学)
;
Tsinghua University(清华大学)
机构
*
Tsinghua University(清华大学)
;
Zhongguancun Academy(中关村学院)
;
Infinigence AI
;
Peking University(北京大学)
;
UC Berkeley(加州大学伯克利分校)
;
Harbin Institute of Technology(哈尔滨工程学院)
;
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所)
CommentsAccepted to RSS 2026. This is the technical report of the RLinf Team, focusing on the algorithm side. For the system-level design, please refer to arXiv:2509.15965. The open-sourced code link: https://github.com/RLinf/RLinf
Grounding Large Language Models as Generalizable Policies in Network Control
大语言模型作为网络优化的通用策略
Duo Wu, Linjia Kang, Zhimin Wang, Fangxin Wang, Wei Zhang, Chongbo Sun, Xuefeng Tao, Wei Yang, Le Zhang, Wenwu Zhu, Peng Cui, Zhi Wang
机构
*
Bytedance(字节跳动)
;
Shenzhen International Graduate School(深圳国际研究生院)
;
Tsinghua University(清华大学)
;
The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳))
;
Department of Computer Science and Technology(计算机科学与技术系)