4 code implementations • 27 Feb 2024 • Shuming Ma, Hongyu Wang, Lingxiao Ma, Lei Wang, Wenhui Wang, Shaohan Huang, Li Dong, Ruiping Wang, Jilong Xue, Furu Wei
Recent research, such as BitNet, is paving the way for a new era of 1-bit Large Language Models (LLMs).
8 code implementations • 17 Jul 2023 • Yutao Sun, Li Dong, Shaohan Huang, Shuming Ma, Yuqing Xia, Jilong Xue, Jianyong Wang, Furu Wei
In this work, we propose Retentive Network (RetNet) as a foundation architecture for large language models, simultaneously achieving training parallelism, low-cost inference, and good performance.
no code implementations • 8 Apr 2023 • Xiaonan Nie, Xupeng Miao, Zilong Wang, Zichao Yang, Jilong Xue, Lingxiao Ma, Gang Cao, Bin Cui
We first present an empirical analysis on the problems and opportunities of training MoE models, which motivates us to overcome the routing imbalance and fluctuation problems by a dynamic expert management and device placement mechanism.
2 code implementations • 29 Dec 2021 • Xiaonan Nie, Xupeng Miao, Shijie Cao, Lingxiao Ma, Qibin Liu, Jilong Xue, Youshan Miao, Yi Liu, Zhi Yang, Bin Cui
Then it diversifies the experts and continues to train the MoE with a novel Dense-to-Sparse gate (DTS-Gate).
no code implementations • 19 Oct 2018 • Lingxiao Ma, Zhi Yang, Youshan Miao, Jilong Xue, Ming Wu, Lidong Zhou, Yafei Dai
This evolution has led to large graph-based irregular and sparse models that go beyond what existing deep learning frameworks are designed for.
no code implementations • 22 May 2018 • Jilong Xue, Youshan Miao, Cheng Chen, Ming Wu, Lintao Zhang, Lidong Zhou
Its computation is typically characterized by a simple tensor data abstraction to model multi-dimensional matrices, a data-flow graph to model computation, and iterative executions with relatively frequent synchronizations, thereby making it substantially different from Map/Reduce style distributed big data computation.