HLHaibin Lin

Haibin Lin

*This is an algorithmically generated author record based on publicly available information.

Haibin Lin has a total of 120 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    MegaScale-MoE: Large-Scale Communication-Efficient Training of Mixture-of-Experts Models in Production

    by Chao Jin, Ziheng Jiang, Zhihao Bai, Zheng Zhong, Juncai Liu, Xiang Li, Ningxin Zheng, Xi Wang, Cong Xie, Qi Huang, et al.

    We present MegaScale-MoE, a production system tailored for the efficient training of large-scale mixture-of-experts (MoE) models. MoE emerges as a promising architecture to scale large language models (LLMs) to unprecedented sizes, thereby enhancing model performance. However, existing MoE training systems experience a degradation in training efficiency, exacerbated by the escalating scale of MoE models and the continuous evolution of hardware. Recognizing the pivotal role of efficient communication in enhancing MoE training, MegaScale-MoE customizes communication-efficient parallelism strategies for attention and FFNs in each MoE layer and adopts a holistic approach to overlap communication with computation at both inter- and intra-operator levels. Additionally, MegaScale-MoE applies communication compression with adjusted communication patterns to lower precision, further improving training efficiency. When training a 352B MoE model on 1,440 NVIDIA Hopper GPUs, MegaScale-MoE achieves a training throughput of 1.41M tokens/s, improving the efficiency by 1.88X compared to Megatron-LM. We share our operational experience in accelerating MoE training and hope that by offering our insights in system design, this work will motivate future research in MoE systems.


  • Open Access

    MegaScale-Data: Scaling DataLoader for Multisource Large Foundation Model Training

    by Juntao Zhao, Qi Lu, Wei Jia, Borui Wan, Lei Zuo, Junda Feng, Jianyu Jiang, Yangrui Chen, Shuaishuai Cao, Jialing He, et al.

    Modern frameworks for training large foundation models (LFMs) employ dataloaders in a data-parallel manner, with each loader processing a disjoint subset of training data. When preparing data for LFM training that originates from multiple, distinct sources, two fundamental challenges arise. First, due to the quadratic computational complexity of the attention operator, the non-uniform sample distribution over data-parallel ranks leads to significant workload imbalance among dataloaders, degrading the training efficiency. Second, supporting diverse data sources requires per-dataset file access states that are redundantly replicated across parallel loaders, consuming excessive memory. This also hinders dynamic data mixing (e.g., curriculum learning) and causes redundant access/memory overhead in hybrid parallelism. We present MegaScale-Data, an industrial-grade distributed data loading architecture for multisource LFMs training, with three key innovations: (1) Disaggregated data preprocessing via role-specific actors (Source Loaders/Data Constructors) to eliminate source and parallelism redundant data access and ensure multisource scalability. (2) Centralized and declarative data plane for load-time multisource orchestration, such as long-short context, multimodality, and curriculum learning. (3) Multi-level auto-partitioning and scaling mechanism for source loaders under heterogeneous preprocessing costs. We also contribute our designs and operational experience in deployment and fault tolerance. MegaScale-Data achieves up to: (1) 4.5X end-to-end training throughput improvement, and (2) 13.5X reduction in CPU memory usage.

  • Open Access

    Laminar: A Scalable Asynchronous RL Post-Training Framework

    by Guangming Sheng, Yuxuan Tong, Borui Wan, Wang Zhang, Chaobo Jia, Xibin Wu, Yuqi Wu, Xiang Li, Chi Zhang, Yanghua Peng, et al.

    Reinforcement learning (RL) post-training for Large Language Models (LLMs) is now scaling to large clusters and running for extended durations to enhance model reasoning performance. However, the scalability of existing RL frameworks is limited, as extreme long-tail skewness in RL trajectory generation causes severe GPU underutilization. Current asynchronous RL systems attempt to mitigate this, but they rely on global weight synchronization between the actor and all rollouts, which creates a rigid model update schedule. This global synchronization is ill-suited for the highly skewed and evolving distribution of trajectory generation latency in RL training, crippling training efficiency. Our key insight is that efficient scaling requires breaking this lockstep through trajectory-level asynchrony, which generates and consumes each trajectory independently. We propose Laminar, a scalable and robust RL post-training system built on a fully decoupled architecture. First, we replace global updates with a tier of relay workers acting as a distributed parameter service. This enables asynchronous and fine-grained weight synchronization, allowing rollouts to pull the latest weight anytime without stalling the actor's training loop. Second, a dynamic repack mechanism consolidates long-tail trajectories onto a few dedicated rollouts, maximizing generation throughput. The fully decoupled design also isolates failures, ensuring robustness for long-running jobs. Our evaluation on a 1024–GPU cluster shows that Laminar achieves up to 5.48X training throughput speedup over state-of-the-art systems, while reducing model convergence time.


  • Open Access

    MegaScale-Omni: A Hyper-Scale, Workload-Resilient System for MultiModal LLM Training in Production

    by Chunyu Xue, Yangrui Chen, Jianyu Jiang, Ningxin Zheng, Junda Feng, Jingji Chen, Shixiong Zhao, Shen Yan, Yi Lin, Lei Shi, et al.

    As the foundational component of versatile AI applications, training an multimodal large language model (MLLM) relies on multimodal datasets with dynamic modality mixture proportions and sample length distributions. However, existing MLLM systems remain inefficient under dynamic workloads, due to statically coupled decisions of resource allocation and model parallelization between encoders and the LLM backbone. This paper presents MegaScale-Omni, an industrial-grade MLLM training system tailored for dynamic workload adaption and hyper-scale deployment. MegaScale-Omni is built upon the training scheme of encoder-LLM multiplexing with three key innovations: (1) Decoupled parallelism strategies with long-short sequence parallelism for encoders to process variable-length samples, and full-fledged 5D parallelism for the LLM backbone, both organized under a communication-efficient parallelization layout. (2) Unified encoder-LLM representations for flexible, extensible colocation, and a new paradigm of encoder-LLM joint pipeline with workload resilience. (3) Workload balancing techniques via decentralized grouped reordering in data loaders and adaptive resharding from encoder to LLM ranks. MegaScale-Omni is deployed as the foundation of our in-house large-scale MLLM training tasks with thousands of GPUs. Our experimental results demonstrate 1.27×–7.57× throughput improvement under production-grade dynamic workloads, as compared to four state-of-the-art systems.

  • Open Access

    It Takes Two to Entangle

    by Zhanghan Wang, Ding Ding, Hang Zhu, Haibin Lin, Aurojit Panda


  • Open Access

    SwiftSpec: Disaggregated Speculative Decoding and Fused Kernels for Low-Latency LLM Inference

    by Ziyi Zhang, Ziheng Jiang, Chengquan Jiang, Menghan Yu, Size Zheng, Haibin Lin, Xin Liu, Henry Hoffmann

  • Open Access

    Robust LLM Training Infrastructure at ByteDance

    by Borui Wan, Gaohong Liu, Zuquan Song, Jun Wang, Yun Zhang, Guangming Sheng, Shuguang Wang, Houmin Wei, Chenyuan Wang, Weiqiang Lou, et al.


  • Open Access

    MegaScale-Infer: Efficient Mixture-of-Experts Model Serving with Disaggregated Expert Parallelism

    by Ruidong Zhu, Ziheng Jiang, Chao Jin, Peng Wu, Cesar A. Stuardo, Dongyang Wang, Xinlei Zhang, Huaping Zhou, Haoran Wei, Yang Cheng, et al.


  • Open Access

    ByteScale: Communication-Efficient Scaling of LLM Training with a 2048K Context Length on 16384 GPUs

    by Hao Ge, Junda Feng, Qi Huang, Fangcheng Fu, Xiaonan Nie, Lei Zuo, Haibin Lin, Bin Cui, Xin Liu


  • Open Access

    From ATOP to ZCube: Automated Topology Optimization Pipeline and A Highly Cost-Effective Network Topology for Large Model Training

    by Zihan Yan, Dan Li, Li Chen, Dian Xiong, Kaihui Gao, Yiwei Zhang, Rui Yan, Menglei Zhang, Bochun Zhang, Zhuo Jiang, et al.