Author name cluster

Xupeng Miao

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

8 papers

2 author rows

ICML Conference 2025 Conference Paper

Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs

Youhe Jiang
Fangcheng Fu
Xiaozhe Yao
Guoliang He
Xupeng Miao
Ana Klimovic
Bin Cui 0001
Binhang Yuan

Recent advancements in Large Language Models (LLMs) have led to increasingly diverse requests, accompanied with varying resource (compute and memory) demands to serve them. However, this in turn degrades the cost-efficiency of LLM serving as common practices primarily rely on homogeneous GPU resources. In response to this problem, this work conducts a thorough study about serving LLMs over heterogeneous GPU resources on cloud platforms. The rationale is that different GPU types exhibit distinct compute and memory characteristics, aligning well with the divergent resource demands of diverse requests. Particularly, through comprehensive benchmarking, we discover that the cost-efficiency of LLM serving can be substantially optimized by meticulously determining GPU composition, deployment configurations, and workload assignments. Subsequently, we design a scheduling algorithm via mixed-integer linear programming, aiming at deducing the most cost-efficient serving plan under the constraints of price budget and real-time GPU availability. Remarkably, our approach effectively outperforms homogeneous and heterogeneous baselines under a wide array of scenarios, covering diverse workload traces, varying GPU availablilities, and multi-model serving. This casts new light on more accessible and efficient LLM serving over heterogeneous cloud resources.

Details

ICLR Conference 2025 Conference Paper

NetMoE: Accelerating MoE Training through Dynamic Sample Placement

Xinyi Liu
Yujie Wang
Fangcheng Fu
Xupeng Miao
Shenhan Zhu
Xiaonan Nie
Bin Cui 0001

Mixture of Experts (MoE) is a widely used technique to expand model sizes for better model quality while maintaining the computation cost constant. In a nutshell, an MoE model consists of multiple experts in each model layer and routes the training tokens to only a fixed number of experts rather than all. In distributed training, as experts are distributed among different GPUs, All-to-All communication is necessary to exchange the training tokens among the GPUs after each time of expert routing. Due to the frequent and voluminous data exchanges, All-to-All communication has become a notable challenge to training efficiency. In this paper, we manage to accelerate All-to-All communication in MoE models from the training sample perspective, which is unexplored so far. In particular, we put forward the observation that tokens in the same training sample have certain levels of locality in expert routing. Motivated by this, we develop NetMoE, which takes such locality into account and dynamically rearranges the placement of training samples to minimize All-to-All communication costs. Specifically, we model the All-to-All communication given the sample placement and formulate an integer programming problem to deduce the optimal placement in polynomial time. Experiments with 32 GPUs show that NetMoE achieves a maximum efficiency improvement of $1.67 \times$ compared with current MoE training frameworks.

Details

AAAI Conference 2024 Conference Paper

Accelerating Text-to-Image Editing via Cache-Enabled Sparse Diffusion Inference

Zihao Yu
Haoyang Li
Fangcheng Fu
Xupeng Miao
Bin Cui

Due to the recent success of diffusion models, text-to-image generation is becoming increasingly popular and achieves a wide range of applications. Among them, text-to-image editing, or continuous text-to-image generation, attracts lots of attention and can potentially improve the quality of generated images. It's common to see that users may want to slightly edit the generated image by making minor modifications to their input textual descriptions for several rounds of diffusion inference. However, such an image editing process suffers from the low inference efficiency of many existing diffusion models even using GPU accelerators. To solve this problem, we introduce Fast Image Semantically Edit (FISEdit), a cached-enabled sparse diffusion model inference engine for efficient text-to-image editing. The key intuition behind our approach is to utilize the semantic mapping between the minor modifications on the input text and the affected regions on the output image. For each text editing step, FISEdit can 1) automatically identify the affected image regions and 2) utilize the cached unchanged regions' feature map to accelerate the inference process. For the former, we measure the differences between cached and ad hoc feature maps given the modified textual description, extract the region with significant differences, and capture the affected region by masks. For the latter, we develop an efficient sparse diffusion inference engine that only computes the feature maps for the affected region while reusing the cached statistics for the rest of the image. Finally, extensive empirical results show that FISEdit can be 3.4 times and 4.4 times faster than existing methods on NVIDIA TITAN RTX and A100 GPUs respectively, and even generates more satisfactory images.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

LSH-MoE: Communication-efficient MoE Training via Locality-Sensitive Hashing

Xiaonan Nie
Qibin Liu
Fangcheng Fu
Shenhan Zhu
Xupeng Miao
Xiaoyang Li
Yang Zhang
Shouda Liu

Larger transformer models perform better on various downstream tasks but require more cost to scale up the model size. To efficiently enlarge models, the Mixture-of-Expert (MoE) architecture is widely adopted, which consists of a gate network and a series of experts and keep the training cost constant by routing the input data to a fixed number of experts instead of all. In existing large-scale MoE training systems, experts would be distributed among different GPUs for parallelization, and thus input data requires additional all-to-all communication to access the target expert and conduct corresponding computation. However, upon evaluating the training process of three mainstream MoE models on commonly used GPU clusters, we found that the all-to-all communication ratio averaged around 45\%, which significantly hinders the training efficiency and scalability of MoE models. In this paper, we propose LSH-MoE, a communication-efficient MoE training framework using locality-sensitive hashing (LSH). We first present the problems of scaling MoE training in existing systems and highlight the potential of exploiting token similarity to facilitate data compression. Then, we introduce an efficient LSH-based compression technique, which utilizes the cross-polytope hashing for rapid clustering and implements a residual-based error compensation scheme to alleviate the adverse impact of compression. To verify the effectiveness of our methods, we conduct experiments on both language models (e. g. , RoBERTa, GPT, and T5) and vision models (e. g. , Swin) for both pre-training and fine-tuning tasks. The results demonstrate that our method substantially outperforms its counterparts across different tasks by 1. 28-2. 2$\times$ of speedup.

PDF Details DOI

IJCAI Conference 2024 Conference Paper

X-former Elucidator: Reviving Efficient Attention for Long Context Language Modeling

Xupeng Miao
Shenhan Zhu
Fangcheng Fu
Ziyu Guo
Zhi Yang
Yaofeng Tu
Zhihao Jia
Bin Cui

Transformer-based LLMs are becoming increasingly important in various AI applications. However, apart from the success of LLMs, the explosive demand of long context handling capabilities is a key and in-time problem for both academia and industry. Due to the limitations from the quadratic complexity of the attention mechanism, long context scenarios require much more resources for LLM development and deployment, bringing huge challenges to the underlying AI infrastructure. Meanwhile, we observe that there is a trend of reviving previous efficient attention mechanisms to latest LLMs. However, it still remains an open question about how to select from these diverse approaches in practice. In this paper, we answer this question from several aspects. First, we revisit these latest long-context LLM innovations and discuss their relationship with prior approaches with a novel and comprehensive taxonomy. Next, we conduct a thorough evaluation over various types of workloads considering both efficiency and effectiveness. Finally, we provide an in-depth analysis, summarize our key findings, and offer insightful suggestions on the trade-offs of designing and deploying efficient attention mechanisms for Transformer-based LLMs.

PDF Details DOI

AAAI Conference 2023 Conference Paper

CALIP: Zero-Shot Enhancement of CLIP with Parameter-Free Attention

Ziyu Guo
Renrui Zhang
Longtian Qiu
Xianzheng Ma
Xupeng Miao
Xuming He
Bin Cui

Contrastive Language-Image Pre-training (CLIP) has been shown to learn visual representations with promising zero-shot performance. To further improve its downstream accuracy, existing works propose additional learnable modules upon CLIP and fine-tune them by few-shot training sets. However, the resulting extra training cost and data requirement severely hinder the efficiency for model deployment and knowledge transfer. In this paper, we introduce a free-lunch enhancement method, CALIP, to boost CLIP's zero-shot performance via a parameter-free attention module. Specifically, we guide visual and textual representations to interact with each other and explore cross-modal informative features via attention. As the pre-training has largely reduced the embedding distances between two modalities, we discard all learnable parameters in the attention and bidirectionally update the multi-modal features, enabling the whole process to be parameter-free and training-free. In this way, the images are blended with textual-aware signals and the text representations become visual-guided for better adaptive zero-shot alignment. We evaluate CALIP on various benchmarks of 14 datasets for both 2D image and 3D point cloud few-shot classification, showing consistent zero-shot performance improvement over CLIP. Based on that, we further insert a small number of linear layers in CALIP's attention module and verify our robustness under the few-shot settings, which also achieves leading performance compared to existing methods. Those extensive experiments demonstrate the superiority of our approach for efficient enhancement of CLIP. Code is available at https://github.com/ZiyuGuo99/CALIP.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

Model-enhanced Vector Index

Hailin Zhang
Yujing Wang
Qi Chen
Ruiheng Chang
Ting Zhang
Ziming Miao
Yingyan Hou
Yang Ding

Embedding-based retrieval methods construct vector indices to search for document representations that are most similar to the query representations. They are widely used in document retrieval due to low latency and decent recall performance. Recent research indicates that deep retrieval solutions offer better model quality, but are hindered by unacceptable serving latency and the inability to support document updates. In this paper, we aim to enhance the vector index with end-to-end deep generative models, leveraging the differentiable advantages of deep retrieval models while maintaining desirable serving efficiency. We propose Model-enhanced Vector Index (MEVI), a differentiable model-enhanced index empowered by a twin-tower representation model. MEVI leverages a Residual Quantization (RQ) codebook to bridge the sequence-to-sequence deep retrieval and embedding-based models. To substantially reduce the inference time, instead of decoding the unique document ids in long sequential steps, we first generate some semantic virtual cluster ids of candidate documents in a small number of steps, and then leverage the well-adapted embedding vectors to further perform a fine-grained search for the relevant documents in the candidate virtual clusters. We empirically show that our model achieves better performance on the commonly used academic benchmarks MSMARCO Passage and Natural Questions, with comparable serving latency to dense retrieval solutions.

PDF Details

IJCAI Conference 2023 Conference Paper

OSDP: Optimal Sharded Data Parallel for Distributed Deep Learning

Youhe Jiang
Fangcheng Fu
Xupeng Miao
Xiaonan Nie
Bin Cui

Large-scale deep learning models contribute to significant performance improvements on varieties of downstream tasks. Current data and model parallelism approaches utilize model replication and partition techniques to support the distributed training of ultra-large models. However, directly deploying these systems often leads to sub-optimal training efficiency due to the complex model architectures and the strict device memory constraints. In this paper, we propose Optimal Sharded Data Parallel (OSDP), an automated parallel training system that combines the advantages from both data and model parallelism. Given the model description and the device information, OSDP makes trade-offs between the memory consumption and the hardware utilization, thus automatically generates the distributed computation graph and maximizes the overall system throughput. In addition, OSDP introduces operator splitting to further alleviate peak memory footprints during training with negligible overheads, which enables the trainability of larger models as well as the higher throughput. Extensive experimental results of OSDP on multiple different kinds of large-scale models demonstrate that the proposed strategy outperforms the state-of-the-art in multiple regards.

PDF Details DOI