Arrow Research search

Author name cluster

Yanzhi Wang

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.

45 papers
2 author rows

Possible papers

45

AAMAS Conference 2026 Conference Paper

Structured Agent Distillation for Large Language Model Agents

  • Jun Liu
  • Zhenglun Kong
  • Peiyan Dong
  • Changdi Yang
  • Tianqin Li
  • Yanyue Xie
  • Yifan Gong
  • Xuan Shen

Largelanguagemodels(LLMs)exhibitstrongcapabilitiesasdecisionmaking agents by interleaving reasoning and actions, as seen in ReAct-style frameworks. Yet, their practical deployment is constrained by high inference costs and large model sizes. We propose Structured Agent Distillation, the first framework to distill a ReAct-based LLM agent into a smaller model while preserving both reasoning fidelity and action consistency. Our method introduces a structured, span-level distillation strategy that explicitly segments trajectories into reasoning and action spans, enabling fine-grained alignment beyond standard token-level imitation. Unlike other advanced distillation methods, Our method segments trajectories into [REASON] and [ACT] spans, applying segmentspecific losses to align each component with the teacher’s behavior. This structure-aware supervision enables compact agents to better Corresponding Authors: Pu Zhao, Hao Tang. This work is licensed under a Creative Commons Attribution International 4. 0 License. Proc. of the 25th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2026), C. Amato, L. Dennis, V. Mascardi, J. Thangarajah (eds.), May 25 – 29, 2026, Paphos, Cyprus. © 2026 International Foundation for Autonomous Agents and Multiagent Systems (www. ifaamas. org). https: //doi. org/10. 65109/OLHJ8062 replicate the teacher’s decision process. Experiments on ALFWorld, HotPotQA-ReAct, and WebShop show that our approach consistently outperforms token-level and imitation learning baselines, achieving significant compression with minimal performance drop. Scaling and ablation results further highlight the importance of span-level alignment for efficient and deployable agents. We will release code upon acceptance.

PDF

NeurIPS Conference 2025 Conference Paper

ALTER: All-in-One Layer Pruning and Temporal Expert Routing for Efficient Diffusion Generation

  • Xiaomeng Yang
  • Lei Lu
  • Qihui Fan
  • Changdi Yang
  • Juyi Lin
  • Yanzhi Wang
  • Xuan Zhang
  • Shangqian Gao

Diffusion models have demonstrated exceptional capabilities in generating high-fidelity images. However, their iterative denoising process results in significant computational overhead during inference, limiting their practical deployment in resource-constrained environments. Existing acceleration methods often adopt uniform strategies that fail to capture the temporal variations during diffusion generation, while the commonly adopted sequential $\textit{pruning-then-fine-tuning strategy}$ suffers from sub-optimality due to the misalignment between pruning decisions made on pretrained weights and the model’s final parameters. To address these limitations, we introduce $\textbf{ALTER}$: $\textbf{A}$ll-in-One $\textbf{L}$ayer Pruning and $\textbf{T}$emporal $\textbf{E}$xpoert $\textbf{R}$outing, a unified framework that transforms diffusion models into a mixture of efficient temporal experts. ALTER achieves a single-stage optimization that unifies layer pruning, expert routing, and model fine-tuning by employing a trainable hypernetwork, which dynamically generates layer pruning decisions and manages timestep routing to specialized, pruned expert sub-networks throughout the ongoing fine-tuning of the UNet. This unified co-optimization strategy enables significant efficiency gains while preserving high generative quality. Specifically, ALTER achieves same-level visual fidelity to the original 50-step Stable Diffusion v2. 1 model while utilizing only 25. 9\% of its total MACs with just 20 inference steps and delivering a 3. 64$\times$ speedup through 35\% sparsity.

PDF Details

EAAI Journal 2025 Journal Article

Dynamic Gaussian Graph Operator: Learning parametric partial differential equations in arbitrary discrete mechanics problems

  • Chu Wang
  • Jinhong Wu
  • Yanzhi Wang
  • Zhijian Zha
  • Qi Zhou

Deep learning methods have access to be employed for solving physical systems governed by parametric partial differential equations (PDEs) due to massive scientific data. It has been refined to operator learning that focuses on learning non-linear mapping between infinite-dimensional function spaces, offering interface from observations to solutions. However, state-of-the-art neural operators are limited to constant and uniform discretization, thereby leading to deficiency in generalization on arbitrary discretization schemes for computational domain. In this work, we propose a novel operator learning algorithm, referred to as Dynamic Gaussian Graph Operator (DGGO) that expands neural operators to learning parametric PDEs in arbitrary discrete mechanics problems. The Dynamic Gaussian Graph (DGG) kernel learns to map the observation vectors defined in general Euclidean space to metric vectors defined in high-dimensional uniform metric space. The DGG integral kernel is parameterized by Gaussian kernel weighted Riemann sum approximating and using dynamic message passing graph to depict the interrelation within the integral term. Fourier Neural Operator is selected to localize the metric vectors on spatial and frequency domains. Metric vectors are regarded as located on latent uniform domain, wherein spatial and spectral transformation offer highly regular constraints on solution space. The efficiency and robustness of DGGO are validated by applying it to solve numerical arbitrary discrete mechanics problems in comparison with mainstream neural operators. Ablation experiments are implemented to demonstrate the effectiveness of spatial transformation in the DGG kernel. The proposed method is utilized to forecast stress field of hyper-elastic material with geometrically variable void as engineering application.

Details DOI

IJCAI Conference 2025 Conference Paper

FairSMOE: Mitigating Multi-Attribute Fairness Problem with Sparse Mixture-of-Experts

  • Changdi Yang
  • Zheng Zhan
  • Ci Zhang
  • Yifan Gong
  • Yize Li
  • Zichong Meng
  • Jun Liu
  • Xuan Shen

Real‐world datasets usually contain multiple attributes, making it essential to ensure fairness across all of them simultaneously. However, different attributes may vary in difficulty, and no existing approaches have effectively addressed this issue. Consequently, an attribute‐adaptive strategy is needed to achieve fairness for all attributes. Multi‐task Learning (MTL) leverages shared information to optimize multiple tasks concurrently, while Sparsely‐Gated Mixture‐of‐Experts (SMoE) can dynamically allocate computational resources to the most needed tasks. In this work, we formulate multi‐attribute fairness issue as an MTL problem and employ SMoE to achieve desirable performance across all attributes simultaneously. We first analyze the feasibility and find the potentiality by formalizing multi-attribute fairness problem into a MTL problem and mitigating it by using SMoE. However, vanilla SMoE could lead to over-utilization problem which causes sub-optimal performance. We then proposed an innovative SMoE framework for multi-attribute fair image classification, which further improves multi-attribute fairness by redesigning the MoE layer and routing policy with fairness consideration. Extensive experiments demonstrated the effectiveness. Taking a DeiT-Small as the backbone, we achieve 77. 25% and 86. 01% accuracy on the ISIC2019 and CelebA dataset respectively with Multi-attribute Predictive Quality Disparity (PQD) score of 0. 801 and 0. 787, beating current state-of-the-art methods Muffin, InfoFair and MultiFair.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Harmony in Divergence: Towards Fast, Accurate, and Memory-efficient Zeroth-order LLM Fine-tuning

  • Qitao Tan
  • Jun Liu
  • Zheng Zhan
  • Caiwen Ding
  • Yanzhi Wang
  • Xiaolong Ma
  • Jaewoo Lee
  • Jin Lu

Large language models (LLMs) excel across various tasks, but standard first-order (FO) fine-tuning demands considerable memory, significantly limiting real-world deployment. Recently, zeroth-order (ZO) optimization stood out as a promising memory-efficient training paradigm, avoiding backward passes and relying solely on forward passes for gradient estimation, making it attractive for resource-constrained scenarios. However, ZO method lags far behind FO method in both convergence speed and accuracy. To bridge the gap, we introduce a novel layer-wise divergence analysis that uncovers the distinct update pattern of FO and ZO optimization. Aiming to resemble the learning capacity of FO method from the findings, we propose \textbf{Di}vergence-driven \textbf{Z}eroth-\textbf{O}rder (\textbf{DiZO}) optimization. DiZO conducts divergence-driven layer adaptation by incorporating projections to ZO updates, generating diverse-magnitude updates precisely scaled to layer-wise individual optimization needs. Our results demonstrate that DiZO significantly reduces the needed iterations for convergence without sacrificing throughput, cutting training GPU hours by up to 48\% on various datasets. Moreover, DiZO consistently outperforms the representative ZO baselines in fine-tuning RoBERTa-large, OPT-series, and Llama-series on downstream tasks and, in some cases, even surpasses memory-intensive FO fine-tuning. Our code is released at \url{https: //github. com/Skilteee/DiZO}.

PDF Details

AAAI Conference 2025 Conference Paper

LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers

  • Xuan Shen
  • Zhao Song
  • Yufa Zhou
  • Bo Chen
  • Yanyu Li
  • Yifan Gong
  • Kai Zhang
  • Hao Tan

Diffusion Transformers have emerged as the preeminent models for a wide array of generative tasks, demonstrating superior performance and efficacy across various applications. The promising results come at the cost of slow inference, as each denoising step requires running the whole transformer model with a large amount of parameters. In this paper, we show that performing the full computation of the model at each diffusion step is unnecessary, as some computations can be skipped by lazily reusing the results of previous steps. Furthermore, we show that the lower bound of similarity between outputs at consecutive steps is notably high, and this similarity can be linearly approximated using the inputs. To verify our demonstrations, we propose the **LazyDiT**, a lazy learning framework that efficiently leverages cached results from earlier steps to skip redundant computations. Specifically, we incorporate lazy learning layers into the model, effectively trained to maximize laziness, enabling dynamic skipping of redundant computations. Experimental results show that LazyDiT outperforms the DDIM sampler across multiple diffusion transformer models at various resolutions. Furthermore, we implement our method on mobile devices, achieving better performance than DDIM with similar latency.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Numerical Pruning for Efficient Autoregressive Models

  • Xuan Shen
  • Zhao Song
  • Yufa Zhou
  • Bo Chen
  • Jing Liu
  • Ruiyi Zhang
  • Ryan A. Rossi
  • Hao Tan

Transformers have emerged as the leading architecture in deep learning, proving to be versatile and highly effective across diverse domains beyond language and image processing. However, their impressive performance often incurs high computational costs due to their substantial model size. This paper focuses on compressing decoder-only transformer-based autoregressive models through structural weight pruning to improve the model efficiency while preserving performance for both language and image generation tasks. Specifically, we propose a training-free pruning method that calculates a numerical score with Newton's method for the Attention and MLP modules, respectively. Besides, we further propose another compensation algorithm to recover the pruned model for better performance. To verify the effectiveness of our method, we provide both theoretical support and extensive experiments. Our experiments show that our method achieves state-of-the-art performance with reduced memory usage and faster generation speeds on GPUs.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

Routing Mamba: Scaling State Space Models with Mixture-of-Experts Projection

  • Zheng Zhan
  • Liliang Ren
  • Shuohang Wang
  • Liyuan Liu
  • Yang Liu
  • Yeyun Gong
  • Yanzhi Wang
  • Yelong Shen

State Space Models (SSMs) offer remarkable performance gains in efficient sequence modeling, with constant per-step inference-time computation and memory complexity. Recent advances, such as Mamba, further enhance SSMs with input-dependent gating and hardware-aware implementations, positioning them as strong alternatives to Transformers for long sequence modeling. However, efficiently scaling the expressive power of SSMs, particularly with Mixture of Experts (MoE), remains challenging, as naive integration attempts often falter or degrade performance. In this work, we introduce Routing Mamba (RoM), a novel approach that scales SSM parameters using sparse mixtures of linear projection experts. By sharing routing decisions between projection layers and lightweight sub-modules within Mamba across experts, RoM leverages synergies among linear projection experts for effective and efficient sparse scaling of Mamba layers. At a scale of 1. 3B active parameters (10B total) and 16K training sequence length, RoM achieves language modeling performance equivalent to a dense Mamba model requiring over 2. 3$\times$ more active parameters, and demonstrates consistent perplexity across context lengths. Experimental results further show RoM effectively scales hybrid language models, yielding a 23% FLOPS saving compared to dense Mamba scaling for similar performance. We release our training codebase at https: //github. com/zhanzheng8585/Routing-Mamba.

PDF Details

AAAI Conference 2025 Conference Paper

Toward Adaptive Large Language Models Structured Pruning via Hybrid-grained Weight Importance Assessment

  • Jun Liu
  • Zhenglun Kong
  • Pu Zhao
  • Changdi Yang
  • Xuan Shen
  • Hao Tang
  • Geng Yuan
  • Wei Niu

Structured pruning for large language models (LLMs) has garnered significant academic interest due to its ability to efficiently compress and accelerate LLMs by eliminating redundant weight groups at a coarse-grained granularity. Current structured pruning methods for LLMs typically depend on a singular granularity for assessing weight importance, resulting in notable performance degradation in downstream tasks. Intriguingly, our empirical investigations reveal that utilizing unstructured pruning, which achieves better performance retention by pruning weights at a finer granularity, \emph{i.e.}, individual weights, yields significantly varied sparse LLM structures when juxtaposed to structured pruning. This suggests that evaluating both holistic and individual assessments for weight importance are essential for LLM pruning. Building on this insight, we introduce the Hybrid-grained Weight Importance Assessment (HyWIA), a novel method that merges fine-grained and coarse-grained evaluations of weight importance for the pruning of LLMs. Leveraging an attention mechanism, HyWIA adaptively determines the optimal blend of granularity in weight importance assessments in an end-to-end pruning manner. Extensive experiments on LLaMA-V1/V2, Vicuna, Baichuan, and Bloom across various benchmarks demonstrate the effectiveness of HyWIA in pruning LLMs. For example, HyWIA surpasses the cutting-edge LLM-Pruner by an average margin of 2.82% in accuracy across seven downstream tasks when pruning LLaMA-7B by 50%.

PDF Details DOI

AAAI Conference 2024 Conference Paper

Agile-Quant: Activation-Guided Quantization for Faster Inference of LLMs on the Edge

  • Xuan Shen
  • Peiyan Dong
  • Lei Lu
  • Zhenglun Kong
  • Zhengang Li
  • Ming Lin
  • Chao Wu
  • Yanzhi Wang

Large Language Models (LLMs) stand out for their impressive performance in intricate language modeling tasks. However, their demanding computational and memory needs pose obstacles for broad use on edge devices. Quantization is then introduced to boost LLMs' on-device efficiency. Recent works show that 8-bit or lower weight quantization is feasible with minimal impact on end-to-end task performance, while the activation is still not quantized. On the other hand, mainstream commodity edge devices still struggle to execute these sub-8-bit quantized networks effectively. In this paper, we propose Agile-Quant, an Activation-Guided quantization framework for faster Inference of popular Large Language Models (LLMs) on the Edge. Considering the hardware profiling and activation analysis, we first introduce a basic activation quantization strategy to balance the trade-off of task performance and real inference speed. Then we leverage the activation-aware token pruning technique to reduce the outliers and the adverse impact on attentivity. Ultimately, we utilize the SIMD-based 4-bit multiplier and our efficient TRIP matrix multiplication to implement the accelerator for LLMs on the edge. We apply our framework on different scales of LLMs including LLaMA, OPT, and BLOOM with 4-bit or 8-bit for the activation and 4-bit for the weight quantization. Experiments show that Agile-Quant achieves simultaneous quantization of model weights and activations while maintaining task performance comparable to existing weight-only quantization methods. Moreover, in the 8- and 4-bit scenario, Agile-Quant achieves an on-device speedup of up to 2.55x compared to its FP16 counterparts across multiple edge devices, marking a pioneering advancement in this domain.

PDF Details DOI

IJCAI Conference 2024 Conference Paper

Digital Avatars: Framework Development and Their Evaluation

  • Timothy Rupprecht
  • Sung-En Chang
  • Yushu Wu
  • Lei Lu
  • Enfu Nan
  • Chih-hsiang Li
  • Caiyue Lai
  • Zhimin Li

We present a novel prompting strategy for artificial intelligence driven digital avatars. To better quantify how our prompting strategy affects anthropomorphic features like humor, authenticity, and favorability we present Crowd Vote - an adaptation of Crowd Score that allows for judges to elect a large language model (LLM) candidate over competitors answering the same or similar prompts. To visualize the responses of our LLM, and the effectiveness of our prompting strategy we propose an end-to-end framework for creating high-fidelity artificial intelligence (AI) driven digital avatars. This pipeline effectively captures an individual's essence for interaction and our streaming algorithm delivers a high-quality digital avatar with real-time audio-video streaming from server to mobile device. Both our visualization tool, and our Crowd Vote metrics demonstrate our AI driven digital avatars have state-of-the-art humor, authenticity, and favorability outperforming all competitors and baselines. In the case of our Donald Trump and Joe Biden avatars, their authenticity and favorability are rated higher than even their real-world equivalents.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Exploring Token Pruning in Vision State Space Models

  • Zheng Zhan
  • Zhenglun Kong
  • Yifan Gong
  • Yushu Wu
  • Zichong Meng
  • Hangyu Zheng
  • Xuan Shen
  • Stratis Ioannidis

State Space Models (SSMs) have the advantage of keeping linear computational complexity compared to attention modules in transformers, and have been applied to vision tasks as a new type of powerful vision foundation model. Inspired by the observations that the final prediction in vision transformers (ViTs) is only based on a subset of most informative tokens, we take the novel step of enhancing the efficiency of SSM-based vision models through token-based pruning. However, direct applications of existing token pruning techniques designed for ViTs fail to deliver good performance, even with extensive fine-tuning. To address this issue, we revisit the unique computational characteristics of SSMs and discover that naive application disrupts the sequential token positions. This insight motivates us to design a novel and general token pruning method specifically for SSM-based vision models. We first introduce a pruning-aware hidden state alignment method to stabilize the neighborhood of remaining tokens for performance enhancement. Besides, based on our detailed analysis, we propose a token importance evaluation method adapted for SSM models, to guide the token pruning. With efficient implementation and practical acceleration methods, our method brings actual speedup. Extensive experiments demonstrate that our approach can achieve significant computation reduction with minimal impact on performance across different tasks. Notably, we achieve 81. 7\% accuracy on ImageNet with a 41. 6\% reduction in the FLOPs for pruned PlainMamba-L3. Furthermore, our work provides deeper insights into understanding the behavior of SSM-based vision models for future research.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Fast and Memory-Efficient Video Diffusion Using Streamlined Inference

  • Zheng Zhan
  • Yushu Wu
  • Yifan Gong
  • Zichong Meng
  • Zhenglun Kong
  • Changdi Yang
  • Geng Yuan
  • Pu Zhao

The rapid progress in artificial intelligence-generated content (AIGC), especially with diffusion models, has significantly advanced development of high-quality video generation. However, current video diffusion models exhibit demanding computational requirements and high peak memory usage, especially for generating longer and higher-resolution videos. These limitations greatly hinder the practical application of video diffusion models on standard hardware platforms. To tackle this issue, we present a novel, training-free framework named Streamlined Inference, which leverages the temporal and spatial properties of video diffusion models. Our approach integrates three core components: Feature Slicer, Operator Grouping, and Step Rehash. Specifically, Feature Slicer effectively partitions input features into sub-features and Operator Grouping processes each sub-feature with a group of consecutive operators, resulting in significant memory reduction without sacrificing the quality or speed. Step Rehash further exploits the similarity between adjacent steps in diffusion, and accelerates inference through skipping unnecessary steps. Extensive experiments demonstrate that our approach significantly reduces peak memory and computational overhead, making it feasible to generate high-quality videos on a single consumer GPU (e. g. , reducing peak memory of Animatediff from 42GB to 11GB, featuring faster inference on 2080Ti).

PDF Details DOI

IJCAI Conference 2024 Conference Paper

FasterVD: On Acceleration of Video Diffusion Models

  • Pinrui Yu
  • Dan Luo
  • Timothy Rupprecht
  • Lei Lu
  • Zhenglun Kong
  • Pu Zhao
  • Yanyu Li
  • Octavia Camps

Equipped with Denoising Diffusion Probabilistic Models, video content generation has gained significant research interest recently. However, diffusion pipelines call for intensive computation and model storage, which poses challenges for their wide and efficient deployment. In this work, we address this issue by integrating LCM-LoRA to reduce the denoising steps and escalating the video generation process by frame skipping and interpolation. Our framework achieves an approximately 10× inference acceleration for high-quality realistic video generation on commonly available GPUs.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

Search for Efficient Large Language Models

  • Xuan Shen
  • Pu Zhao
  • Yifan Gong
  • Zhenglun Kong
  • Zheng Zhan
  • Yushu Wu
  • Ming Lin
  • Chao Wu

Large Language Models (LLMs) have long held sway in the realms of artificial intelligence research. Numerous efficient techniques, including weight pruning, quantization, and distillation, have been embraced to compress LLMs, targeting memory reduction and inference acceleration, which underscore the redundancy in LLMs. However, most model compression techniques concentrate on weight optimization, overlooking the exploration of optimal architectures. Besides, traditional architecture search methods, limited by the elevated complexity with extensive parameters, struggle to demonstrate their effectiveness on LLMs. In this paper, we propose a training-free architecture search framework to identify optimal subnets that preserve the fundamental strengths of the original LLMs while achieving inference acceleration. Furthermore, after generating subnets that inherit specific weights from the original LLMs, we introduce a reformation algorithm that utilizes the omitted weights to rectify the inherited weights with a small amount of calibration data. Compared with SOTA training-free structured pruning works that can generate smaller networks, our method demonstrates superior performance across standard benchmarks. Furthermore, our generated subnets can directly reduce the usage of GPU memory and achieve inference acceleration.

PDF Details DOI

EAAI Journal 2023 Journal Article

A structurally re-parameterized convolution neural network-based method for gearbox fault diagnosis in edge computing scenarios

  • Yanzhi Wang
  • Jinhong Wu
  • Ziyang Yu
  • Jiexiang Hu
  • Qi Zhou

Gearboxes operate in harsh environments. Cloud-based techniques have been previously adopted for fault diagnosis in Gearboxes. Cloud-based fault diagnosis methods are prone to time delays and loss of information. Therefore, edge computing-based fault diagnosis becomes an option. However, with limited hardware resources for edge devices, balancing the diagnostic capabilities of the model with operating performance becomes a challenge. This paper proposes a lightweight convolutional neural network for gearbox fault diagnosis in edge computing scenarios to achieve an accurate diagnosis and lightweight deployment of models. By constructing the Mel-Frequency Cepstral Coefficients (MFCC) feature matrix of input data, the methodology can suppress noise interference and improve diagnostic accuracy. Based on the structural re-parameterization, the model structure transforms from multiple branches at training time to a single branch at inference time. This improves the inference speed of the model and reduces the hardware cost when the model is deployed while ensuring that the diagnostic capability of the model remains unchanged. Validation experiments were conducted on a public dataset and a custom experimental device, using the NVIDIA Jetson Xavier NX kit as the edge computing platform. According to the experiment result, after extracting the MFCC feature matrix, the average diagnostic accuracy rate in the noisy environment of the presented methodology is improved by 12. 22% and 9. 44%, respectively. After structural re-parameterization, the Memory of the model decreases by 52. 58%, and the inference speed is increased by 38. 83%.

Details DOI

IJCAI Conference 2023 Conference Paper

Data Level Lottery Ticket Hypothesis for Vision Transformers

  • Xuan Shen
  • Zhenglun Kong
  • Minghai Qin
  • Peiyan Dong
  • Geng Yuan
  • Xin Meng
  • Hao Tang
  • Xiaolong Ma

The conventional lottery ticket hypothesis (LTH) claims that there exists a sparse subnetwork within a dense neural network and a proper random initialization method, called the winning ticket, such that it can be trained from scratch to almost as good as the dense counterpart. Meanwhile, the research of LTH in vision transformers (ViTs) is scarcely evaluated. In this paper, we first show that the conventional winning ticket is hard to find at weight level of ViTs by existing methods. Then, we generalize the LTH for ViTs to input data consisting of image patches inspired by the input dependence of ViTs. That is, there exists a subset of input image patches such that a ViT can be trained from scratch by using only this subset of patches and achieve similar accuracy to the ViTs trained by using all image patches. We call this subset of input patches the winning tickets, which represent a significant amount of information in the input data. We use a ticket selector to generate the winning tickets based on the informativeness of patches for various types of ViT, including DeiT, LV-ViT, and Swin Transformers. The experiments show that there is a clear difference between the performance of models trained with winning tickets and randomly selected subsets, which verifies our proposed theory. We elaborate the analogical similarity between our proposed Data-LTH-ViTs and the conventional LTH for further verifying the integrity of our theory. The Source codes are available at https: //github. com/shawnricecake/vit-lottery-ticket-input.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

HotBEV: Hardware-oriented Transformer-based Multi-View 3D Detector for BEV Perception

  • Peiyan Dong
  • Zhenglun Kong
  • Xin Meng
  • Pinrui Yu
  • Yifan Gong
  • Geng Yuan
  • Hao Tang
  • Yanzhi Wang

The bird's-eye-view (BEV) perception plays a critical role in autonomous driving systems, involving the accurate and efficient detection and tracking of objects from a top-down perspective. To achieve real-time decision-making in self-driving scenarios, low-latency computation is essential. While recent approaches to BEV detection have focused on improving detection precision using Lift-Splat-Shoot (LSS)-based or transformer-based schemas, the substantial computational and memory burden of these approaches increases the risk of system crashes when multiple on-vehicle tasks run simultaneously. Unfortunately, there is a dearth of literature on efficient BEV detector paradigms, let alone achieving realistic speedups. Unlike existing works that focus on reducing computation costs, this paper focuses on developing an efficient model design that prioritizes actual on-device latency. To achieve this goal, we propose a latency-aware design methodology that considers key hardware properties, such as memory access cost and degree of parallelism. Given the prevalence of GPUs as the main computation platform for autonomous driving systems, we develop a theoretical latency prediction model and introduce efficient building operators. By leveraging these operators and following an effective local-to-global visual modeling process, we propose a hardware-oriented backbone that is also optimized for strong feature capturing and fusing. Using these insights, we present a new hardware-oriented framework for efficient yet accurate camera-view BEV detectors. Experiments show that HotBEV achieves a 2\%$\sim$23\% NDS gain, and 2\%$\sim$7. 8\% mAP gain with a 1. 1$\times$$\sim$3. 4$\times$ speedups compared to existing works on V100; On multiple GPU devices such as GPU GTX 2080 and the low-end GTX 1080, HotBEV achieves 1. 1$\times$$\sim$6. 3$\times$ faster than others.

PDF Details

NeurIPS Conference 2023 Conference Paper

PackQViT: Faster Sub-8-bit Vision Transformers via Full and Packed Quantization on the Mobile

  • Peiyan Dong
  • Lei Lu
  • Chao Wu
  • Cheng Lyu
  • Geng Yuan
  • Hao Tang
  • Yanzhi Wang

While Vision Transformers (ViTs) have undoubtedly made impressive strides in computer vision (CV), their intricate network structures necessitate substantial computation and memory resources. A decision-making process for CV tasks typically entails performing computations with low latency, which is a tricky problem for ViT models. Model quantization is a widely-used technique to optimize the hardware efficiency of deep neural networks. Full quantization under Sub-8-bit precision, in particular, is a promising solution to reduce inference latency significantly. Unfortunately, current commodity hardware, such as CPUs and GPUs, still struggles to efficiently execute these sub-8-bit quantized networks, as their SIMD instructions only support a granularity of 8 bits or wider. Also, there is a scarcity of literature that presents a full quantization paradigm for ViTs. In this paper, we propose an activation-aware fully sub-8-bit quantization-aware training (QAT) framework called PackQViT for efficient yet accurate ViT acceleration on mobile devices to facilitate real-time AI-powered decision-making. Specifically, in revisiting data activation within the ViT dataflow, two characteristics are relevant to quantization strategy and precision: the long-tailed distribution and systematic channel-wise outliers. In response, we employ either log2 quantization or clipping to address the long-tailed distribution and incorporate outlier-aware training for residual link quantization to regulate the various channel-wise outliers more consistently. Notably, due to the systematic fixed pattern, outlier-aware training approach can predict the channel indices and regularized scales of outliers in advance, thus avoiding the runtime data-adaptive selection during inference. Furthermore, we employ Int-$2^{n}$-Softmax, Int-LayerNorm, and Integer GELU to enable integer-only computation flow. Finally, we develop a SIMD-based 4-bit packed multiplier to achieve end-to-end ViT acceleration on mobile phones. Compared to prior studies on ViT quantization using 8-bit precision, PackQViT surpasses other works by an improved accuracy ranging from 0. 4\% to 17. 9\% for various widely used ViTs on ImageNet dataset; under 4-bit precision, PackQViT demonstrates 0. 4%$\sim$2. 8% higher accuracy. Compared to the baseline multiplier, our implementations on the Realme GT Android smartphone with Snapdragon 870 SoC CPU achieve 2. 6x$\sim$3. 7x speedup under 8-bit scenario and 3. 8x$\sim$5. 9x speedup under 4-bit which ensures practical real-time performance.

PDF Details

AAAI Conference 2023 Conference Paper

Peeling the Onion: Hierarchical Reduction of Data Redundancy for Efficient Vision Transformer Training

  • Zhenglun Kong
  • Haoyu Ma
  • Geng Yuan
  • Mengshu Sun
  • Yanyue Xie
  • Peiyan Dong
  • Xin Meng
  • Xuan Shen

Vision transformers (ViTs) have recently obtained success in many applications, but their intensive computation and heavy memory usage at both training and inference time limit their generalization. Previous compression algorithms usually start from the pre-trained dense models and only focus on efficient inference, while time-consuming training is still unavoidable. In contrast, this paper points out that the million-scale training data is redundant, which is the fundamental reason for the tedious training. To address the issue, this paper aims to introduce sparsity into data and proposes an end-to-end efficient training framework from three sparse perspectives, dubbed Tri-Level E-ViT. Specifically, we leverage a hierarchical data redundancy reduction scheme, by exploring the sparsity under three levels: number of training examples in the dataset, number of patches (tokens) in each example, and number of connections between tokens that lie in attention weights. With extensive experiments, we demonstrate that our proposed technique can noticeably accelerate training for various ViT architectures while maintaining accuracy. Remarkably, under certain ratios, we are able to improve the ViT accuracy rather than compromising it. For example, we can achieve 15.2% speedup with 72.6% (+0.4) Top-1 accuracy on Deit-T, and 15.7% speedup with 79.9% (+0.1) Top-1 accuracy on Deit-S. This proves the existence of data redundancy in ViT. Our code is released at https://github.com/ZLKong/Tri-Level-ViT

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Seconds

  • Yanyu Li
  • Huan Wang
  • Qing Jin
  • Ju Hu
  • Pavlo Chemerys
  • Yun Fu
  • Yanzhi Wang
  • Sergey Tulyakov

Text-to-image diffusion models can create stunning images from natural language descriptions that rival the work of professional artists and photographers. However, these models are large, with complex network architectures and tens of denoising iterations, making them computationally expensive and slow to run. As a result, high-end GPUs and cloud-based inference are required to run diffusion models at scale. This is costly and has privacy implications, especially when user data is sent to a third party. To overcome these challenges, we present a generic approach that, for the first time, unlocks running text-to-image diffusion models on mobile devices in **less than 2 seconds**. We achieve so by introducing efficient network architecture and improving step distillation. Specifically, we propose an efficient UNet by identifying the redundancy of the original model and reducing the computation of the image decoder via data distillation. Further, we enhance the step distillation by exploring training strategies and introducing regularization from classifier-free guidance. Our extensive experiments on MS-COCO show that our model with $8$ denoising steps achieves better FID and CLIP scores than Stable Diffusion v$1. 5$ with $50$ steps. Our work democratizes content creation by bringing powerful text-to-image diffusion models to the hands of users.

PDF Details

AAAI Conference 2023 Conference Paper

Towards Real-Time Segmentation on the Edge

  • Yanyu Li
  • Changdi Yang
  • Pu Zhao
  • Geng Yuan
  • Wei Niu
  • Jiexiong Guan
  • Hao Tang
  • Minghai Qin

The research in real-time segmentation mainly focuses on desktop GPUs. However, autonomous driving and many other applications rely on real-time segmentation on the edge, and current arts are far from the goal. In addition, recent advances in vision transformers also inspire us to re-design the network architecture for dense prediction task. In this work, we propose to combine the self attention block with lightweight convolutions to form new building blocks, and employ latency constraints to search an efficient sub-network. We train an MLP latency model based on generated architecture configurations and their latency measured on mobile devices, so that we can predict the latency of subnets during search phase. To the best of our knowledge, we are the first to achieve over 74% mIoU on Cityscapes with semi-real-time inference (over 15 FPS) on mobile GPU from an off-the-shelf phone.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

Advancing Model Pruning via Bi-level Optimization

  • Yihua Zhang
  • Yuguang Yao
  • Parikshit Ram
  • Pu Zhao
  • Tianlong Chen
  • Mingyi Hong
  • Yanzhi Wang
  • Sijia Liu

The deployment constraints in practical applications necessitate the pruning of large-scale deep learning models, i. e. , promoting their weight sparsity. As illustrated by the Lottery Ticket Hypothesis (LTH), pruning also has the potential of improving their generalization ability. At the core of LTH, iterative magnitude pruning (IMP) is the predominant pruning method to successfully find ‘winning tickets’. Yet, the computation cost of IMP grows prohibitively as the targeted pruning ratio increases. To reduce the computation overhead, various efficient ‘one-shot’ pruning methods have been developed, but these schemes are usually unable to find winning tickets as good as IMP. This raises the question of how to close the gap between pruning accuracy and pruning efficiency? To tackle it, we pursue the algorithmic advancement of model pruning. Specifically, we formulate the pruning problem from a fresh and novel viewpoint, bi-level optimization (BLO). We show that the BLO interpretation provides a technically-grounded optimization base for an efficient implementation of the pruning-retraining learning paradigm used in IMP. We also show that the proposed bi-level optimization-oriented pruning method (termed BiP) is a special class of BLO problems with a bi-linear problem structure. By leveraging such bi-linearity, we theoretically show that BiP can be solved as easily as first-order optimization, thus inheriting the computation efficiency. Through extensive experiments on both structured and unstructured pruning with 5 model architectures and 4 data sets, we demonstrate that BiP can find better winning tickets than IMP in most cases, and is computationally as efficient as the one-shot pruning schemes, demonstrating $2-7\times$ speedup over IMP for the same level of model accuracy and sparsity.

PDF Details

NeurIPS Conference 2022 Conference Paper

EfficientFormer: Vision Transformers at MobileNet Speed

  • Yanyu Li
  • Geng Yuan
  • Yang Wen
  • Ju Hu
  • Georgios Evangelidis
  • Sergey Tulyakov
  • Yanzhi Wang
  • Jian Ren

Vision Transformers (ViT) have shown rapid progress in computer vision tasks, achieving promising results on various benchmarks. However, due to the massive number of parameters and model design, e. g. , attention mechanism, ViT-based models are generally times slower than lightweight convolutional networks. Therefore, the deployment of ViT for real-time applications is particularly challenging, especially on resource-constrained hardware such as mobile devices. Recent efforts try to reduce the computation complexity of ViT through network architecture search or hybrid design with MobileNet block, yet the inference speed is still unsatisfactory. This leads to an important question: can transformers run as fast as MobileNet while obtaining high performance? To answer this, we first revisit the network architecture and operators used in ViT-based models and identify inefficient designs. Then we introduce a dimension-consistent pure transformer (without MobileNet blocks) as a design paradigm. Finally, we perform latency-driven slimming to get a series of final models dubbed EfficientFormer. Extensive experiments show the superiority of EfficientFormer in performance and speed on mobile devices. Our fastest model, EfficientFormer-L1, achieves $79. 2\%$ top-1 accuracy on ImageNet-1K with only $1. 6$ ms inference latency on iPhone 12 (compiled with CoreML), which runs as fast as MobileNetV2$\times 1. 4$ ($1. 6$ ms, $74. 7\%$ top-1), and our largest model, EfficientFormer-L7, obtains $83. 3\%$ accuracy with only $7. 0$ ms latency. Our work proves that properly designed transformers can reach extremely low latency on mobile devices while maintaining high performance.

PDF Details

NeurIPS Conference 2022 Conference Paper

Layer Freezing & Data Sieving: Missing Pieces of a Generic Framework for Sparse Training

  • Geng Yuan
  • Yanyu Li
  • Sheng Li
  • Zhenglun Kong
  • Sergey Tulyakov
  • Xulong Tang
  • Yanzhi Wang
  • Jian Ren

Recently, sparse training has emerged as a promising paradigm for efficient deep learning on edge devices. The current research mainly devotes the efforts to reducing training costs by further increasing model sparsity. However, increasing sparsity is not always ideal since it will inevitably introduce severe accuracy degradation at an extremely high sparsity level. This paper intends to explore other possible directions to effectively and efficiently reduce sparse training costs while preserving accuracy. To this end, we investigate two techniques, namely, layer freezing and data sieving. First, the layer freezing approach has shown its success in dense model training and fine-tuning, yet it has never been adopted in the sparse training domain. Nevertheless, the unique characteristics of sparse training may hinder the incorporation of layer freezing techniques. Therefore, we analyze the feasibility and potentiality of using the layer freezing technique in sparse training and find it has the potential to save considerable training costs. Second, we propose a data sieving method for dataset-efficient training, which further reduces training costs by ensuring only a partial dataset is used throughout the entire training process. We show that both techniques can be well incorporated into the sparse training algorithm to form a generic framework, which we dub SpFDE. Our extensive experiments demonstrate that SpFDE can significantly reduce training costs while preserving accuracy from three dimensions: weight sparsity, layer freezing, and dataset sieving. Our code and models will be released.

PDF Details

IJCAI Conference 2022 Conference Paper

Pruning-as-Search: Efficient Neural Architecture Search via Channel Pruning and Structural Reparameterization

  • Yanyu Li
  • Pu Zhao
  • Geng Yuan
  • Xue Lin
  • Yanzhi Wang
  • Xin Chen

Neural architecture search (NAS) and network pruning are widely studied efficient AI techniques, but not yet perfect. NAS performs exhaustive candidate architecture search, incurring tremendous search cost. Though (structured) pruning can simply shrink model dimension, it remains unclear how to decide the per-layer sparsity automatically and optimally. In this work, we revisit the problem of layer-width optimization and propose Pruning-as-Search (PaS), an end-to-end channel pruning method to search out desired sub-network automatically and efficiently. Specifically, we add a depth-wise binary convolution to learn pruning policies directly through gradient descent. By combining the structural reparameterization and PaS, we successfully searched out a new family of VGG-like and lightweight networks, which enable the flexibility of arbitrary width with respect to each layer instead of each stage. Experimental results show that our proposed architecture outperforms prior arts by around 1. 0% top-1 accuracy under similar inference speed on ImageNet-1000 classification task. Furthermore, we demonstrate the effectiveness of our width search on complex tasks including instance segmentation and image translation. Code and models are released.

PDF Details DOI

IJCAI Conference 2022 Conference Paper

Real-Time Portrait Stylization on the Edge

  • Yanyu Li
  • Xuan Shen
  • Geng Yuan
  • Jiexiong Guan
  • Wei Niu
  • Hao Tang
  • Bin Ren
  • Yanzhi Wang

In this work we demonstrate real-time portrait stylization, specifically, translating self-portrait into cartoon or anime style on mobile devices. We propose a latency-driven differentiable architecture search method, maintaining realistic generative quality. With our framework, we obtain 10× computation reduction on the generative model and achieve real-time video stylization on off-the-shelf smartphone using mobile GPUs.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

SparCL: Sparse Continual Learning on the Edge

  • Zifeng Wang
  • Zheng Zhan
  • Yifan Gong
  • Geng Yuan
  • Wei Niu
  • Tong Jian
  • Bin Ren
  • Stratis Ioannidis

Existing work in continual learning (CL) focuses on mitigating catastrophic forgetting, i. e. , model performance deterioration on past tasks when learning a new task. However, the training efficiency of a CL system is under-investigated, which limits the real-world application of CL systems under resource-limited scenarios. In this work, we propose a novel framework called Sparse Continual Learning (SparCL), which is the first study that leverages sparsity to enable cost-effective continual learning on edge devices. SparCL achieves both training acceleration and accuracy preservation through the synergy of three aspects: weight sparsity, data efficiency, and gradient sparsity. Specifically, we propose task-aware dynamic masking (TDM) to learn a sparse network throughout the entire CL process, dynamic data removal (DDR) to remove less informative training data, and dynamic gradient masking (DGM) to sparsify the gradient updates. Each of them not only improves efficiency, but also further mitigates catastrophic forgetting. SparCL consistently improves the training efficiency of existing state-of-the-art (SOTA) CL methods by at most 23X less training FLOPs, and, surprisingly, further improves the SOTA accuracy by at most 1. 7%. SparCL also outperforms competitive baselines obtained from adapting SOTA sparse training methods to the CL setting in both efficiency and accuracy. We also evaluate the effectiveness of SparCL on a real mobile phone, further indicating the practical potential of our method.

PDF Details

AAAI Conference 2021 System Paper

A Compression-Compilation Co-Design Framework Towards Real-Time Object Detection on Mobile Devices

  • Yuxuan Cai
  • Geng Yuan
  • Hongjia Li
  • Wei Niu
  • Yanyu Li
  • Xulong Tang
  • Bin Ren
  • Yanzhi Wang

The rapid development and wide utilization of object detection techniques have aroused requirements for both accuracy and speed of object detectors. In this work, we propose a compression-compilation co-design framework to achieve real-time YOLOv4 inference on mobile devices. We propose a novel fine-grained structured pruning, which maintain high accuracy while achieving high hardware parallelism. Our pruned YOLOv4 achieves 48. 9 mAP and 17 FPS inference speed on an off-the-shelf Samsung Galaxy S20 smartphone, which is 5. 5× faster than the original state-of-the-art detector YOLOv4.

PDF Details

IJCAI Conference 2021 Conference Paper

A Compression-Compilation Framework for On-mobile Real-time BERT Applications

  • Wei Niu
  • Zhenglun Kong
  • Geng Yuan
  • Weiwen Jiang
  • Jiexiong Guan
  • Caiwen Ding
  • Pu Zhao
  • Sijia Liu

Transformer-based deep learning models have increasingly demonstrated high accuracy on many natural language processing (NLP) tasks. In this paper, we propose a compression-compilation co-design framework that can guarantee the identified model meets both resource and real-time specifications of mobile devices. Our framework applies a compiler-aware neural architecture optimization method (CANAO), which can generate the optimal compressed model that balances both accuracy and latency. We are able to achieve up to 7. 8x speedup compared with TensorFlow-Lite with only minor accuracy loss. We present two types of BERT applications on mobile devices: Question Answering (QA) and Text Generation. Both can be executed in real-time with latency as low as 45ms. Videos for demonstrating the framework can be found on https: //www. youtube. com/watch? v=_WIRvK_2PZI

PDF Details DOI

NeurIPS Conference 2021 Conference Paper

MEST: Accurate and Fast Memory-Economic Sparse Training Framework on the Edge

  • Geng Yuan
  • Xiaolong Ma
  • Wei Niu
  • Zhengang Li
  • Zhenglun Kong
  • Ning Liu
  • Yifan Gong
  • Zheng Zhan

Recently, a new trend of exploring sparsity for accelerating neural network training has emerged, embracing the paradigm of training on the edge. This paper proposes a novel Memory-Economic Sparse Training (MEST) framework targeting for accurate and fast execution on edge devices. The proposed MEST framework consists of enhancements by Elastic Mutation (EM) and Soft Memory Bound (&S) that ensure superior accuracy at high sparsity ratios. Different from the existing works for sparse training, this current work reveals the importance of sparsity schemes on the performance of sparse training in terms of accuracy as well as training speed on real edge devices. On top of that, the paper proposes to employ data efficiency for further acceleration of sparse training. Our results suggest that unforgettable examples can be identified in-situ even during the dynamic exploration of sparsity masks in the sparse training process, and therefore can be removed for further training speedup on edge devices. Comparing with state-of-the-art (SOTA) works on accuracy, our MEST increases Top-1 accuracy significantly on ImageNet when using the same unstructured sparsity scheme. Systematical evaluation on accuracy, training speed, and memory footprint are conducted, where the proposed MEST framework consistently outperforms representative SOTA works. A reviewer strongly against our work based on his false assumptions and misunderstandings. On top of the previous submission, we employ data efficiency for further acceleration of sparse training. And we explore the impact of model sparsity, sparsity schemes, and sparse training algorithms on the number of removable training examples. Our codes are publicly available at: https: //github. com/boone891214/MEST.

PDF Details

AAAI Conference 2021 Conference Paper

RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices

  • Wei Niu
  • Mengshu Sun
  • Zhengang Li
  • Jou-An Chen
  • Jiexiong Guan
  • Xipeng Shen
  • Yanzhi Wang
  • Sijia Liu

Mobile devices are becoming an important carrier for deep learning tasks, as they are being equipped with powerful, highend mobile CPUs and GPUs. However, it is still a challenging task to execute 3D Convolutional Neural Networks (CNNs) targeting for real-time performance, besides high inference accuracy. The reason is more complex model structure and higher model dimensionality overwhelm the available computation/storage resources on mobile devices. A natural way may be turning to deep learning weight pruning techniques. However, the direct generalization of existing 2D CNN weight pruning methods to 3D CNNs is not ideal for fully exploiting mobile parallelism while achieving high inference accuracy. This paper proposes RT3D, a model compression and mobile acceleration framework for 3D CNNs, seamlessly integrating neural network weight pruning and compiler code generation techniques. We propose and investigate two structured sparsity schemes i. e. , the vanilla structured sparsity and kernel group structured (KGS) sparsity that are mobile acceleration friendly. The vanilla sparsity removes whole kernel groups, while KGS sparsity is a more fine-grained structured sparsity that enjoys higher flexibility while exploiting full on-device parallelism. We propose a reweighted regularization pruning algorithm to achieve the proposed sparsity schemes. The inference time speedup due to sparsity is approaching the pruning rate of the whole model FLOPs (floating point operations). RT3D demonstrates up to 29. 1× speedup in end-to-end inference time comparing with current mobile frameworks supporting 3D CNNs, with moderate 1% ∼ 1. 5% accuracy loss. The endto-end inference time for 16 video frames could be within 150 ms, when executing representative C3D and R(2+1)D models on a cellphone. For the first time, real-time execution of 3D CNNs is achieved on off-the-shelf mobiles.

PDF Details

NeurIPS Conference 2021 Conference Paper

Sanity Checks for Lottery Tickets: Does Your Winning Ticket Really Win the Jackpot?

  • Xiaolong Ma
  • Geng Yuan
  • Xuan Shen
  • Tianlong Chen
  • Xuxi Chen
  • Xiaohan Chen
  • Ning Liu
  • Minghai Qin

There have been long-standing controversies and inconsistencies over the experiment setup and criteria for identifying the "winning ticket" in literature. To reconcile such, we revisit the definition of lottery ticket hypothesis, with comprehensive and more rigorous conditions. Under our new definition, we show concrete evidence to clarify whether the winning ticket exists across the major DNN architectures and/or applications. Through extensive experiments, we perform quantitative analysis on the correlations between winning tickets and various experimental factors, and empirically study the patterns of our observations. We find that the key training hyperparameters, such as learning rate and training epochs, as well as the architecture characteristics such as capacities and residual connections, are all highly correlated with whether and when the winning tickets can be identified. Based on our analysis, we summarize a guideline for parameter settings in regards of specific architecture characteristics, which we hope to catalyze the research progress on the topic of lottery ticket hypothesis. Our codes are publicly available at: https: //github. com/boone891214/sanity-check-LTH.

PDF Details

NeurIPS Conference 2021 Conference Paper

ScaleCert: Scalable Certified Defense against Adversarial Patches with Sparse Superficial Layers

  • Husheng Han
  • Kaidi Xu
  • Xing Hu
  • Xiaobing Chen
  • Ling Liang
  • Zidong Du
  • Qi Guo
  • Yanzhi Wang

Adversarial patch attacks that craft the pixels in a confined region of the input images show their powerful attack effectiveness in physical environments even with noises or deformations. Existing certified defenses towards adversarial patch attacks work well on small images like MNIST and CIFAR-10 datasets, but achieve very poor certified accuracy on higher-resolution images like ImageNet. It is urgent to design both robust and effective defenses against such a practical and harmful attack in industry-level larger images. In this work, we propose the certified defense methodology that achieves high provable robustness for high-resolution images and largely improves the practicality for real adoption of the certified defense. The basic insight of our work is that the adversarial patch intends to leverage localized superficial important neurons (SIN) to manipulate the prediction results. Hence, we leverage the SIN-based DNN compression techniques to significantly improve the certified accuracy, by reducing the adversarial region searching overhead and filtering the prediction noises. Our experimental results show that the certified accuracy is increased from 36. 3% (the state-of-the-art certified detection) to 60. 4%on the ImageNet dataset, largely pushing the certified defenses for practical use.

PDF Details

IJCAI Conference 2021 Conference Paper

Towards Fast and Accurate Multi-Person Pose Estimation on Mobile Devices

  • Xuan Shen
  • Geng Yuan
  • Wei Niu
  • Xiaolong Ma
  • Jiexiong Guan
  • Zhengang Li
  • Bin Ren
  • Yanzhi Wang

The rapid development of autonomous driving, abnormal behavior detection, and behavior recognition makes an increasing demand for multi-person pose estimation-based applications, especially on mobile platforms. However, to achieve high accuracy, state-of-the-art methods tend to have a large model size and complex post-processing algorithm, which costs intense computation and long end-to-end latency. To solve this problem, we propose an architecture optimization and weight pruning framework to accelerate inference of multi-person pose estimation on mobile devices. With our optimization framework, we achieve up to 2. 51X faster model inference speed with higher accuracy compared to representative lightweight multi-person pose estimator.

PDF Details DOI

AAAI Conference 2021 Conference Paper

YOLObile: Real-Time Object Detection on Mobile Devices via Compression-Compilation Co-Design

  • Yuxuan Cai
  • Hongjia Li
  • Geng Yuan
  • Wei Niu
  • Yanyu Li
  • Xulong Tang
  • Bin Ren
  • Yanzhi Wang

The rapid development and wide utilization of object detection techniques have aroused attention on both accuracy and speed of object detectors. However, the current state-of-theart object detection works are either accuracy-oriented using a large model but leading to high latency or speed-oriented using a lightweight model but sacrificing accuracy. In this work, we propose YOLObile framework, a real-time object detection on mobile devices via compression-compilation co-design. A novel block-punched pruning scheme is proposed for any kernel size. To improve computational efficiency on mobile devices, a GPU-CPU collaborative scheme is adopted along with advanced compiler-assisted optimizations. Experimental results indicate that our pruning scheme achieves 14× compression rate of YOLOv4 with 49. 0 mAP. Under our YOLObile framework, we achieve 17 FPS inference speed using GPU on Samsung Galaxy S20. By incorporating our proposed GPU-CPU collaborative scheme, the inference speed is increased to 19. 1 FPS, and outperforms the original YOLOv4 by 5× speedup. Source code is at: https: //github. com/nightsnack/YOLObile.

PDF Details

AAAI Conference 2020 Conference Paper

AutoCompress: An Automatic DNN Structured Pruning Framework for Ultra-High Compression Rates

  • Ning Liu
  • Xiaolong Ma
  • Zhiyuan Xu
  • Yanzhi Wang
  • Jian Tang
  • Jieping Ye

Structured weight pruning is a representative model compression technique of DNNs to reduce the storage and computation requirements and accelerate inference. An automatic hyperparameter determination process is necessary due to the large number of flexible hyperparameters. This work proposes AutoCompress, an automatic structured pruning framework with the following key performance improvements: (i) effectively incorporate the combination of structured pruning schemes in the automatic process; (ii) adopt the stateof-art ADMM-based structured weight pruning as the core algorithm, and propose an innovative additional purification step for further weight reduction without accuracy loss; and (iii) develop effective heuristic search method enhanced by experience-based guided search, replacing the prior deep reinforcement learning technique which has underlying incompatibility with the target pruning problem. Extensive experiments on CIFAR-10 and ImageNet datasets demonstrate that AutoCompress is the key to achieve ultra-high pruning rates on the number of weights and FLOPs that cannot be achieved before. As an example, AutoCompress outperforms the prior work on automatic model compression by up to 33× in pruning rate (120× reduction in the actual parameter count) under the same accuracy. Significant inference speedup has been observed from the AutoCompress framework on actual measurements on smartphone. We release models of this work at anonymous link: http: //bit. ly/2VZ63dS.

PDF Details

AAAI Conference 2020 Conference Paper

Embedding Compression with Isotropic Iterative Quantization

  • Siyu Liao
  • Jie Chen
  • Yanzhi Wang
  • Qinru Qiu
  • Bo Yuan

Continuous representation of words is a standard component in deep learning-based NLP models. However, representing a large vocabulary requires significant memory, which can cause problems, particularly on resource-constrained platforms. Therefore, in this paper we propose an isotropic iterative quantization (IIQ) approach for compressing embedding vectors into binary ones, leveraging the iterative quantization technique well established for image retrieval, while satisfying the desired isotropic property of PMI based models. Experiments with pre-trained embeddings (i. e. , GloVe and HDC) demonstrate a more than thirty-fold compression ratio with comparable and sometimes even improved performance over the original real-valued embedding vectors.

PDF Details

AAAI Conference 2020 Conference Paper

PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices

  • Xiaolong Ma
  • Fu-Ming Guo
  • Wei Niu
  • Xue Lin
  • Jian Tang
  • Kaisheng Ma
  • Bin Ren
  • Yanzhi Wang

Model compression techniques on Deep Neural Network (DNN) have been widely acknowledged as an effective way to achieve acceleration on a variety of platforms, and DNN weight pruning is a straightforward and effective method. There are currently two mainstreams of pruning methods representing two extremes of pruning regularity: non-structured, fine-grained pruning can achieve high sparsity and accuracy, but is not hardware friendly; structured, coarse-grained pruning exploits hardware-efficient structures in pruning, but suffers from accuracy drop when the pruning rate is high. In this paper, we introduce PCONV, comprising a new sparsity dimension, – fine-grained pruning patterns inside the coarsegrained structures. PCONV comprises two types of sparsities, Sparse Convolution Patterns (SCP) which is generated from intra-convolution kernel pruning and connectivity sparsity generated from inter-convolution kernel pruning. Essentially, SCP enhances accuracy due to its special vision properties, and connectivity sparsity increases pruning rate while maintaining balanced workload on filter computation. To deploy PCONV, we develop a novel compiler-assisted DNN inference framework and execute PCONV models in real-time without accuracy compromise, which cannot be achieved in prior work. Our experimental results show that, PCONV outperforms three state-of-art end-to-end DNN frameworks, TensorFlow-Lite, TVM, and Alibaba Mobile Neural Network with speedup up to 39. 2×, 11. 4×, and 6. 3×, respectively, with no accuracy loss. Mobile devices can achieve real-time inference on large-scale DNNs.

PDF Details

IJCAI Conference 2020 Conference Paper

Towards Real-Time DNN Inference on Mobile Platforms with Model Pruning and Compiler Optimization

  • Wei Niu
  • Pu Zhao
  • Zheng Zhan
  • Xue Lin
  • Yanzhi Wang
  • Bin Ren

High-end mobile platforms rapidly serve as primary computing devices for a wide range of Deep Neural Network (DNN) applications. However, the constrained computation and storage resources on these devices still pose significant challenges for real-time DNN inference executions. To address this problem, we propose a set of hardware-friendly structured model pruning and compiler optimization techniques to accelerate DNN executions on mobile devices. This demo shows that these optimizations can enable real-time mobile execution of multiple DNN applications, including style transfer, DNN coloring and super resolution.

PDF Details DOI

IJCAI Conference 2019 Conference Paper

Interpreting and Evaluating Neural Network Robustness

  • Fuxun Yu
  • Zhuwei Qin
  • Chenchen Liu
  • Liang Zhao
  • Yanzhi Wang
  • Xiang Chen

Recently, adversarial deception becomes one of the most considerable threats to deep neural networks. However, compared to extensive research in new designs of various adversarial attacks and defenses, the neural networks' intrinsic robustness property is still lack of thorough investigation. This work aims to qualitatively interpret the adversarial attack and defense mechanisms through loss visualization, and establish a quantitative metric to evaluate the model's intrinsic robustness. The proposed robustness metric identifies the upper bound of a model's prediction divergence in the given domain and thus indicates whether the model can maintain a stable prediction. With extensive experiments, our metric demonstrates several advantages over conventional testing accuracy based robustness estimation: (1) it provides a uniformed evaluation to models with different structures and parameter scales; (2) it over-performs conventional accuracy based robustness evaluation and provides a more reliable evaluation that is invariant to different test settings; (3) it can be fast generated without considerable testing cost.

PDF Details

IJCAI Conference 2019 Conference Paper

Protecting Neural Networks with Hierarchical Random Switching: Towards Better Robustness-Accuracy Trade-off for Stochastic Defenses

  • Xiao Wang
  • Siyue Wang
  • Pin-Yu Chen
  • Yanzhi Wang
  • Brian Kulis
  • Xue Lin
  • Sang Chin

Despite achieving remarkable success in various domains, recent studies have uncovered the vulnerability of deep neural networks to adversarial perturbations, creating concerns on model generalizability and new threats such as prediction-evasive misclassification or stealthy reprogramming. Among different defense proposals, stochastic network defenses such as random neuron activation pruning or random perturbation to layer inputs are shown to be promising for attack mitigation. However, one critical drawback of current defenses is that the robustness enhancement is at the cost of noticeable performance degradation on legitimate data, e. g. , large drop in test accuracy. This paper is motivated by pursuing for a better trade-off between adversarial robustness and test accuracy for stochastic network defenses. We propose Defense Efficiency Score (DES), a comprehensive metric that measures the gain in unsuccessful attack attempts at the cost of drop in test accuracy of any defense. To achieve a better DES, we propose hierarchical random switching (HRS), which protects neural networks through a novel randomization scheme. A HRS-protected model contains several blocks of randomly switching channels to prevent adversaries from exploiting fixed model structures and parameters for their malicious purposes. Extensive experiments show that HRS is superior in defending against state-of-the-art white-box and adaptive adversarial misclassification attacks. We also demonstrate the effectiveness of HRS in defending adversarial reprogramming, which is the first defense against adversarial programs. Moreover, in most settings the average DES of HRS is at least 5X higher than current stochastic network defenses, validating its significantly improved robustness-accuracy trade-off.

PDF Details

AAAI Conference 2019 Conference Paper

Universal Approximation Property and Equivalence of Stochastic Computing-Based Neural Networks and Binary Neural Networks

  • Yanzhi Wang
  • Zheng Zhan
  • Liang Zhao
  • Jian Tang
  • Siyue Wang
  • Jiayu Li
  • Bo Yuan
  • Wujie Wen

Large-scale deep neural networks are both memory and computation-intensive, thereby posing stringent requirements on the computing platforms. Hardware accelerations of deep neural networks have been extensively investigated. Specific forms of binary neural networks (BNNs) and stochastic computing-based neural networks (SCNNs) are particularly appealing to hardware implementations since they can be implemented almost entirely with binary operations. Despite the obvious advantages in hardware implementation, these approximate computing techniques are questioned by researchers in terms of accuracy and universal applicability. Also it is important to understand the relative pros and cons of SCNNs and BNNs in theory and in actual hardware implementations. In order to address these concerns, in this paper we prove that the ”ideal” SCNNs and BNNs satisfy the universal approximation property with probability 1 (due to the stochastic behavior), which is a new angle from the original approximation property. The proof is conducted by first proving the property for SCNNs from the strong law of large numbers, and then using SCNNs as a “bridge” to prove for BNNs. Besides the universal approximation property, we also derive an appropriate bound for bit length M in order to provide insights for the actual neural network implementations. Based on the universal approximation property, we further prove that SCNNs and BNNs exhibit the same energy complexity. In other words, they have the same asymptotic energy consumption with the growth of network size. We also provide a detailed analysis of the pros and cons of SCNNs and BNNs for hardware implementations and conclude that SC- NNs are more suitable.

PDF Details

AAAI Conference 2018 Conference Paper

Towards Ultra-High Performance and Energy Efficiency of Deep Learning Systems: An Algorithm-Hardware Co-Optimization Framework

  • Yanzhi Wang
  • Caiwen Ding
  • Zhe Li
  • Geng Yuan
  • Siyu Liao
  • Xiaolong Ma
  • Bo Yuan
  • Xuehai Qian

Hardware accelerations of deep learning systems have been extensively investigated in industry and academia. The aim of this paper is to achieve ultra-high energy efficiency and performance for hardware implementations of deep neural networks (DNNs). An algorithm-hardware co-optimization framework is developed, which is applicable to different DNN types, sizes, and application scenarios. The algorithm part adopts the general block-circulant matrices to achieve a fine-grained tradeoff of accuracy and compression ratio. It applies to both fully-connected and convolutional layers and contains a mathematically rigorous proof of the effectiveness of the method. The proposed algorithm reduces computational complexity per layer from O(n2 ) to O(n log n) and storage complexity from O(n2 ) to O(n), both for training and inference. The hardware part consists of highly efficient Field Programmable Gate Array (FPGA)-based implementations using effective reconfiguration, batch processing, deep pipelining, resource re-using, and hierarchical control. Experimental results demonstrate that the proposed framework achieves at least 152X speedup and 71X energy efficiency gain compared with IBM TrueNorth processor under the same test accuracy. It achieves at least 31X energy efficiency gain compared with the reference FPGA-based work.

PDF Details

ICRA Conference 2017 Conference Paper

Improving contour accuracy of a 2-DOF planar parallel kinematic machine by smart structure based compensation method

  • Yao Jiang 0003
  • Tiemin Li
  • Liping Wang 0001
  • Feifan Chen
  • Yanzhi Wang

High contour accuracy is vital to the multi-axis motion system. Improvement in the contour accuracy of the parallel kinematic machine (PKM) has being a challenging issue in the process of its practical application. In analogy to the intelligent structure of the organisms, this paper proposes a smart structure based compensation (SSBC) method to improve the PKM's contour accuracy. Instead of adopting the advanced control strategies or improving the performances of machine's hardware, this method deals with the contouring issue through innovative mechanical design. The smart structures, which have the ability of perception and control, are introduced into the PKM's components to actively suppress their deformations and vibrations. Meanwhile, the accurate micro motions provided by the smart structure components can be used to compensate for the PKM's contour error caused by the uncoordinated control of the driving system just by using a simple control strategy. The implementation procedure of the SSBC method is first discussed in detail. A 2-DOF PKM is taken as the application object and the smart structures are introduced into its kinematic chains to control their axial deformations and internal forces. Then this PKM's kinematic error is reduced in advance to guarantee the result of the contour accuracy improvement. Finally, several experimental tests are performed to verify the effectiveness the proposed SSBC method in improving the PKM's contour accuracy. Except for the PKM, the SSBC method also provides a novel and effective way for the advanced multi-axis machines to improve their contour accuracies.

Details
v2026.09.13