Author name cluster

Yuan Cheng

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.

21 papers

2 author rows

EAAI Journal 2026 Journal Article

Antinoise Adaptive Time–Frequency Fusion for multivariate time series anomaly detection

Sizhe Huang
Liang Xi
Xunhua Huang
Yuan Cheng
Han Liu

Multivariate time series anomaly detection (MTSAD) is critical for ensuring the security of various related Cyber-Physical Systems (CPS), which are highly dynamic and prone to diverse anomaly patterns. While previous studies have demonstrated the effectiveness of frequency-domain features in enhancing detection performance, existing methods face two key limitations: (i) they fail to balance the contributions of time-domain and frequency-domain features, and (ii) they overlook noise-induced cognitive bias, which is common in various CPS scenarios, leading to misjudgment of normal and abnormal patterns. To address these challenges, we propose an AntiNoise Adaptive Time–Frequency Fusion method for MTSAD (ANAF). ANAF dynamically integrates time- and frequency-domain features with an adaptive weighting strategy and employs a residual structure to preserve temporal dependencies. Furthermore, we design an uncertainty-aware detection strategy with fused features as inputs, to perceive noise-induced cognitive biases and enhance detection robustness and accuracy. Experimental results on four benchmark datasets show that ANAF outperforms state-of-the-art methods, specifically, improving the average F1 score by 11. 09%.

Details DOI

AAAI Conference 2026 Conference Paper

PET2Rep: Towards Vision-Language Model-Drived Automated Radiology Report Generation for Positron Emission Tomography

Yichi Zhang
Wenbo Zhang
Zehui Ling
Gang Feng
Sisi Peng
Deshu Chen
Yuchen Liu
Hongwei Zhang

Positron emission tomography (PET) is a cornerstone of modern oncologic and neurologic imaging, distinguished by its unique ability to illuminate dynamic metabolic processes that transcend the anatomical focus of traditional imaging technologies. Radiology reports are essential for clinical decision making, yet their manual creation is labor-intensive and time-consuming. Recent advancements of vision-language models (VLMs) have shown strong potential in medical applications, presenting a promising avenue for automating report generation. However, existing applications of VLMs in the medical domain have predominantly focused on structural imaging modalities, while the unique characteristics of molecular PET imaging have largely been overlooked. To bridge the gap, we introduce PET2Rep, a large-scale comprehensive benchmark for evaluation of general and medical VLMs for radiology report generation for PET images. PET2Rep stands out as the first dedicated dataset for PET report generation with metabolic information, uniquely capturing whole-body image-report pairs that cover dozens of organs to fill the critical gap in existing benchmarks and mirror real-world clinical comprehensiveness. In addition to widely recognized natural language generation metrics, we introduce a series of clinical efficiency metrics to evaluate the quality of radiotracer uptake pattern description in key organs in generated reports. We conduct a head-to-head comparison of 30 cutting-edge general-purpose and medical-specialized VLMs. The results show that the current state-of-the-art VLMs perform poorly on PET report generation task, falling considerably short of fulfilling practical needs. Moreover, we identify several key insufficiency that need to be addressed to advance the development in medical applications. We believe PET2Rep will serve as a platform for the development and application of VLMs for PET imaging, accelerating the development of trustworthy reporting tools that can genuinely alleviate radiologist burden and enhance patient care.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Structure-based RNA Design by Step-wise Optimization of Latent Diffusion Model

Qi Si
Xuyang Liu
Penglei Wang
Xin Guo
Yuan Qi
Yuan Cheng

RNA inverse folding, designing sequences to form specific 3D structures, is critical for therapeutics, gene regulation, and synthetic biology. Current methods, focused on sequence recovery, struggle to address structural objectives like secondary structure consistency (SS), minimum free energy (MFE), and local distance difference test (LDDT), leading to suboptimal structural accuracy. To tackle this, we propose a reinforcement learning (RL) framework integrated with a latent diffusion model (LDM). Drawing inspiration from the success of diffusion models in RNA inverse folding, which adeptly model complex sequence-structure interactions, we develop an LDM incorporating pre-trained RNA-FM embeddings from a large-scale RNA model. These embeddings capture co-evolutionary patterns, markedly improving sequence recovery accuracy. However, existing approaches, including diffusion-based methods, cannot effectively handle non-differentiable structural objectives. By contrast, RL excels in this task by using policy-driven reward optimization to navigate complex, non-gradient-based objectives, offering a significant advantage over traditional methods. In summary, we propose the Step-wise Optimization of Latent Diffusion Model (SOLD), a novel RL framework that optimizes single-step noise without sampling the full diffusion trajectory, achieving efficient refinement of multiple structural objectives. Experimental results demonstrate SOLD surpasses its LDM baseline and state-of-the-art methods across all metrics, establishing a robust framework for RNA inverse folding with profound implications for biotechnological and therapeutic applications.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Tracing the Heart’s Pathways: ECG Representation Learning from a Cardiac Conduction Perspective

Tan Pan
Yixuan Sun
Chen Jiang
Qiong Gao
Rui Sun
Xingmeng Zhang
Zhenqi Yang
LIMEI HAN

The multi-lead electrocardiogram (ECG) stands as a cornerstone of cardiac diagnosis. Recent strides in electrocardiogram self-supervised learning (eSSL) have brightened prospects for enhancing representation learning without relying on high-quality annotations. Yet earlier eSSL methods suffer a key limitation: they focus on consistent patterns across leads and beats, overlooking the inherent differences in heartbeats rooted in cardiac conduction processes, while subtle but significant variations carry unique physiological signatures. Moreover, representation learning for ECG analysis should align with ECG diagnostic guidelines, which progress from individual heartbeats to single leads and ultimately to lead combinations. This sequential logic, however, is often neglected when applying pre-trained models to downstream tasks. To address these gaps, we propose CLEAR-HUG, a two-stage framework designed to capture subtle variations in cardiac conduction across leads while adhering to ECG diagnostic guidelines. In the first stage, we introduce an eSSL model termed Conduction-LEAd Reconstructor (CLEAR), which captures both specific variations and general commonalities across heartbeats. Treating each heartbeat as a distinct entity, CLEAR employs a simple yet effective sparse attention mechanism to reconstruct signals without interference from other heartbeats. In the second stage, we implement a Hierarchical lead-Unified Group head (HUG) for disease diagnosis, mirroring clinical workflow. Experimental results across six tasks show a 6.84% improvement, validating the effectiveness of CLEAR-HUG. This highlights its ability to enhance representations of cardiac conduction and align patterns with expert diagnostic guidelines.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

ChromFound: Towards A Universal Foundation Model for Single-Cell Chromatin Accessibiltiy Data

Yifeng Jiao
Yuchen Liu
Yu Zhang
Xin Guo
Yushuai Wu
Chen Jiang
Jiyang Li
Hongwei Zhang

The advent of single-cell Assay for Transposase-Accessible Chromatin using sequencing (scATAC-seq) offers an innovative perspective for deciphering regulatory mechanisms by assembling a vast repository of single-cell chromatin accessibility data. While foundation models have achieved significant success in single-cell transcriptomics, there is currently no foundation model for scATAC-seq that supports zero-shot high-quality cell identification and comprehensive multi-omics analysis simultaneously. Key challenges lie in the high dimensionality and sparsity of scATAC-seq data, as well as the lack of a standardized schema for representing open chromatin regions (OCRs). Here, we present ChromFound, a foundation model tailored for scATAC-seq. ChromFound utilizes a hybrid architecture and genome-aware tokenization to effectively capture genome-wide long contexts and regulatory signals from dynamic chromatin landscapes. Pretrained on 1. 97 million cells from 30 tissues and 6 disease conditions, ChromFound demonstrates broad applicability across 6 diverse tasks. Notably, it achieves robust zero-shot performance in generating universal cell representations and exhibits excellent transferability in cell type annotation and cross-omics prediction. By uncovering enhancer-gene links undetected by existing computational methods, ChromFound offers a promising framework for understanding disease risk variants in the noncoding genome. The implementation of ChromFound is available via https: //github. com/JohnsonKlose/ChromFound.

PDF Details

ICML Conference 2025 Conference Paper

We propose Network Automatic Relevance Determination (NARD), an extension of ARD for linearly probabilistic models, to simultaneously model sparse relationships between inputs $X \in \mathbb R^{d \times N}$ and outputs $Y \in \mathbb R^{m \times N}$, while capturing the correlation structure among the $Y$. NARD employs a matrix normal prior which contains a sparsity-inducing parameter to identify and discard irrelevant features, thereby promoting sparsity in the model. Algorithmically, it iteratively updates both the precision matrix and the relationship between $Y$ and the refined inputs. To mitigate the computational inefficiencies of the $\mathcal O(m^3 + d^3)$ cost per iteration, we introduce Sequential NARD, which evaluates features sequentially, and a Surrogate Function Method, leveraging an efficient approximation of the marginal likelihood and simplifying the calculation of determinant and inverse of an intermediate matrix. Combining the Sequential update with the Surrogate Function method further reduces computational costs. The computational complexity per iteration for these three methods is reduced to $\mathcal O(m^3+p^3)$, $\mathcal O(m^3 + d^2)$, $\mathcal O(m^3+p^2)$ respectively, where $p \ll d$ is the final number of features in the model. Our methods demonstrate significant improvements in computational efficiency with comparable performance on both synthetic and real-world datasets.

Details

NeurIPS Conference 2025 Conference Paper

Minimal Semantic Sufficiency Meets Unsupervised Domain Generalization

Tan Pan
Kaiyu Guo
Dongli Xu
Zhaorui Tan
Chen Jiang
Deshu Chen
Xin Guo
Brian Lovell

The generalization ability of deep learning has been extensively studied in supervised settings, yet it remains less explored in unsupervised scenarios. Recently, the Unsupervised Domain Generalization (UDG) task has been proposed to enhance the generalization of models trained with prevalent unsupervised learning techniques, such as Self-Supervised Learning (SSL). UDG confronts the challenge of distinguishing semantics from variations without category labels. Although some recent methods have employed domain labels to tackle this issue, such domain labels are often unavailable in real-world contexts. In this paper, we address these limitations by formalizing UDG as the task of learning a Minimal Sufficient Semantic Representation: a representation that (i) preserves all semantic information shared across augmented views (sufficiency), and (ii) maximally removes information irrelevant to semantics (minimality). We theoretically ground these objectives from the perspective of information theory, demonstrating that optimizing representations to achieve sufficiency and minimality directly reduces out-of-distribution risk. Practically, we implement this optimization through Minimal-Sufficient UDG (MS-UDG), a learnable model by integrating (a) an InfoNCE-based objective to achieve sufficiency; (b) two complementary components to promote minimality: a novel semantic-variation disentanglement loss and a reconstruction-based mechanism for capturing adequate variation. Empirically, MS-UDG sets a new state-of-the-art on popular unsupervised domain-generalization benchmarks, consistently outperforming existing SSL and UDG methods, without category or domain labels during representation learning.

PDF Details

ICLR Conference 2025 Conference Paper

OSTQuant: Refining Large Language Model Quantization with Orthogonal and Scaling Transformations for Better Distribution Fitting

Xing Hu 0010
Yuan Cheng
Dawei Yang
Zhixuan Chen
Zukang Xu
Jiangyong Yu
Chen Xu
Zhihang Yuan

Post-training quantization (PTQ) has emerged as a widely adopted technique for compressing and accelerating Large Language Models (LLMs). The major challenge in LLM quantization is that uneven and heavy-tailed data distributions can expand the quantization range, thereby reducing bit precision for most values. Recent methods attempt to eliminate outliers and balance inter-channel differences by employing linear transformations; however, they remain heuristic and are often overlook optimizing the data distribution across the entire quantization space. In this paper, we introduce Quantization Space Utilization Rate (QSUR), a novel metric that effectively assesses the quantizability of transformed data by measuring the space utilization of the data in the quantization space. We complement QSUR with mathematical derivations that examine the effects and limitations of various transformations, guiding our development of Orthogonal and Scaling Transformation-based Quantization (OSTQuant). OSTQuant employs a learnable equivalent transformation, consisting of an orthogonal transformation and a scaling transformation, to optimize the distributions of weights and activations across the entire quantization space. Futhermore, we propose the KL-Top loss function, designed to mitigate noise during optimization while retaining richer semantic information within the limited calibration data imposed by PTQ. OSTQuant outperforms existing work on various LLMs and benchmarks. In the W4-only setting, it retains 99.5\% of the floating-point accuracy. In the more challenging W4A4KV4 configuration, OSTQuant reduces the performance gap by 32\% on the LLaMA-3-8B model compared to state-of-the-art methods. Code will be available.

Details

AAAI Conference 2025 Conference Paper

Pushing the Limits of BFP on Narrow Precision LLM Inference

Hui Wang
Yuan Cheng
Xiaomeng Han
Zhengpeng Zhao
Dawei Yang
Zhe Jiang

The substantial computational and memory demands of Large Language Models (LLMs) hinder their deployment. Block Floating Point (BFP) has proven effective in accelerating linear operations, a cornerstone of LLM workloads. However, as sequence lengths grow, nonlinear operations, such as Attention, increasingly become performance bottlenecks due to their quadratic computational complexity. These nonlinear operations are predominantly executed using inefficient floating-point formats, which renders the system challenging to optimize software efficiency and hardware overhead. In this paper, we delve into the limitations and potential of applying BFP to nonlinear operations. Given our findings, we introduce a hardware-software co-design framework (DB-Attn), including: (i) DBFP, an advanced BFP version, overcomes nonlinear operation challenges with a pivot-focus strategy for diverse data and an adaptive grouping strategy for flexible exponent sharing. (ii) DH-LUT, a novel lookup table algorithm dedicated to accelerating nonlinear operations with DBFP format. (iii) An RTL-level DBFP-based engine is implemented to support DB-Attn, applicable to FPGA and ASIC. Results show that DB-Attn provides significant performance improvements with negligible accuracy loss, achieving 74% GPU speedup on Softmax of LLaMA and 10x low-overhead performance improvement over SOTA designs.

PDF Details DOI

ICML Conference 2024 Conference Paper

In-context Convergence of Transformers

Yu Huang 0023
Yuan Cheng
Yingbin Liang

Transformers have recently revolutionized many domains in modern machine learning and one salient discovery is their remarkable in-context learning capability, where models can solve an unseen task by utilizing task-specific prompts without further parameters fine-tuning. This also inspired recent theoretical studies aiming to understand the in-context learning mechanism of transformers, which however focused only on linear transformers. In this work, we take the first step toward studying the learning dynamics of a one-layer transformer with softmax attention trained via gradient descent in order to in-context learn linear function classes. We consider a structured data model, where each token is randomly sampled from a set of feature vectors in either balanced or imbalanced fashion. For data with balanced features, we establish the finite-time convergence guarantee with near-zero prediction error by navigating our analysis over two phases of the training dynamics of the attention map. More notably, for data with imbalanced features, we show that the learning dynamics take a stage-wise convergence process, where the transformer first converges to a near-zero prediction error for the query tokens of dominant features, and then converges later to a near-zero error for query tokens of under-represented features, via one and four training phases. Our proof features new techniques for analyzing the competing strengths of two types of attention weights, the change of which determines different training phases.

Details

ICLR Conference 2024 Conference Paper

Provable Benefits of Multi-task RL under Non-Markovian Decision Making Processes

Ruiquan Huang
Yuan Cheng
Jing Yang 0002
Vincent Tan
Yingbin Liang

In multi-task reinforcement learning (RL) under Markov decision processes (MDPs), the presence of shared latent structures among multiple MDPs has been shown to yield significant benefits to the sample efficiency compared to single-task RL. In this paper, we investigate whether such a benefit can extend to more general sequential decision making problems such as predictive state representations (PSRs). The main challenge here is that the large and complex model space makes it hard to identify what types of common latent structure of multi-task PSRs can reduce the model complexity and improve sample efficiency. To this end, we posit a joint model class for tasks and use the notion of $\eta$-bracketing number to quantify its complexity; this number also serves as a general metric to capture the similarity of tasks and thus determines the benefit of multi-task over single-task RL. We first study upstream multi-task learning over PSRs, in which all tasks share the same observation and action spaces. We propose a provably efficient algorithm UMT-PSR for finding near-optimal policies for all PSRs, and demonstrate that the advantage of multi-task learning manifests if the joint model class of PSRs has a smaller $\eta$-bracketing number compared to that of individual single-task learning. We further investigate downstream learning, in which the agent needs to learn a new target task that shares some commonalities with the upstream tasks via a similarity constraint. By exploiting the learned PSRs from the upstream, we develop a sample-efficient algorithm that provably finds a near-optimal policy. Upon specialization to some examples with small $\eta$-bracketing numbers, our results further highlight the benefit compared to directly learning a single-task PSR.

Details

AAAI Conference 2023 Conference Paper

DC-Former: Diverse and Compact Transformer for Person Re-identification

Wen Li
Cheng Zou
Meng Wang
Furong Xu
Jianan Zhao
Ruobing Zheng
Yuan Cheng
Wei Chu

In person re-identification (ReID) task, it is still challenging to learn discriminative representation by deep learning, due to limited data. Generally speaking, the model will get better performance when increasing the amount of data. The addition of similar classes strengthens the ability of the classifier to identify similar identities, thereby improving the discrimination of representation. In this paper, we propose a Diverse and Compact Transformer (DC-Former) that can achieve a similar effect by splitting embedding space into multiple diverse and compact subspaces. Compact embedding subspace helps model learn more robust and discriminative embedding to identify similar classes. And the fusion of these diverse embeddings containing more fine-grained information can further improve the effect of ReID. Specifically, multiple class tokens are used in vision transformer to represent multiple embedding spaces. Then, a self-diverse constraint (SDC) is applied to these spaces to push them away from each other, which makes each embedding space diverse and compact. Further, a dynamic weight controller (DWC) is further designed for balancing the relative importance among them during training. The experimental results of our method are promising, which surpass previous state-of-the-art methods on several commonly used person ReID benchmarks. Our code is available at https://github.com/ant-research/Diverse-and-Compact-Transformer.

PDF Details DOI

ICLR Conference 2023 Conference Paper

Improved Sample Complexity for Reward-free Reinforcement Learning under Low-rank MDPs

Yuan Cheng
Ruiquan Huang
Yingbin Liang
Jing Yang 0002

In reward-free reinforcement learning (RL), an agent explores the environment first without any reward information, in order to achieve certain learning goals afterwards for any given reward. In this paper we focus on reward-free RL under low-rank MDP models, in which both the representation and linear weight vectors are unknown. Although various algorithms have been proposed for reward-free low-rank MDPs, the corresponding sample complexity is still far from being satisfactory. In this work, we first provide the first known sample complexity lower bound that holds for any algorithm under low-rank MDPs. This lower bound implies it is strictly harder to find a near-optimal policy under low-rank MDPs than under linear MDPs. We then propose a novel model-based algorithm, coined RAFFLE, and show it can both find an $\epsilon$-optimal policy and achieve an $\epsilon$-accurate system identification via reward-free exploration, with a sample complexity significantly improving the previous results. Such a sample complexity matches our lower bound in the dependence on $\epsilon$, as well as on $K$ {in the large $d$ regime}, where $d$ and $K$ respectively denote the representation dimension and action space cardinality. Finally, we provide a planning algorithm (without further interaction with true environment) for RAFFLE to learn a near-accurate representation, which is the first known representation learning guarantee under the same setting.

Details

TMLR Journal 2023 Journal Article

Online Min-max Problems with Non-convexity and Non-stationarity

Yu Huang
Yuan Cheng
Yingbin Liang
Longbo Huang

Online min-max optimization has recently gained considerable interest due to its rich applications to game theory, multi-agent reinforcement learning, online robust learning, etc. Theoretical understanding in this field has been mainly focused on convex-concave settings. Online min-max optimization with nonconvex geometries, which captures various online deep learning problems, has yet been studied so far. In this paper, we make the first effort and investigate online nonconvex-strongly-concave min-max optimization in the nonstationary environment. We first introduce a natural notion of local Nash equilibrium (NE)-regret, and then propose a novel algorithm coined TSODA to achieve the optimal regret. We further generalize our study to the setting with stochastic first-order feedback, and show that a variation of TSODA can also achieve the same optimal regret in expectation. Our theoretical results and the superior performance of the proposed method are further validated by empirical experiments. To our best knowledge, this is the first exploration of efficient online nonconvex min-max optimization.

PDF Details

NeurIPS Conference 2023 Conference Paper

Provably Efficient Algorithm for Nonstationary Low-Rank MDPs

Yuan Cheng
Jing Yang
Yingbin Liang

Reinforcement learning (RL) under changing environment models many real-world applications via nonstationary Markov Decision Processes (MDPs), and hence gains considerable interest. However, theoretical studies on nonstationary MDPs in the literature have mainly focused on tabular and linear (mixture) MDPs, which do not capture the nature of unknown representation in deep RL. In this paper, we make the first effort to investigate nonstationary RL under episodic low-rank MDPs, where both transition kernels and rewards may vary over time, and the low-rank model contains unknown representation in addition to the linear state embedding function. We first propose a parameter-dependent policy optimization algorithm called PORTAL, and further improve PORTAL to its parameter-free version of Ada-PORTAL, which is able to tune its hyper-parameters adaptively without any prior knowledge of nonstationarity. For both algorithms, we provide upper bounds on the average dynamic suboptimality gap, which show that as long as the nonstationarity is not significantly large, PORTAL and Ada-PORTAL are sample-efficient and can achieve arbitrarily small average dynamic suboptimality gap with polynomial sample complexity.

PDF Details

AAAI Conference 2022 Conference Paper

Deeply Tensor Compressed Transformers for End-to-End Object Detection

Peining Zhen
Ziyang Gao
Tianshu Hou
Yuan Cheng
Hai-Bao Chen

DEtection TRansformer (DETR) is a recently proposed method that streamlines the detection pipeline and achieves competitive results against two-stage detectors such as Faster- RCNN. The DETR models get rid of complex anchor generation and post-processing procedures thereby making the detection pipeline more intuitive. However, the numerous redundant parameters in transformers make the DETR models computation and storage intensive, which seriously hinder them to be deployed on the resources-constrained devices. In this paper, to obtain a compact end-to-end detection framework, we propose to deeply compress the transformers with low-rank tensor decomposition. The basic idea of the tensor-based compression is to represent the large-scale weight matrix in one network layer with a chain of low-order matrices. Furthermore, we propose a gated multi-head attention (GMHA) module to mitigate the accuracy drop of the tensor-compressed DETR models. In GMHA, each attention head has an independent gate to determine the passed attention value. The redundant attention information can be suppressed by adopting the normalized gates. Lastly, to obtain fully compressed DETR models, a low-bitwidth quantization technique is introduced for further reducing the model storage size. Based on the proposed methods, we can achieve significant parameter and model size reduction while maintaining high detection performance. We conduct extensive experiments on the COCO dataset to validate the effectiveness of our tensor-compressed (tensorized) DETR models. The experimental results show that we can attain 3. 7× full model compression with 482× feed forward network (FFN) parameter reduction and only 0. 6 points accuracy drop.

PDF Details

NeurIPS Conference 2022 Conference Paper

Provable Benefit of Multitask Representation Learning in Reinforcement Learning

Yuan Cheng
Songtao Feng
Jing Yang
Hong Zhang
Yingbin Liang

As representation learning becomes a powerful technique to reduce sample complexity in reinforcement learning (RL) in practice, theoretical understanding of its advantage is still limited. In this paper, we theoretically characterize the benefit of representation learning under the low-rank Markov decision process (MDP) model. We first study multitask low-rank RL (as upstream training), where all tasks share a common representation, and propose a new multitask reward-free algorithm called REFUEL. REFUEL learns both the transition kernel and the near-optimal policy for each task, and outputs a well-learned representation for downstream tasks. Our result demonstrates that multitask representation learning is provably more sample-efficient than learning each task individually, as long as the total number of tasks is above a certain threshold. We then study the downstream RL in both online and offline settings, where the agent is assigned with a new task sharing the same representation as the upstream tasks. For both online and offline settings, we develop a sample-efficient algorithm, and show that it finds a near-optimal policy with the suboptimality gap bounded by the sum of the estimation error of the learned representation in upstream and a vanishing term as the number of downstream samples becomes large. Our downstream results of online and offline RL further capture the benefit of employing the learned representation from upstream as opposed to learning the representation of the low-rank model directly. To the best of our knowledge, this is the first theoretical study that characterizes the benefit of representation learning in exploration-based reward-free multitask RL for both upstream and downstream tasks.

PDF Details

AAAI Conference 2022 Conference Paper

Width & Depth Pruning for Vision Transformers

Fang Yu
Kun Huang
Meng Wang
Yuan Cheng
Wei Chu
Li Cui

Transformer models have demonstrated their promising potential and achieved excellent performance on a series of computer vision tasks. However, the huge computational cost of vision transformers hinders their deployment and application to edge devices. Recent works have proposed to find and remove the unimportant units of vision transformers. Despite achieving remarkable results, these methods take one dimension of network width into consideration and ignore network depth, which is another important dimension for pruning vision transformers. Therefore, we propose a Width & Depth Pruning (WDPruning) framework that reduces both width and depth dimensions simultaneously. Specifically, for width pruning, a set of learnable pruning-related parameters is used to adaptively adjust the width of transformer. For depth pruning, we introduce several shallow classifiers by using the intermediate information of the transformer blocks, which allows images to be classified by shallow classifiers instead of the deeper classifiers. In the inference period, all of the blocks after shallow classifiers can be dropped so they don’t bring additional parameters and computation. Experimental results on benchmark datasets demonstrate that the proposed method can significantly reduce the computational costs of mainstream vision transformers such as DeiT and Swin Transformer with a minor accuracy drop. In particular, on ILSVRC-12, we achieve over 22% pruning ratio of FLOPs by compressing DeiT-Base, even with an increase of 0. 14% Top-1 accuracy.

PDF Details

TIST Journal 2021 Journal Article

S3-Net: A Fast Scene Understanding Network by Single-Shot Segmentation for Autonomous Driving

Yuan Cheng
Yuchao Yang
Hai-Bao Chen
Ngai Wong
Hao Yu

Real-time segmentation and understanding of driving scenes are crucial in autonomous driving. Traditional pixel-wise approaches extract scene information by segmenting all pixels in a frame, and hence are inefficient and slow. Proposal-wise approaches only learn from the proposed object candidates, but still require multiple steps on the expensive proposal methods. Instead, this work presents a fast single-shot segmentation strategy for video scene understanding. The proposed net, called S3-Net, quickly locates and segments target sub-scenes, and meanwhile extracts attention-aware time-series sub-scene features ( ats-features ) as inputs to an attention-aware spatio-temporal model (ASM). Utilizing tensorization and quantization techniques, S3-Net is intended to be lightweight for edge computing. Experiments results on CityScapes, UCF11, HMDB51, and MOMENTS datasets demonstrate that the proposed S3-Net achieves an accuracy improvement of 8.1% versus the 3D-CNN based approach on UCF11, a storage reduction of 6.9× and an inference speed of 22.8 FPS on CityScapes with a GTX1080Ti GPU.

Details DOI

AAAI Conference 2020 System Paper

Automatic Car Damage Assessment System: Reading and Understanding Videos as Professional Insurance Inspectors

Wei Zhang
Yuan Cheng
Xin Guo
Qingpei Guo
Jian Wang
Qing Wang
Chen Jiang
Meng Wang

We demonstrate a car damage assessment system in car insurance ﬁeld based on artiﬁcial intelligence techniques, which can exempt insurance inspectors from checking cars on site and help people without professional knowledge to evaluate car damages when accidents happen. Unlike existing approaches, we utilize videos instead of photos to interact with users to make the whole procedure as simple as possible. We adopt object and video detection and segmentation techniques in computer vision, and take advantage of multiple frames extracted from videos to achieve high damage recognition accuracy. The system uploads video streams captured by mobile devices, recognizes car damage on the cloud asynchronously and then returns damaged components and repair costs to users. The system evaluates car damages and returns results automatically and effectively in seconds, which reduces laboratory costs and decreases insurance claim time signiﬁcantly.

PDF Details

TCS Journal 2015 Journal Article

Segmenting microarray images using a contour-based method

Mihaela Păun
Yang Li
Yuan Cheng
Iris Tusa
Andrei Păun

In this work we describe a new segmentation technique for the Affymetrix microarray images. We prove that our method can offer better predictions on the gene levels as opposed to the standard Affymetrix segmentation implemented in the Affymetrix GeneChip Operating Software (GCOS). To check the accuracy and show the benefits of the new segmentation method we use a previously implemented methodology to simulate microarray images with realistic features. Using such an “artificial” image provides us with the actual levels for each spot and each gene investigated in the microarray. Using this information we then proceed to segment the same image twice (with GCOS and our new method). The two segmentations will produce two sets of gene levels that are then compared to the known gene levels (known since the moment of generating the “artificial” image). Using this methodology we are able to show statistically (using 50 replicates of the same steps of generating the image, segmenting, comparing the results) that in some cases our new method greatly outperforms the GCOS implemented segmentation method, while in the rest of the cases performs in similar fashion.

Details DOI