Hao Gao

Liuji Chen, Hao Gao, Jinghao Zhang et al.

Tool learning has emerged as a powerful auxiliary mechanism that extends the capabilities of large language models (LLMs), enabling them to address complex tasks that demand real-time relevance or high-precision operations. However, beneath this strength lie significant security risks. Prior studies have primarily concentrated on corrupting the outputs of invoked tools, while largely overlooking the vulnerability of the tool selection process itself. To bridge this gap, we introduce a black-box, text-based attack that substantially increases the likelihood of a target tool being selected. We propose SEEM, a two-level coarse-to-fine perturbation method that operates at both the word and character levels. Through comprehensive experiments, we show that merely perturbing the textual information of tools can markedly raise the probability of the target tool being prioritized and ranked higher among candidates. Our findings expose critical weaknesses in the tool selection mechanism and lay the groundwork for developing defenses to secure this essential process.

Ge Wang, Jie Zhang, Hao Gao et al.

Current tomographic imaging systems need major improvements, especially when multi-dimensional, multi-scale, multi-temporal and multi-parametric phenomena are under investigation. Both preclinical and clinical imaging now depend on in vivo tomography, often requiring separate evaluations by different imaging modalities to define morphologic details, delineate interval changes due to disease or interventions, and study physiological functions that have interconnected aspects. Over the past decade, fusion of multimodality images has emerged with two different approaches: post-hoc image registration and combined acquisition on PET-CT, PET-MRI and other hybrid scanners. There are intrinsic limitations for both the post-hoc image analysis and dual/triple modality approaches defined by registration errors and physical constraints in the acquisition chain. We envision that tomography will evolve beyond current modality fusion and towards grand fusion, a large scale fusion of all or many imaging modalities, which may be referred to as omni-tomography or multi-tomography. Unlike modality fusion, grand fusion is here proposed for truly simultaneous but often localized reconstruction in terms of all or many relevant imaging mechanisms such as CT, MRI, PET, SPECT, US, optical, and possibly more. In this paper, the technical basis for omni-tomography is introduced and illustrated with a top-level design of a next generation scanner, interior tomographic reconstructions of representative modalities, and anticipated applications of omni-tomography.

Xiaoqi Wang, Jian Xiong, Hao Gao et al.

The optimization objective of regression-based blind image quality assessment (IQA) models is to minimize the mean prediction error across the training dataset, which can lead to biased parameter estimation due to potential training data biases. To mitigate this issue, we propose a regression-free framework for image quality evaluation, which is based upon retrieving locally similar instances by incorporating semantic and distortion feature spaces. The approach is motivated by the observation that the human visual system (HVS) exhibits analogous perceptual responses to semantically similar image contents impaired by identical distortions, which we term as content-distortion consistency. The proposed method constructs a hierarchical k-nearest neighbor (k-NN) algorithm for instance retrieval through two classification modules: semantic classification (SC) module and distortion classification (DC) module. Given a test image and an IQA database, the SC module retrieves multiple pristine images semantically similar to the test image. The DC module then retrieves instances based on distortion similarity from the distorted images that correspond to each retrieved pristine image. Finally, quality prediction is obtained by aggregating the subjective scores of the retrieved instances. Without training on subjective quality scores, the proposed regression-free method achieves competitive, even superior performance compared to state-of-the-art regression-based methods on authentic and synthetic distortion IQA benchmarks.

Arash Rabbani, Hao Gao, Dirk Husmeier

In this study, we have tailored a pixel tracking method for temporal extrapolation of the ventricular segmentation masks in cardiac magnetic resonance images. The pixel tracking process starts from the end-diastolic frame of the heart cycle using the available manually segmented images to predict the end-systolic segmentation mask. The superpixels approach is used to divide the raw images into smaller cells and in each time frame, new labels are assigned to the image cells which leads to tracking the movement of the heart wall elements through different frames. The tracked masks at the end of systole are compared with the already available manually segmented masks and dice scores are found to be between 0.81 to 0.84. Considering the fact that the proposed method does not necessarily require a training dataset, it could be an attractive alternative approach to deep learning segmentation methods in scenarios where training data are limited.

Hao Gao, Shaoyu Chen, Yifan Zhu et al.

High-level autonomous driving requires motion planners capable of modeling multimodal future uncertainties while remaining robust in closed-loop interactions. Although diffusion-based planners are effective at modeling complex trajectory distributions, they often suffer from stochastic instabilities and the lack of corrective negative feedback when trained purely with imitation learning. To address these issues, we propose RAD-2, a unified generator-discriminator framework for closed-loop planning. Specifically, a diffusion-based generator is used to produce diverse trajectory candidates, while an RL-optimized discriminator reranks these candidates according to their long-term driving quality. This decoupled design avoids directly applying sparse scalar rewards to the full high-dimensional trajectory space, thereby improving optimization stability. To further enhance reinforcement learning, we introduce Temporally Consistent Group Relative Policy Optimization, which exploits temporal coherence to alleviate the credit assignment problem. In addition, we propose On-policy Generator Optimization, which converts closed-loop feedback into structured longitudinal optimization signals and progressively shifts the generator toward high-reward trajectory manifolds. To support efficient large-scale training, we introduce BEV-Warp, a high-throughput simulation environment that performs closed-loop evaluation directly in Bird's-Eye View feature space via spatial warping. RAD-2 reduces the collision rate by 56% compared with strong diffusion-based planners. Real-world deployment further demonstrates improved perceived safety and driving smoothness in complex urban traffic.

Yuehao Song, Shaoyu Chen, Hao Gao et al.

Vision-language models (VLMs) enhance the planning capability of end-to-end (E2E) driving policy by leveraging high-level semantic reasoning. However, existing approaches often overlook the dual-system consistency between VLM's high-level decision and E2E's low-level planning. As a result, the generated trajectories may misalign with the intended driving decisions, leading to weakened top-down guidance and decision-following ability of the system. To address this issue, we propose Senna-2, an advanced VLM-E2E driving policy that explicitly aligns the two systems for consistent decision-making and planning. Our method follows a consistency-oriented three-stage training paradigm. In the first stage, we conduct driving pre-training to achieve preliminary decision-making and planning, with a decision adapter transmitting VLM decisions to E2E policy in the form of implicit embeddings. In the second stage, we align the VLM and the E2E policy in an open-loop setting. In the third stage, we perform closed-loop alignment via bottom-up Hierarchical Reinforcement Learning in 3DGS environments to reinforce the safety and efficiency. Extensive experiments demonstrate that Senna-2 achieves superior dual-system consistency (19.3% F1 score improvement) and significantly enhances driving safety in both open-loop (5.7% FDE reduction) and closed-loop settings (30.6% AF-CR reduction).

Jie Wang, Jiu-Cheng Xie, Xianyan Li et al.

Constructing vivid 3D head avatars for given subjects and realizing a series of animations on them is valuable yet challenging. This paper presents GaussianHead, which models the actional human head with anisotropic 3D Gaussians. In our framework, a motion deformation field and multi-resolution tri-plane are constructed respectively to deal with the head's dynamic geometry and complex texture. Notably, we impose an exclusive derivation scheme on each Gaussian, which generates its multiple doppelgangers through a set of learnable parameters for position transformation. With this design, we can compactly and accurately encode the appearance information of Gaussians, even those fitting the head's particular components with sophisticated structures. In addition, an inherited derivation strategy for newly added Gaussians is adopted to facilitate training acceleration. Extensive experiments show that our method can produce high-fidelity renderings, outperforming state-of-the-art approaches in reconstruction, cross-identity reenactment, and novel view synthesis tasks. Our code is available at: https://github.com/chiehwangs/gaussian-head.

Hao Gao, Shaoyu Chen, Bo Jiang et al.

Existing end-to-end autonomous driving (AD) algorithms typically follow the Imitation Learning (IL) paradigm, which faces challenges such as causal confusion and an open-loop gap. In this work, we propose RAD, a 3DGS-based closed-loop Reinforcement Learning (RL) framework for end-to-end Autonomous Driving. By leveraging 3DGS techniques, we construct a photorealistic digital replica of the real physical world, enabling the AD policy to extensively explore the state space and learn to handle out-of-distribution scenarios through large-scale trial and error. To enhance safety, we design specialized rewards to guide the policy in effectively responding to safety-critical events and understanding real-world causal relationships. To better align with human driving behavior, we incorporate IL into RL training as a regularization term. We introduce a closed-loop evaluation benchmark consisting of diverse, previously unseen 3DGS environments. Compared to IL-based methods, RAD achieves stronger performance in most closed-loop metrics, particularly exhibiting a 3x lower collision rate. Abundant closed-loop results are presented in the supplementary material. Code is available at https://github.com/hustvl/RAD for facilitating future research.

Haowei Sun, Kai Zhou, Hao Gao et al.

Visual Active Tracking (VAT) aims to control cameras to follow a target in 3D space, which is critical for applications like drone navigation and security surveillance. However, it faces two key bottlenecks in real-world deployment: confusion from visually similar distractors caused by insufficient instance-level discrimination and severe failure under occlusions due to the absence of active planning. To address these, we propose OA-VAT, a unified pipeline with three complementary modules. First, a training-free Instance-Aware Offline Prototype Initialization aggregates multi-view augmented features via DINOv3 to construct discriminative instance prototypes, mitigating distractor confusion. Second, an Online Prototype Enhancement Tracker enhances prototypes online and integrates a confidence-aware Kalman filter for stable tracking under appearance and motion changes. Third, an Occlusion-Aware Trajectory Planner, trained on our new Planning-20k dataset, uses conditional diffusion to generate obstacle-avoiding paths for occlusion recovery. Experiments demonstrate OA-VAT achieves 0.93 average SR on UnrealCV (+2.2% vs. SOTA TrackVLA), 90.8% average CAR on real-world datasets (+12.1% vs. SOTA GC-VAT), and 81.6% TSR on a DJI Tello drone. Running at 35 FPS on an RTX 3090, it delivers robust, real-time performance for practical deployment.

Shiyu Tan, Zixuan Zhao, Hao Gao et al.

Boundary Representation (BRep) is the standard format for Computer-Aided Design (CAD), yet reconstructing high-quality BReps from single-view images remains challenging due to the complexity of topological constraints and operation sequences. We present Img2CADSeq, a multi-stage pipeline that overcomes these limitations by encoding CAD sequences into a three-level hierarchical codebook. Guided by an importance prioritization, this strategy values profiles over details, compressing long sequences into a stable discrete latent space. To bridge the modality gap, we leverage a coarse-to-fine point cloud intermediate, aligning 2D visual features with 3D CAD sequences via contrastive learning to condition a VQ-Diffusion model. Supported by newly introduced CAD-220K and PrintCAD datasets, our approach ensures robust industrial domain adaptation. Extensive experiments demonstrate that Img2CADSeq significantly outperforms state-of-the-art methods, producing standard STEP files that can be directly used in commercial CAD software.

Shaoyu Chen, Bo Jiang, Hao Gao et al.

Learning a human-like driving policy from large-scale driving demonstrations is promising, but the uncertainty and non-deterministic nature of planning make it challenging. In this work, to cope with the uncertainty problem, we propose VADv2, an end-to-end driving model based on probabilistic planning. VADv2 takes multi-view image sequences as input in a streaming manner, transforms sensor data into environmental token embeddings, outputs the probabilistic distribution of action, and samples one action to control the vehicle. Only with camera sensors, VADv2 achieves state-of-the-art closed-loop performance on the CARLA Town05 benchmark, significantly outperforming all existing methods. It runs stably in a fully end-to-end manner, even without the rule-based wrapper. Closed-loop demos are presented at https://hgao-cv.github.io/VADv2.

Yang Yan, Zhong Chen, Cai Xu et al.

Patient-reported outcomes (PROs) directly collected from cancer patients being treated with radiation therapy play a vital role in assisting clinicians in counseling patients regarding likely toxicities. Precise prediction and evaluation of symptoms or health status associated with PROs are fundamental to enhancing decision-making and planning for the required services and support as patients transition into survivorship. However, the raw PRO data collected from hospitals exhibits some intrinsic challenges such as incomplete item reports and imbalance patient toxicities. To the end, in this study, we explore various machine learning techniques to predict patient outcomes related to health status such as pain levels and sleep discomfort using PRO datasets from a cancer photon/proton therapy center. Specifically, we deploy six advanced machine learning classifiers -- Random Forest (RF), XGBoost, Gradient Boosting (GB), Support Vector Machine (SVM), Multi-Layer Perceptron with Bagging (MLP-Bagging), and Logistic Regression (LR) -- to tackle a multi-class imbalance classification problem across three prevalent cancer types: head and neck, prostate, and breast cancers. To address the class imbalance issue, we employ an oversampling strategy, adjusting the training set sample sizes through interpolations of in-class neighboring samples, thereby augmenting minority classes without deviating from the original skewed class distribution. Our experimental findings across multiple PRO datasets indicate that the RF and XGB methods achieve robust generalization performance, evidenced by weighted AUC and detailed confusion matrices, in categorizing outcomes as mild, intermediate, and severe post-radiation therapy. These results underscore the models' effectiveness and potential utility in clinical settings.

Jiateng Liu, Hao Gao, Jiu-Cheng Xie et al.

This paper presents an effective Gaussian management framework for high-fidelity scene reconstruction of appearance and geometry. Departing from recent Gaussian Splatting (GS) methods that rely on indiscriminate attribute assignment, our approach introduces a novel densification strategy called \emph{GauSep} that selectively activates Gaussian color or normal attributes. Together with a tailored rendering pipeline, termed \emph{Separate Rendering}, this strategy alleviates gradient conflicts arising from dual supervision and yields improved reconstruction quality. In addition, we develop \emph{GauRep}, an adaptive and integrated Gaussian representation that reduces redundancy both at the individual and global levels, effectively balancing model capacity and number of parameters. To provide reliable geometric supervision essential for effective management, we also introduce \emph{CoRe}, a novel surface reconstruction module that distills normal fields from the SDF branch to the Gaussian branch through a confidence mechanism. Notably, our management framework is model-agnostic and can be seamlessly incorporated into other architectures, simultaneously improving performance and reducing model size. Extensive experiments demonstrate that our approach achieves superior performance in reconstructing both appearance and geometry compared with state-of-the-art methods, while using significantly fewer parameters.

Jiu-Cheng Xie, Jun Yang, Wenqing Wang et al.

Face aging has received continuous research attention over the past two decades. Although previous works on this topic have achieved impressive success, two longstanding problems remain unsettled: 1) generating diverse and plausible facial aging patterns at the target age stage; 2) measuring the rationality of identity variation between the original portrait and its syntheses with age progression or regression. In this paper, we introduce ${\rm{DLAT}}^{\boldsymbol{+}}$ to realize Diverse and Lifespan Age Transformation on human faces, where the diversity jointly manifests in the transformation of facial textures and shapes. Apart from the diversity mechanism embedded in the model, multiple consistency restrictions are leveraged to keep it away from counterfactual aging syntheses. Moreover, we propose a new metric to assess the rationality of Identity Deviation under Age Gaps (IDAG) between the input face and its series of age-transformed generations, which is based on statistical laws summarized from plenty of genuine face-aging data. Extensive experimental results demonstrate the uniqueness and effectiveness of our method in synthesizing diverse and perceptually reasonable faces across the whole lifetime.

Jiangli Shao, Yongqing Wang, Hao Gao et al.

Nowadays, users are encouraged to activate across multiple online social networks simultaneously. Anchor link prediction, which aims to reveal the correspondence among different accounts of the same user across networks, has been regarded as a fundamental problem for user profiling, marketing, cybersecurity, and recommendation. Existing methods mainly address the prediction problem by utilizing profile, content, or structural features of users in symmetric ways. However, encouraged by online services, users would also post asymmetric information across networks, such as geo-locations and texts. It leads to an emerged challenge in aligning users with asymmetric information across networks. Instead of similarity evaluation applied in previous works, we formalize correlation between geo-locations and texts and propose a novel anchor link prediction framework for matching users across networks. Moreover, our model can alleviate the label scarcity problem by introducing external data. Experimental results on real-world datasets show that our approach outperforms existing methods and achieves state-of-the-art results.

Hao Gao, Yongqing Wang, Shanshan Lyu et al.

Nowadays online users prefer to join multiple social media for the purpose of socialized online service. The problem \textit{anchor link prediction} is formalized to link user data with the common ground on user profile, content and network structure across social networks. Most of the traditional works concentrated on learning matching function with explicit or implicit features on observed user data. However, the low quality of observed user data confuses the judgment on anchor links, resulting in the matching collision problem in practice. In this paper, we explore local structure consistency and then construct a matching graph in order to circumvent matching collisions. Furthermore, we propose graph convolution networks with mini-batch strategy, efficiently solving anchor link prediction on matching graph. The experimental results on three real application scenarios show the great potentials of our proposed method in both prediction accuracy and efficiency. In addition, the visualization of learned embeddings provides us a qualitative way to understand the inference of anchor links on the matching graph.

Shen-Yi Zhao, Hao Gao, Wu-Jun Li

Stochastic gradient decent~(SGD) and its variants, including some accelerated variants, have become popular for training in machine learning. However, in all existing SGD and its variants, the sample size in each iteration~(epoch) of training is the same as the size of the full training set. In this paper, we propose a new method, called \underline{ada}ptive \underline{s}ample \underline{s}election~(ADASS), for training acceleration. During different epoches of training, ADASS only need to visit different training subsets which are adaptively selected from the full training set according to the Lipschitz constants of the loss functions on samples. It means that in ADASS the sample size in each epoch of training can be smaller than the size of the full training set, by discarding some samples. ADASS can be seamlessly integrated with existing optimization methods, such as SGD and momentum SGD, for training acceleration. Theoretical results show that the learning accuracy of ADASS is comparable to that of counterparts with full training set. Furthermore, empirical results on both shallow models and deep models also show that ADASS can accelerate the training process of existing methods without sacrificing accuracy.

Shen-Yi Zhao, Hao Gao, Wu-Jun Li

Distributed stochastic gradient descent~(DSGD) has been widely used for optimizing large-scale machine learning models, including both convex and non-convex models. With the rapid growth of model size, huge communication cost has been the bottleneck of traditional DSGD. Recently, many communication compression methods have been proposed. Memory-based distributed stochastic gradient descent~(M-DSGD) is one of the efficient methods since each worker communicates a sparse vector in each iteration so that the communication cost is small. Recent works propose the convergence rate of M-DSGD when it adopts vanilla SGD. However, there is still a lack of convergence theory for M-DSGD when it adopts momentum SGD. In this paper, we propose a universal convergence analysis for M-DSGD by introducing \emph{transformation equation}. The transformation equation describes the relation between traditional DSGD and M-DSGD so that we can transform M-DSGD to its corresponding DSGD. Hence we get the convergence rate of M-DSGD with momentum for both convex and non-convex problems. Furthermore, we combine M-DSGD and stagewise learning that the learning rate of M-DSGD in each stage is a constant and is decreased by stage, instead of iteration. Using the transformation equation, we propose the convergence rate of stagewise M-DSGD which bridges the gap between theory and practice.

Chang-Wei Shi, Shen-Yi Zhao, Yin-Peng Xie et al.

With the rapid growth of data, distributed momentum stochastic gradient descent~(DMSGD) has been widely used in distributed learning, especially for training large-scale deep models. Due to the latency and limited bandwidth of the network, communication has become the bottleneck of distributed learning. Communication compression with sparsified gradient, abbreviated as \emph{sparse communication}, has been widely employed to reduce communication cost. All existing works about sparse communication in DMSGD employ local momentum, in which the momentum only accumulates stochastic gradients computed by each worker locally. In this paper, we propose a novel method, called \emph{\underline{g}}lobal \emph{\underline{m}}omentum \emph{\underline{c}}ompression~(GMC), for sparse communication. Different from existing works that utilize local momentum, GMC utilizes global momentum. Furthermore, to enhance the convergence performance when using more aggressive sparsification compressors (e.g., RBGS), we extend GMC to GMC+. We theoretically prove the convergence of GMC and GMC+. To the best of our knowledge, this is the first work that introduces global momentum for sparse communication in distributed learning. Empirical results demonstrate that, compared with the local momentum counterparts, our GMC and GMC+ can achieve higher test accuracy and exhibit faster convergence, especially under non-IID data distribution.

Shen-Yi Zhao, Hao Gao, Wu-Jun Li

Due to its efficiency and ease to implement, stochastic gradient descent (SGD) has been widely used in machine learning. In particular, SGD is one of the most popular optimization methods for distributed learning. Recently, quantized SGD (QSGD), which adopts quantization to reduce the communication cost in SGD-based distributed learning, has attracted much attention. Although several QSGD methods have been proposed, some of them are heuristic without theoretical guarantee, and others have high quantization variance which makes the convergence become slow. In this paper, we propose a new method, called Quantized Epoch-SGD (QESGD), for communication-efficient distributed learning. QESGD compresses (quantizes) the parameter with variance reduction, so that it can get almost the same performance as that of SGD with less communication cost. QESGD is implemented on the Parameter Server framework, and empirical results on distributed deep learning show that QESGD can outperform other state-of-the-art quantization methods to achieve the best performance.

Gong-Duo Zhang, Shen-Yi Zhao, Hao Gao et al.

Linear classification has been widely used in many high-dimensional applications like text classification. To perform linear classification for large-scale tasks, we often need to design distributed learning methods on a cluster of multiple machines. In this paper, we propose a new distributed learning method, called feature-distributed stochastic variance reduced gradient (FD-SVRG) for high-dimensional linear classification. Unlike most existing distributed learning methods which are instance-distributed, FD-SVRG is feature-distributed. FD-SVRG has lower communication cost than other instance-distributed methods when the data dimensionality is larger than the number of data instances. Experimental results on real data demonstrate that FD-SVRG can outperform other state-of-the-art distributed methods for high-dimensional linear classification in terms of both communication cost and wall-clock time, when the dimensionality is larger than the number of instances in training data.

Xiaodong Cun, Feng Xu, Chi-Man Pun et al.

Researches in novel viewpoint synthesis majorly focus on interpolation from multi-view input images. In this paper, we focus on a more challenging and ill-posed problem that is to synthesize novel viewpoints from one single input image. To achieve this goal, we propose a novel deep learning-based technique. We design a full resolution network that extracts local image features with the same resolution of the input, which contributes to derive high resolution and prevent blurry artifacts in the final synthesized images. We also involve a pre-trained depth estimation network into our system, and thus 3D information is able to be utilized to infer the flow field between the input and the target image. Since the depth network is trained by depth order information between arbitrary pairs of points in the scene, global image features are also involved into our system. Finally, a synthesis layer is used to not only warp the observed pixels to the desired positions but also hallucinate the missing pixels with recorded pixels. Experiments show that our technique performs well on images of various scenes, and outperforms the state-of-the-art techniques.