Wentao Zhu

Xiaoxuan Ma, Jiajun Su, Chunyu Wang et al.

Inspired by the success of volumetric 3D pose estimation, some recent human mesh estimators propose to estimate 3D skeletons as intermediate representations, from which, the dense 3D meshes are regressed by exploiting the mesh topology. However, body shape information is lost in extracting skeletons, leading to mediocre performance. The advanced motion capture systems solve the problem by placing dense physical markers on the body surface, which allows to extract realistic meshes from their non-rigid motions. However, they cannot be applied to wild images without markers. In this work, we present an intermediate representation, named virtual markers, which learns 64 landmark keypoints on the body surface based on the large-scale mocap data in a generative style, mimicking the effects of physical markers. The virtual markers can be accurately detected from wild images and can reconstruct the intact meshes with realistic shapes by simple interpolation. Our approach outperforms the state-of-the-art methods on three datasets. In particular, it surpasses the existing methods by a notable margin on the SURREAL dataset, which has diverse body shapes. Code is available at https://github.com/ShirleyMaxx/VirtualMarker

M. Jorge Cardoso, Wenqi Li, Richard Brown et al.

Artificial Intelligence (AI) is having a tremendous impact across most areas of science. Applications of AI in healthcare have the potential to improve our ability to detect, diagnose, prognose, and intervene on human disease. For AI models to be used clinically, they need to be made safe, reproducible and robust, and the underlying software framework must be aware of the particularities (e.g. geometry, physiology, physics) of medical data being processed. This work introduces MONAI, a freely available, community-supported, and consortium-led PyTorch-based framework for deep learning in healthcare. MONAI extends PyTorch to support medical data, with a particular focus on imaging, and provide purpose-specific AI model architectures, transformations and utilities that streamline the development and deployment of medical AI models. MONAI follows best practices for software-development, providing an easy-to-use, robust, well-documented, and well-tested software framework. MONAI preserves the simple, additive, and compositional approach of its underlying PyTorch libraries. MONAI is being used by and receiving contributions from research, clinical and industrial teams from around the world, who are pursuing applications spanning nearly every aspect of healthcare.

Hao Zhu, Wayne Wu, Wentao Zhu et al.

Large-scale datasets have played indispensable roles in the recent success of face generation/editing and significantly facilitated the advances of emerging research fields. However, the academic community still lacks a video dataset with diverse facial attribute annotations, which is crucial for the research on face-related videos. In this work, we propose a large-scale, high-quality, and diverse video dataset with rich facial attribute annotations, named the High-Quality Celebrity Video Dataset (CelebV-HQ). CelebV-HQ contains 35,666 video clips with the resolution of 512x512 at least, involving 15,653 identities. All clips are labeled manually with 83 facial attributes, covering appearance, action, and emotion. We conduct a comprehensive analysis in terms of age, ethnicity, brightness stability, motion smoothness, head pose diversity, and data quality to demonstrate the diversity and temporal coherence of CelebV-HQ. Besides, its versatility and potential are validated on two representative tasks, i.e., unconditional video generation and video facial attribute editing. Furthermore, we envision the future potential of CelebV-HQ, as well as the new opportunities and challenges it would bring to related research directions. Data, code, and models are publicly available. Project page: https://celebv-hq.github.io.

Shiqiang Zhu, Ting Yu, Tao Xu et al.

Computing is a critical driving force in the development of human civilization. In recent years, we have witnessed the emergence of intelligent computing, a new computing paradigm that is reshaping traditional computing and promoting digital revolution in the era of big data, artificial intelligence and internet-of-things with new computing theories, architectures, methods, systems, and applications. Intelligent computing has greatly broadened the scope of computing, extending it from traditional computing on data to increasingly diverse computing paradigms such as perceptual intelligence, cognitive intelligence, autonomous intelligence, and human-computer fusion intelligence. Intelligence and computing have undergone paths of different evolution and development for a long time but have become increasingly intertwined in recent years: intelligent computing is not only intelligence-oriented but also intelligence-driven. Such cross-fertilization has prompted the emergence and rapid advancement of intelligent computing. Intelligent computing is still in its infancy and an abundance of innovations in the theories, systems, and applications of intelligent computing are expected to occur soon. We present the first comprehensive survey of literature on intelligent computing, covering its theory fundamentals, the technological fusion of intelligence and computing, important applications, challenges, and future perspectives. We believe that this survey is highly timely and will provide a comprehensive reference and cast valuable insights into intelligent computing for academic and industrial researchers and practitioners.

Jianxu Shangguan, Jing Xu, Hang Ye et al.

Creating lifelike digital humans with genuine social intelligence requires unifying cognitive reasoning and multimodal generation within a coherent framework. Current approaches treat these as separate tasks: Large Language Models excel at dialogue but lack embodied expression, while diffusion-based talking head models achieve visual fidelity but ignore social cognition. To bridge this gap, we propose a closed-loop dual-agent framework integrating perception, social reasoning, and expression into a continuous interaction cycle. The perception module analyzes partners' multimodal behaviors from video, while the social reasoning module infers hidden mental states through Theory of Mind and selects responses via an ensemble mechanism. The expression module then generates emotion-controllable dual-agent videos synthesizing both speaker speech and expression alongside listener reactive behaviors, capturing bidirectional dynamics absent in prior work. We construct a hierarchical Persona-Scenario dataset with psychologically grounded personas and private social goals to support evaluation under information asymmetry. Experiments on this dataset demonstrate competitive or superior performance on both dialogue quality and video generation metrics. Notably, our method surpasses even the full-information Script mode on key dialogue quality dimensions, suggesting that explicit mental state inference under uncertainty can elicit more thoughtful dialogue than unrestricted information access.

Wentao Zhu, Yufang Huang, Xiufeng Xie et al.

The short-form videos have explosive popularity and have dominated the new social media trends. Prevailing short-video platforms,~\textit{e.g.}, Kuaishou (Kwai), TikTok, Instagram Reels, and YouTube Shorts, have changed the way we consume and create content. For video content creation and understanding, the shot boundary detection (SBD) is one of the most essential components in various scenarios. In this work, we release a new public Short video sHot bOundary deTection dataset, named SHOT, consisting of 853 complete short videos and 11,606 shot annotations, with 2,716 high quality shot boundary annotations in 200 test videos. Leveraging this new data wealth, we propose to optimize the model design for video SBD, by conducting neural architecture search in a search space encapsulating various advanced 3D ConvNets and Transformers. Our proposed approach, named AutoShot, achieves higher F1 scores than previous state-of-the-art approaches, e.g., outperforming TransNetV2 by 4.2%, when being derived and evaluated on our newly constructed SHOT dataset. Moreover, to validate the generalizability of the AutoShot architecture, we directly evaluate it on another three public datasets: ClipShots, BBC and RAI, and the F1 scores of AutoShot outperform previous state-of-the-art approaches by 1.1%, 0.9% and 1.2%, respectively. The SHOT dataset and code can be found in https://github.com/wentaozhu/AutoShot.git .

Jianghao Wu, Guotai Wang, Ran Gu et al.

Domain Adaptation (DA) is important for deep learning-based medical image segmentation models to deal with testing images from a new target domain. As the source-domain data are usually unavailable when a trained model is deployed at a new center, Source-Free Domain Adaptation (SFDA) is appealing for data and annotation-efficient adaptation to the target domain. However, existing SFDA methods have a limited performance due to lack of sufficient supervision with source-domain images unavailable and target-domain images unlabeled. We propose a novel Uncertainty-aware Pseudo Label guided (UPL) SFDA method for medical image segmentation. Specifically, we propose Target Domain Growing (TDG) to enhance the diversity of predictions in the target domain by duplicating the pre-trained model's prediction head multiple times with perturbations. The different predictions in these duplicated heads are used to obtain pseudo labels for unlabeled target-domain images and their uncertainty to identify reliable pseudo labels. We also propose a Twice Forward pass Supervision (TFS) strategy that uses reliable pseudo labels obtained in one forward pass to supervise predictions in the next forward pass. The adaptation is further regularized by a mean prediction-based entropy minimization term that encourages confident and consistent results in different prediction heads. UPL-SFDA was validated with a multi-site heart MRI segmentation dataset, a cross-modality fetal brain segmentation dataset, and a 3D fetal tissue segmentation dataset. It improved the average Dice by 5.54, 5.01 and 6.89 percentage points for the three tasks compared with the baseline, respectively, and outperformed several state-of-the-art SFDA methods.

Jue Wang, Wentao Zhu, Pichao Wang et al.

Effective modeling of complex spatiotemporal dependencies in long-form videos remains an open problem. The recently proposed Structured State-Space Sequence (S4) model with its linear complexity offers a promising direction in this space. However, we demonstrate that treating all image-tokens equally as done by S4 model can adversely affect its efficiency and accuracy. To address this limitation, we present a novel Selective S4 (i.e., S5) model that employs a lightweight mask generator to adaptively select informative image tokens resulting in more efficient and accurate modeling of long-term spatiotemporal dependencies in videos. Unlike previous mask-based token reduction methods used in transformers, our S5 model avoids the dense self-attention calculation by making use of the guidance of the momentum-updated S4 model. This enables our model to efficiently discard less informative tokens and adapt to various long-form video understanding tasks more effectively. However, as is the case for most token reduction methods, the informative image tokens could be dropped incorrectly. To improve the robustness and the temporal horizon of our model, we propose a novel long-short masked contrastive learning (LSMCL) approach that enables our model to predict longer temporal context using shorter input videos. We present extensive comparative results using three challenging long-form video understanding datasets (LVU, COIN and Breakfast), demonstrating that our approach consistently outperforms the previous state-of-the-art S4 model by up to 9.6% accuracy while reducing its memory footprint by 23%.

Wentao Zhu, Xiaoxuan Ma, Zhaoyang Liu et al.

We present a unified perspective on tackling various human-centric video tasks by learning human motion representations from large-scale and heterogeneous data resources. Specifically, we propose a pretraining stage in which a motion encoder is trained to recover the underlying 3D motion from noisy partial 2D observations. The motion representations acquired in this way incorporate geometric, kinematic, and physical knowledge about human motion, which can be easily transferred to multiple downstream tasks. We implement the motion encoder with a Dual-stream Spatio-temporal Transformer (DSTformer) neural network. It could capture long-range spatio-temporal relationships among the skeletal joints comprehensively and adaptively, exemplified by the lowest 3D pose estimation error so far when trained from scratch. Furthermore, our proposed framework achieves state-of-the-art performance on all three downstream tasks by simply finetuning the pretrained motion encoder with a simple regression head (1-2 layers), which demonstrates the versatility of the learned motion representations. Code and models are available at https://motionbert.github.io/

Hang Ye, Wentao Zhu, Chunyu Wang et al.

While the voxel-based methods have achieved promising results for multi-person 3D pose estimation from multi-cameras, they suffer from heavy computation burdens, especially for large scenes. We present Faster VoxelPose to address the challenge by re-projecting the feature volume to the three two-dimensional coordinate planes and estimating X, Y, Z coordinates from them separately. To that end, we first localize each person by a 3D bounding box by estimating a 2D box and its height based on the volume features projected to the xy-plane and z-axis, respectively. Then for each person, we estimate partial joint coordinates from the three coordinate planes separately which are then fused to obtain the final 3D pose. The method is free from costly 3D-CNNs and improves the speed of VoxelPose by ten times and meanwhile achieves competitive accuracy as the state-of-the-art methods, proving its potential in real-time applications.

Wentao Zhu, Yuan Jin, Gege Ma et al.

The accurate diagnosis on pathological subtypes for lung cancer is of significant importance for the follow-up treatments and prognosis managements. In this paper, we propose self-generating hybrid feature network (SGHF-Net) for accurately classifying lung cancer subtypes on computed tomography (CT) images. Inspired by studies stating that cross-scale associations exist in the image patterns between the same case's CT images and its pathological images, we innovatively developed a pathological feature synthetic module (PFSM), which quantitatively maps cross-modality associations through deep neural networks, to derive the "gold standard" information contained in the corresponding pathological images from CT images. Additionally, we designed a radiological feature extraction module (RFEM) to directly acquire CT image information and integrated it with the pathological priors under an effective feature fusion framework, enabling the entire classification model to generate more indicative and specific pathologically related features and eventually output more accurate predictions. The superiority of the proposed model lies in its ability to self-generate hybrid features that contain multi-modality image information based on a single-modality input. To evaluate the effectiveness, adaptability, and generalization ability of our model, we performed extensive experiments on a large-scale multi-center dataset (i.e., 829 cases from three hospitals) to compare our model and a series of state-of-the-art (SOTA) classification models. The experimental results demonstrated the superiority of our model for lung cancer subtypes classification with significant accuracy improvements in terms of accuracy (ACC), area under the curve (AUC), and F1 score.

Runqi Wang, Yuxiang Bao, Baochang Zhang et al.

Humans can continuously learn new knowledge. However, machine learning models suffer from drastic dropping in performance on previous tasks after learning new tasks. Cognitive science points out that the competition of similar knowledge is an important cause of forgetting. In this paper, we design a paradigm for lifelong learning based on meta-learning and associative mechanism of the brain. It tackles the problem from two aspects: extracting knowledge and memorizing knowledge. First, we disrupt the sample's background distribution through a background attack, which strengthens the model to extract the key features of each task. Second, according to the similarity between incremental knowledge and base knowledge, we design an adaptive fusion of incremental knowledge, which helps the model allocate capacity to the knowledge of different difficulties. It is theoretically analyzed that the proposed learning paradigm can make the models of different tasks converge to the same optimum. The proposed method is validated on the MNIST, CIFAR100, CUB200 and ImageNet100 datasets.

Hai Ci, Mingdong Wu, Wentao Zhu et al.

Learning 3D human pose prior is essential to human-centered AI. Here, we present GFPose, a versatile framework to model plausible 3D human poses for various applications. At the core of GFPose is a time-dependent score network, which estimates the gradient on each body joint and progressively denoises the perturbed 3D human pose to match a given task specification. During the denoising process, GFPose implicitly incorporates pose priors in gradients and unifies various discriminative and generative tasks in an elegant framework. Despite the simplicity, GFPose demonstrates great potential in several downstream tasks. Our experiments empirically show that 1) as a multi-hypothesis pose estimator, GFPose outperforms existing SOTAs by 20% on Human3.6M dataset. 2) as a single-hypothesis pose estimator, GFPose achieves comparable results to deterministic SOTAs, even with a vanilla backbone. 3) GFPose is able to produce diverse and realistic samples in pose denoising, completion and generation tasks. Project page https://sites.google.com/view/gfpose/

Wentao Zhu, Xiaoxuan Ma, Dongwoo Ro et al.

Human motion generation aims to generate natural human pose sequences and shows immense potential for real-world applications. Substantial progress has been made recently in motion data collection technologies and generation methods, laying the foundation for increasing interest in human motion generation. Most research within this field focuses on generating human motions based on conditional signals, such as text, audio, and scene contexts. While significant advancements have been made in recent years, the task continues to pose challenges due to the intricate nature of human motion and its implicit relationship with conditional signals. In this survey, we present a comprehensive literature review of human motion generation, which, to the best of our knowledge, is the first of its kind in this field. We begin by introducing the background of human motion and generative models, followed by an examination of representative methods for three mainstream sub-tasks: text-conditioned, audio-conditioned, and scene-conditioned human motion generation. Additionally, we provide an overview of common datasets and evaluation metrics. Lastly, we discuss open problems and outline potential future research directions. We hope that this survey could provide the community with a comprehensive glimpse of this rapidly evolving field and inspire novel ideas that address the outstanding challenges.

Runqi Wang, Linlin Yang, Baochang Zhang et al.

Research on the generalization ability of deep neural networks (DNNs) has recently attracted a great deal of attention. However, due to their complex architectures and large numbers of parameters, measuring the generalization ability of specific DNN models remains an open challenge. In this paper, we propose to use multiple factors to measure and rank the relative generalization of DNNs based on a new concept of confidence dimension (CD). Furthermore, we provide a feasible framework in our CD to theoretically calculate the upper bound of generalization based on the conventional Vapnik-Chervonenk dimension (VC-dimension) and Hoeffding's inequality. Experimental results on image classification and object detection demonstrate that our CD can reflect the relative generalization ability for different DNNs. In addition to full-precision DNNs, we also analyze the generalization ability of binary neural networks (BNNs), whose generalization ability remains an unsolved problem. Our CD yields a consistent and reliable measure and ranking for both full-precision DNNs and BNNs on all the tasks.

Yukang Lin, Jiahao Shao, Shuoran Jiang et al.

Watermarking acts as a critical safeguard in text generated by Large Language Models (LLMs). By embedding identifiable signals into model outputs, watermarking enables reliable attribution and enhances the security of machine-generated content. Existing approaches typically embed signals by manipulating token generation probabilities. Despite their effectiveness, these methods inherently face a trade-off between detectability and text quality: the signal strength and randomness required for robust watermarking tend to degrade the performance of downstream tasks. In this paper, we design a novel embedding scheme that controls seed pools to facilitate diverse parallel generation of watermarked text. Based on that scheme, we propose WaterSearch, a sentence-level, search-based watermarking framework adaptable to a wide range of existing methods. WaterSearch enhances text quality by jointly optimizing two key aspects: 1) distribution fidelity and 2) watermark signal characteristics. Furthermore, WaterSearch is complemented by a sentence-level detection method with strong attack robustness. We evaluate our method on three popular LLMs across ten diverse tasks. Extensive experiments demonstrate that our method achieves an average performance improvement of 51.01\% over state-of-the-art baselines at a watermark detectability strength of 95\%. In challenging scenarios such as short text generation and low-entropy output generation, our method yields performance gains of 47.78\% and 36.47\%, respectively. Moreover, under different attack senarios including insertion, synonym substitution and paraphrase attasks, WaterSearch maintains high detectability, further validating its robust anti-attack capabilities. Our code is available at \href{https://github.com/Yukang-Lin/WaterSearch}{https://github.com/Yukang-Lin/WaterSearch}.

Yuanfeng Wu, Shaojie Li, Zhiqiang Du et al.

Whole slide image (WSI) processing is becoming part of the key components of standard clinical diagnosis for various diseases. However, the direct application of conventional image processing algorithms to WSI faces certain obstacles because of WSIs' distinct property: the super-high resolution. The performance of most WSI-related tasks relies on the efficacy of the backbone which extracts WSI patch feature representations. Hence, we proposed BROW, a foundation model for extracting better feature representations for WSIs, which can be conveniently adapted to downstream tasks without or with slight fine-tuning. The model takes transformer architecture, pretrained using self-distillation framework. To improve model's robustness, techniques such as patch shuffling have been employed. Additionally, the model leverages the unique properties of WSIs, utilizing WSI's multi-scale pyramid to incorporate an additional global view, thereby further enhancing its performance. We used both private and public data to make up a large pretraining dataset, containing more than 11000 slides, over 180M extracted patches, encompassing WSIs related to various organs and tissues. To assess the effectiveness of \ourmodel, we run a wide range of downstream tasks, including slide-level subtyping, patch-level classification and nuclei instance segmentation. The results confirmed the efficacy, robustness and good generalization ability of the proposed model. This substantiates its potential as foundation model for WSI feature extraction and highlights promising prospects for its application in WSI processing.

Wentao Zhu, Mohamed Omar

Audio event has a hierarchical architecture in both time and frequency and can be grouped together to construct more abstract semantic audio classes. In this work, we develop a multiscale audio spectrogram Transformer (MAST) that employs hierarchical representation learning for efficient audio classification. Specifically, MAST employs one-dimensional (and two-dimensional) pooling operators along the time (and frequency domains) in different stages, and progressively reduces the number of tokens and increases the feature dimensions. MAST significantly outperforms AST~\cite{gong2021ast} by 22.2\%, 4.4\% and 4.7\% on Kinetics-Sounds, Epic-Kitchens-100 and VGGSound in terms of the top-1 accuracy without external training data. On the downloaded AudioSet dataset, which has over 20\% missing audios, MAST also achieves slightly better accuracy than AST. In addition, MAST is 5x more efficient in terms of multiply-accumulates (MACs) with 42\% reduction in the number of parameters compared to AST. Through clustering metrics and visualizations, we demonstrate that the proposed MAST can learn semantically more separable feature representations from audio signals.

Gaoren Lin, Huangxuan Zhao, Yuan Xiong et al.

Text-guided medical segmentation enhances segmentation accuracy by utilizing clinical reports as auxiliary information. However, existing methods typically rely on unaligned image and text encoders, which necessitate complex interaction modules for multimodal fusion. While CLIP provides a pre-aligned multimodal feature space, its direct application to medical imaging is limited by three main issues: insufficient preservation of fine-grained anatomical structures, inadequate modeling of complex clinical descriptions, and domain-specific semantic misalignment. To tackle these challenges, we propose TGC-Net, a CLIP-based framework focusing on parameter-efficient, task-specific adaptations. Specifically, it incorporates a Semantic-Structural Synergy Encoder (SSE) that augments CLIP's ViT with a CNN branch for multi-scale structural refinement, a Domain-Augmented Text Encoder (DATE) that injects large-language-model-derived medical knowledge, and a Vision-Language Calibration Module (VLCM) that refines cross-modal correspondence in a unified feature space. Experiments on five datasets across chest X-ray and thoracic CT modalities demonstrate that TGC-Net achieves state-of-the-art performance with substantially fewer trainable parameters, including notable Dice gains on challenging benchmarks.

Xiaoxuan Ma, Stephan P. Kaufhold, Jiajun Su et al.

Understanding the behavior of non-human primates is crucial for improving animal welfare, modeling social behavior, and gaining insights into distinctively human and phylogenetically shared behaviors. However, the lack of datasets on non-human primate behavior hinders in-depth exploration of primate social interactions, posing challenges to research on our closest living relatives. To address these limitations, we present ChimpACT, a comprehensive dataset for quantifying the longitudinal behavior and social relations of chimpanzees within a social group. Spanning from 2015 to 2018, ChimpACT features videos of a group of over 20 chimpanzees residing at the Leipzig Zoo, Germany, with a particular focus on documenting the developmental trajectory of one young male, Azibo. ChimpACT is both comprehensive and challenging, consisting of 163 videos with a cumulative 160,500 frames, each richly annotated with detection, identification, pose estimation, and fine-grained spatiotemporal behavior labels. We benchmark representative methods of three tracks on ChimpACT: (i) tracking and identification, (ii) pose estimation, and (iii) spatiotemporal action detection of the chimpanzees. Our experiments reveal that ChimpACT offers ample opportunities for both devising new methods and adapting existing ones to solve fundamental computer vision tasks applied to chimpanzee groups, such as detection, pose estimation, and behavior analysis, ultimately deepening our comprehension of communication and sociality in non-human primates.

Wentao Zhu, Jason Qin, Yuke Lou et al.

Humans exhibit a remarkable capacity for anticipating the actions of others and planning their own actions accordingly. In this study, we strive to replicate this ability by addressing the social motion prediction problem. We introduce a new benchmark, a novel formulation, and a cognition-inspired framework. We present Wusi, a 3D multi-person motion dataset under the context of team sports, which features intense and strategic human interactions and diverse pose distributions. By reformulating the problem from a multi-agent reinforcement learning perspective, we incorporate behavioral cloning and generative adversarial imitation learning to boost learning efficiency and generalization. Furthermore, we take into account the cognitive aspects of the human social action planning process and develop a cognitive hierarchy framework to predict strategic human social interactions. We conduct comprehensive experiments to validate the effectiveness of our proposed dataset and approach. Code and data are available at https://walter0807.github.io/Social-CH/.

Haoru Wang, Wentao Zhu, Luyi Miao et al.

Human motion generation is a critical task with a wide range of applications. Achieving high realism in generated motions requires naturalness, smoothness, and plausibility. Despite rapid advancements in the field, current generation methods often fall short of these goals. Furthermore, existing evaluation metrics typically rely on ground-truth-based errors, simple heuristics, or distribution distances, which do not align well with human perceptions of motion quality. In this work, we propose a data-driven approach to bridge this gap by introducing a large-scale human perceptual evaluation dataset, MotionPercept, and a human motion critic model, MotionCritic, that capture human perceptual preferences. Our critic model offers a more accurate metric for assessing motion quality and could be readily integrated into the motion generation pipeline to enhance generation quality. Extensive experiments demonstrate the effectiveness of our approach in both evaluating and improving the quality of generated human motions by aligning with human perceptions. Code and data are publicly available at https://motioncritic.github.io/.

Yangfan Ni, Duo Zhang, Gege Ma et al.

Accurate reorientation and segmentation of the left ventricular (LV) is essential for the quantitative analysis of myocardial perfusion imaging (MPI), in which one critical step is to reorient the reconstructed transaxial nuclear cardiac images into standard short-axis slices for subsequent image processing. Small-scale LV myocardium (LV-MY) region detection and the diverse cardiac structures of individual patients pose challenges to LV segmentation operation. To mitigate these issues, we propose an end-to-end model, named as multi-scale spatial transformer UNet (MS-ST-UNet), that involves the multi-scale spatial transformer network (MSSTN) and multi-scale UNet (MSUNet) modules to perform simultaneous reorientation and segmentation of LV region from nuclear cardiac images. The proposed method is trained and tested using two different nuclear cardiac image modalities: 13N-ammonia PET and 99mTc-sestamibi SPECT. We use a multi-scale strategy to generate and extract image features with different scales. Our experimental results demonstrate that the proposed method significantly improves the reorientation and segmentation performance. This joint learning framework promotes mutual enhancement between reorientation and segmentation tasks, leading to cutting edge performance and an efficient image processing workflow. The proposed end-to-end deep network has the potential to reduce the burden of manual delineation for cardiac images, thereby providing multimodal quantitative analysis assistance for physicists.

Rui Yang, Matthew Yu Heng Wong, Huitao Li et al.

The rapid growth of medical knowledge and increasing complexity of clinical practice pose challenges. In this context, large language models (LLMs) have demonstrated value; however, inherent limitations remain. Retrieval-augmented generation (RAG) technologies show potential to enhance their clinical applicability. This study reviewed RAG applications in medicine. We found that research primarily relied on publicly available data, with limited application in private data. For retrieval, approaches commonly relied on English-centric embedding models, while LLMs were mostly generic, with limited use of medical-specific LLMs. For evaluation, automated metrics evaluated generation quality and task performance, whereas human evaluation focused on accuracy, completeness, relevance, and fluency, with insufficient attention to bias and safety. RAG applications were concentrated on question answering, report generation, text summarization, and information extraction. Overall, medical RAG remains at an early stage, requiring advances in clinical validation, cross-linguistic adaptation, and support for low-resource settings to enable trustworthy and responsible global use.

Yuan Jin, Gege Ma, Geng Chen et al.

The accurate diagnosis of pathological subtypes of lung cancer is of paramount importance for follow-up treatments and prognosis managements. Assessment methods utilizing deep learning technologies have introduced novel approaches for clinical diagnosis. However, the majority of existing models rely solely on single-modality image input, leading to limited diagnostic accuracy. To this end, we propose a novel deep learning network designed to accurately classify lung cancer subtype with multi-dimensional and multi-modality images, i.e., CT and pathological images. The strength of the proposed model lies in its ability to dynamically process both paired CT-pathological image sets as well as independent CT image sets, and consequently optimize the pathology-related feature extractions from CT images. This adaptive learning approach enhances the flexibility in processing multi-dimensional and multi-modality datasets and results in performance elevating in the model testing phase. We also develop a contrastive constraint module, which quantitatively maps the cross-modality associations through network training, and thereby helps to explore the "gold standard" pathological information from the corresponding CT scans. To evaluate the effectiveness, adaptability, and generalization ability of our model, we conducted extensive experiments on a large-scale multi-center dataset and compared our model with a series of state-of-the-art classification models. The experimental results demonstrated the superiority of our model for lung cancer subtype classification, showcasing significant improvements in accuracy metrics such as ACC, AUC, and F1-score.

Shikun Ban, Juling Fan, Xiaoxuan Ma et al.

Estimating robot pose from RGB images is a crucial problem in computer vision and robotics. While previous methods have achieved promising performance, most of them presume full knowledge of robot internal states, e.g. ground-truth robot joint angles. However, this assumption is not always valid in practical situations. In real-world applications such as multi-robot collaboration or human-robot interaction, the robot joint states might not be shared or could be unreliable. On the other hand, existing approaches that estimate robot pose without joint state priors suffer from heavy computation burdens and thus cannot support real-time applications. This work introduces an efficient framework for real-time robot pose estimation from RGB images without requiring known robot states. Our method estimates camera-to-robot rotation, robot state parameters, keypoint locations, and root depth, employing a neural network module for each task to facilitate learning and sim-to-real transfer. Notably, it achieves inference in a single feed-forward pass without iterative optimization. Our approach offers a 12-time speed increase with state-of-the-art accuracy, enabling real-time holistic robot pose estimation for the first time. Code and models are available at https://github.com/Oliverbansk/Holistic-Robot-Pose-Estimation.

Xiaosong Wang, Xiaofan Zhang, Guotai Wang et al.

The emerging trend of advancing generalist artificial intelligence, such as GPTv4 and Gemini, has reshaped the landscape of research (academia and industry) in machine learning and many other research areas. However, domain-specific applications of such foundation models (e.g., in medicine) remain untouched or often at their very early stages. It will require an individual set of transfer learning and model adaptation techniques by further expanding and injecting these models with domain knowledge and data. The development of such technologies could be largely accelerated if the bundle of data, algorithms, and pre-trained foundation models were gathered together and open-sourced in an organized manner. In this work, we present OpenMEDLab, an open-source platform for multi-modality foundation models. It encapsulates not only solutions of pioneering attempts in prompting and fine-tuning large language and vision models for frontline clinical and bioinformatic applications but also building domain-specific foundation models with large-scale multi-modal medical data. Importantly, it opens access to a group of pre-trained foundation models for various medical image modalities, clinical text, protein engineering, etc. Inspiring and competitive results are also demonstrated for each collected approach and model in a variety of benchmarks for downstream tasks. We welcome researchers in the field of medical artificial intelligence to continuously contribute cutting-edge methods and models to OpenMEDLab, which can be accessed via https://github.com/openmedlab.

Yanshuo Wang, Yuan Xu, Xuesong Li et al.

Egocentric assistants often rely on first-person view data to capture user behavior and context for personalized services. Since different users exhibit distinct habits, preferences, and routines, such personalization is essential for truly effective assistance. However, effectively integrating long-term user data for personalization remains a key challenge. To address this, we introduce EgoSelf, a system that includes a graph-based interaction memory constructed from past observations and a dedicated learning task for personalization. The memory captures temporal and semantic relationships among interaction events and entities, from which user-specific profiles are derived. The personalized learning task is formulated as a prediction problem where the model predicts possible future interactions from individual user's historical behavior recorded in the graph. Extensive experiments demonstrate the effectiveness of EgoSelf as a personalized egocentric assistant. Code is available at \href{https://abie-e.github.io/egoself_project/}{https://abie-e.github.io/egoself\_project/}.

Zishi Li, Xiaoxuan Ma, Qiuyan Shang et al.

Temporal repetition counting aims to quantify the repeated action cycles within a video. The majority of existing methods rely on the similarity correlation matrix to characterize the repetitiveness of actions, but their scalability is hindered due to the quadratic computational complexity. In this work, we introduce a novel approach that employs an action query representation to localize repeated action cycles with linear computational complexity. Based on this representation, we further develop two key components to tackle the essential challenges of temporal repetition counting. Firstly, to facilitate open-set action counting, we propose the dynamic update scheme on action queries. Unlike static action queries, this approach dynamically embeds video features into action queries, offering a more flexible and generalizable representation. Secondly, to distinguish between actions of interest and background noise actions, we incorporate inter-query contrastive learning to regularize the video representations corresponding to different action queries. As a result, our method significantly outperforms previous works, particularly in terms of long video sequences, unseen actions, and actions at various speeds. On the challenging RepCountA benchmark, we outperform the state-of-the-art method TransRAC by 26.5% in OBO accuracy, with a 22.7% mean error decrease and 94.1% computational burden reduction. Code is available at https://github.com/lizishi/DeTRC.

Zhenhua Huang, Kunhao Li, Shaojie Wang et al.

Despite the Graph Neural Networks' (GNNs) proficiency in analyzing graph data, achieving high-accuracy and interpretable predictions remains challenging. Existing GNN interpreters typically provide post-hoc explanations disjointed from GNNs' predictions, resulting in misrepresentations. Self-explainable GNNs offer built-in explanations during the training process. However, they cannot exploit the explanatory outcomes to augment prediction performance, and they fail to provide high-quality explanations of node features and require additional processes to generate explainable subgraphs, which is costly. To address the aforementioned limitations, we propose a self-explained and self-supervised graph neural network (SES) to bridge the gap between explainability and prediction. SES comprises two processes: explainable training and enhanced predictive learning. During explainable training, SES employs a global mask generator co-trained with a graph encoder and directly produces crucial structure and feature masks, reducing time consumption and providing node feature and subgraph explanations. In the enhanced predictive learning phase, mask-based positive-negative pairs are constructed utilizing the explanations to compute a triplet loss and enhance the node representations by contrastive learning.

Kai Guan, Rongyuan Wu, Shuai Li et al.

In real-world scenarios, the performance of semantic segmentation often deteriorates when processing low-quality (LQ) images, which may lack clear semantic structures and high-frequency details. Although image restoration techniques offer a promising direction for enhancing degraded visual content, conventional real-world image restoration (Real-IR) models primarily focus on pixel-level fidelity and often fail to recover task-relevant semantic cues, limiting their effectiveness when directly applied to downstream vision tasks. Conversely, existing segmentation models trained on high-quality data lack robustness under real-world degradations. In this paper, we propose Restoration Adaptation for Semantic Segmentation (RASS), which effectively integrates semantic image restoration into the segmentation process, enabling high-quality semantic segmentation on the LQ images directly. Specifically, we first propose a Semantic-Constrained Restoration (SCR) model, which injects segmentation priors into the restoration model by aligning its cross-attention maps with segmentation masks, encouraging semantically faithful image reconstruction. Then, RASS transfers semantic restoration knowledge into segmentation through LoRA-based module merging and task-specific fine-tuning, thereby enhancing the model's robustness to LQ images. To validate the effectiveness of our framework, we construct a real-world LQ image segmentation dataset with high-quality annotations, and conduct extensive experiments on both synthetic and real-world LQ benchmarks. The results show that SCR and RASS significantly outperform state-of-the-art methods in segmentation and restoration tasks. Code, models, and datasets will be available at https://github.com/Ka1Guan/RASS.git.

Mingxuan Liu, Yuhe Ke, Wentao Zhu et al.

The integration of large language models (LLMs) into healthcare holds promise to enhance clinical decision-making, yet their susceptibility to biases remains a critical concern. Gender has long influenced physician behaviors and patient outcomes, raising concerns that LLMs assuming human-like roles, such as clinicians or medical educators, may replicate or amplify gender-related biases. Using case studies from the New England Journal of Medicine Challenge (NEJM), we assigned genders (female, male, or unspecified) to multiple open-source and proprietary LLMs. We evaluated their response consistency across LLM-gender assignments regarding both LLM-based diagnosis and models' judgments on the clinical relevance or necessity of patient gender. In our findings, diagnoses were relatively consistent across LLM genders for most models. However, for patient gender's relevance and necessity in LLM-based diagnosis, all models demonstrated substantial inconsistency across LLM genders, particularly for relevance judgements. Some models even displayed a systematic female-male disparity in their interpretation of patient gender. These findings present an underexplored bias that could undermine the reliability of LLMs in clinical practice, underscoring the need for routine checks of identity-assignment consistency when interacting with LLMs to ensure reliable and equitable AI-supported clinical care.

Shuwei Shao, Zhongcai Pei, Weihai Chen et al.

Recently, self-supervised learning technology has been applied to calculate depth and ego-motion from monocular videos, achieving remarkable performance in autonomous driving scenarios. One widely adopted assumption of depth and ego-motion self-supervised learning is that the image brightness remains constant within nearby frames. Unfortunately, the endoscopic scene does not meet this assumption because there are severe brightness fluctuations induced by illumination variations, non-Lambertian reflections and interreflections during data collection, and these brightness fluctuations inevitably deteriorate the depth and ego-motion estimation accuracy. In this work, we introduce a novel concept referred to as appearance flow to address the brightness inconsistency problem. The appearance flow takes into consideration any variations in the brightness pattern and enables us to develop a generalized dynamic image constraint. Furthermore, we build a unified self-supervised framework to estimate monocular depth and ego-motion simultaneously in endoscopic scenes, which comprises a structure module, a motion module, an appearance module and a correspondence module, to accurately reconstruct the appearance and calibrate the image brightness. Extensive experiments are conducted on the SCARED dataset and EndoSLAM dataset, and the proposed unified framework exceeds other self-supervised approaches by a large margin. To validate our framework's generalization ability on different patients and cameras, we train our model on SCARED but test it on the SERV-CT and Hamlyn datasets without any fine-tuning, and the superior results reveal its strong generalization ability. Code will be available at: \url{https://github.com/ShuweiShao/AF-SfMLearner}.

Tianli Zhao, Xi Sheryl Zhang, Wentao Zhu et al.

For practical deep neural network design on mobile devices, it is essential to consider the constraints incurred by the computational resources and the inference latency in various applications. Among deep network acceleration related approaches, pruning is a widely adopted practice to balance the computational resource consumption and the accuracy, where unimportant connections can be removed either channel-wisely or randomly with a minimal impact on model accuracy. The channel pruning instantly results in a significant latency reduction, while the random weight pruning is more flexible to balance the latency and accuracy. In this paper, we present a unified framework with Joint Channel pruning and Weight pruning (JCW), and achieves a better Pareto-frontier between the latency and accuracy than previous model compression approaches. To fully optimize the trade-off between the latency and accuracy, we develop a tailored multi-objective evolutionary algorithm in the JCW framework, which enables one single search to obtain the optimal candidate architectures for various deployment requirements. Extensive experiments demonstrate that the JCW achieves a better trade-off between the latency and accuracy against various state-of-the-art pruning methods on the ImageNet classification dataset. Our codes are available at https://github.com/jcw-anonymous/JCW.

Wentao Zhu, Tianlong Kong, Shun Lu et al.

Recently, x-vector has been a successful and popular approach for speaker verification, which employs a time delay neural network (TDNN) and statistics pooling to extract speaker characterizing embedding from variable-length utterances. Improvement upon the x-vector has been an active research area, and enormous neural networks have been elaborately designed based on the x-vector, eg, extended TDNN (E-TDNN), factorized TDNN (F-TDNN), and densely connected TDNN (D-TDNN). In this work, we try to identify the optimal architectures from a TDNN based search space employing neural architecture search (NAS), named SpeechNAS. Leveraging the recent advances in the speaker recognition, such as high-order statistics pooling, multi-branch mechanism, D-TDNN and angular additive margin softmax (AAM) loss with a minimum hyper-spherical energy (MHE), SpeechNAS automatically discovers five network architectures, from SpeechNAS-1 to SpeechNAS-5, of various numbers of parameters and GFLOPs on the large-scale text-independent speaker recognition dataset VoxCeleb1. Our derived best neural network achieves an equal error rate (EER) of 1.02% on the standard test set of VoxCeleb1, which surpasses previous TDNN based state-of-the-art approaches by a large margin. Code and trained weights are in https://github.com/wentaozhu/speechnas.git

Wentao Zhu, Can Zhao, Wenqi Li et al.

Deep Learning (DL) models are becoming larger, because the increase in model size might offer significant accuracy gain. To enable the training of large deep networks, data parallelism and model parallelism are two well-known approaches for parallel training. However, data parallelism does not help reduce memory footprint per device. In this work, we introduce Large deep 3D ConvNets with Automated Model Parallelism (LAMP) and investigate the impact of both input's and deep 3D ConvNets' size on segmentation accuracy. Through automated model parallelism, it is feasible to train large deep 3D ConvNets with a large input patch, even the whole image. Extensive experiments demonstrate that, facilitated by the automated model parallelism, the segmentation accuracy can be improved through increasing model size and input context size, and large input yields significant inference speedup compared with sliding window of small patches in the inference. Code is available\footnote{https://monai.io/research/lamp-automated-model-parallelism}.

Wentao Zhu, Yufang Huang, Liang Zeng et al.

Methods: Our deep learning model, called AnatomyNet, segments OARs from head and neck CT images in an end-to-end fashion, receiving whole-volume HaN CT images as input and generating masks of all OARs of interest in one shot. AnatomyNet is built upon the popular 3D U-net architecture, but extends it in three important ways: 1) a new encoding scheme to allow auto-segmentation on whole-volume CT images instead of local patches or subsets of slices, 2) incorporating 3D squeeze-and-excitation residual blocks in encoding layers for better feature representation, and 3) a new loss function combining Dice scores and focal loss to facilitate the training of the neural model. These features are designed to address two main challenges in deep-learning-based HaN segmentation: a) segmenting small anatomies (i.e., optic chiasm and optic nerves) occupying only a few slices, and b) training with inconsistent data annotations with missing ground truth for some anatomical structures. Results: We collected 261 HaN CT images to train AnatomyNet, and used MICCAI Head and Neck Auto Segmentation Challenge 2015 as a benchmark dataset to evaluate the performance of AnatomyNet. The objective is to segment nine anatomies: brain stem, chiasm, mandible, optic nerve left, optic nerve right, parotid gland left, parotid gland right, submandibular gland left, and submandibular gland right. Compared to previous state-of-the-art results from the MICCAI 2015 competition, AnatomyNet increases Dice similarity coefficient by 3.3% on average. AnatomyNet takes about 0.12 seconds to fully segment a head and neck CT image of dimension 178 x 302 x 225, significantly faster than previous methods. In addition, the model is able to process whole-volume CT images and delineate all OARs in one pass, requiring little pre- or post-processing. https://github.com/wentaozhu/AnatomyNet-for-anatomical-segmentation.git.

Wentao Zhu, Yeeleng S. Vang, Yufang Huang et al.

Recently deep learning has been witnessing widespread adoption in various medical image applications. However, training complex deep neural nets requires large-scale datasets labeled with ground truth, which are often unavailable in many medical image domains. For instance, to train a deep neural net to detect pulmonary nodules in lung computed tomography (CT) images, current practice is to manually label nodule locations and sizes in many CT images to construct a sufficiently large training dataset, which is costly and difficult to scale. On the other hand, electronic medical records (EMR) contain plenty of partial information on the content of each medical image. In this work, we explore how to tap this vast, but currently unexplored data source to improve pulmonary nodule detection. We propose DeepEM, a novel deep 3D ConvNet framework augmented with expectation-maximization (EM), to mine weakly supervised labels in EMRs for pulmonary nodule detection. Experimental results show that DeepEM can lead to 1.5\% and 3.9\% average improvement in free-response receiver operating characteristic (FROC) scores on LUNA16 and Tianchi datasets, respectively, demonstrating the utility of incomplete information in EMRs for improving deep learning algorithms.\footnote{https://github.com/uci-cbcl/DeepEM-for-Weakly-Supervised-Detection.git}

Wentao Zhu, Chaochun Liu, Wei Fan et al.

In this work, we present a fully automated lung computed tomography (CT) cancer diagnosis system, DeepLung. DeepLung consists of two components, nodule detection (identifying the locations of candidate nodules) and classification (classifying candidate nodules into benign or malignant). Considering the 3D nature of lung CT data and the compactness of dual path networks (DPN), two deep 3D DPN are designed for nodule detection and classification respectively. Specifically, a 3D Faster Regions with Convolutional Neural Net (R-CNN) is designed for nodule detection with 3D dual path blocks and a U-net-like encoder-decoder structure to effectively learn nodule features. For nodule classification, gradient boosting machine (GBM) with 3D dual path network features is proposed. The nodule classification subnetwork was validated on a public dataset from LIDC-IDRI, on which it achieved better performance than state-of-the-art approaches and surpassed the performance of experienced doctors based on image modality. Within the DeepLung system, candidate nodules are detected first by the nodule detection subnetwork, and nodule diagnosis is conducted by the classification subnetwork. Extensive experimental results demonstrate that DeepLung has performance comparable to experienced doctors both for the nodule-level and patient-level diagnosis on the LIDC-IDRI dataset.\footnote{https://github.com/uci-cbcl/DeepLung.git}

Wentao Zhu, Xiang Xiang, Trac D. Tran et al.

Mass segmentation provides effective morphological features which are important for mass diagnosis. In this work, we propose a novel end-to-end network for mammographic mass segmentation which employs a fully convolutional network (FCN) to model a potential function, followed by a CRF to perform structured learning. Because the mass distribution varies greatly with pixel position, the FCN is combined with a position priori. Further, we employ adversarial training to eliminate over-fitting due to the small sizes of mammogram datasets. Multi-scale FCN is employed to improve the segmentation performance. Experimental results on two public datasets, INbreast and DDSM-BCRP, demonstrate that our end-to-end network achieves better performance than state-of-the-art approaches. \footnote{https://github.com/wentaozhu/adversarial-deep-structural-networks.git}

Zhenhua Huang, Kunhao Li, Shaojie Wang et al.

Graph neural networks (GNNs) are widely applied in graph data modeling. However, existing GNNs are often trained in a task-driven manner that fails to fully capture the intrinsic nature of the graph structure, resulting in sub-optimal node and graph representations. To address this limitation, we propose a novel Graph structure Prompt Learning method (GPL) to enhance the training of GNNs, which is inspired by prompt mechanisms in natural language processing. GPL employs task-independent graph structure losses to encourage GNNs to learn intrinsic graph characteristics while simultaneously solving downstream tasks, producing higher-quality node and graph representations. In extensive experiments on eleven real-world datasets, after being trained by GPL, GNNs significantly outperform their original performance on node classification, graph classification, and edge prediction tasks (up to 10.28%, 16.5%, and 24.15%, respectively). By allowing GNNs to capture the inherent structural prompts of graphs in GPL, they can alleviate the issue of over-smooth and achieve new state-of-the-art performances, which introduces a novel and effective direction for GNN research with potential applications in various domains.

Wentao Zhu, Zhining Zhang, Yizhou Wang

Understanding and attributing mental states, known as Theory of Mind (ToM), emerges as a fundamental capability for human social reasoning. While Large Language Models (LLMs) appear to possess certain ToM abilities, the mechanisms underlying these capabilities remain elusive. In this study, we discover that it is possible to linearly decode the belief status from the perspectives of various agents through neural activations of language models, indicating the existence of internal representations of self and others' beliefs. By manipulating these representations, we observe dramatic changes in the models' ToM performance, underscoring their pivotal role in the social reasoning process. Additionally, our findings extend to diverse social reasoning tasks that involve different causal inference patterns, suggesting the potential generalizability of these representations.

Chengyang Li, Kaiyi Xiong, Yuan Xu et al.

Embodied foundation models have achieved significant breakthroughs in robotic manipulation, yet they still depend heavily on large-scale robot demonstrations. Although recent works have explored leveraging human data to alleviate this dependency, effectively extracting transferable knowledge remains a significant challenge due to the inherent embodiment gap between human and robot. We argue that the intention underlying human actions can serve as a powerful intermediate representation for bridging this gap. In this paper, we introduce a novel framework that explicitly learns and transfers human intention to facilitate robotic manipulation. Specifically, we model intention through gaze, as it naturally precedes physical actions and serves as an observable proxy for human intent. Our model is first pretrained on a large-scale egocentric human dataset to capture human intention and its synergy with action, followed by finetuning on a small set of robot and human data. During inference, the model adopts a Chain-of-Thought reasoning paradigm, sequentially predicting intention before executing the action. Extensive evaluations in simulation and real-world settings, across long-horizon and fine-grained tasks, and under few-shot and robustness benchmarks, show that our method consistently outperforms strong baselines, generalizes better, and achieves state-of-the-art performance.

Wentao Zhu

In recent years, researchers combine both audio and video signals to deal with challenges where actions are not well represented or captured by visual cues. However, how to effectively leverage the two modalities is still under development. In this work, we develop a multiscale multimodal Transformer (MMT) that leverages hierarchical representation learning. Particularly, MMT is composed of a novel multiscale audio Transformer (MAT) and a multiscale video Transformer [43]. To learn a discriminative cross-modality fusion, we further design multimodal supervised contrastive objectives called audio-video contrastive loss (AVC) and intra-modal contrastive loss (IMC) that robustly align the two modalities. MMT surpasses previous state-of-the-art approaches by 7.3% and 2.1% on Kinetics-Sounds and VGGSound in terms of the top-1 accuracy without external training data. Moreover, the proposed MAT significantly outperforms AST [28] by 22.2%, 4.4% and 4.7% on three public benchmark datasets, and is about 3% more efficient based on the number of FLOPs and 9.8% more efficient based on GPU memory usage.

Wentao Zhu

Audio and video are two most common modalities in the mainstream media platforms, e.g., YouTube. To learn from multimodal videos effectively, in this work, we propose a novel audio-video recognition approach termed audio video Transformer, AVT, leveraging the effective spatio-temporal representation by the video Transformer to improve action recognition accuracy. For multimodal fusion, simply concatenating multimodal tokens in a cross-modal Transformer requires large computational and memory resources, instead we reduce the cross-modality complexity through an audio-video bottleneck Transformer. To improve the learning efficiency of multimodal Transformer, we integrate self-supervised objectives, i.e., audio-video contrastive learning, audio-video matching, and masked audio and video learning, into AVT training, which maps diverse audio and video representations into a common multimodal representation space. We further propose a masked audio segment loss to learn semantic audio activities in AVT. Extensive experiments and ablation studies on three public datasets and two in-house datasets consistently demonstrate the effectiveness of the proposed AVT. Specifically, AVT outperforms its previous state-of-the-art counterparts on Kinetics-Sounds by 8%. AVT also surpasses one of the previous state-of-the-art video Transformers [25] by 10% on VGGSound by leveraging the audio signal. Compared to one of the previous state-of-the-art multimodal methods, MBT [32], AVT is 1.3% more efficient in terms of FLOPs and improves the accuracy by 3.8% on Epic-Kitchens-100.

Wentao Zhu

Vision transformers (ViTs) have achieved promising results on a variety of Computer Vision tasks, however their quadratic complexity in the number of input tokens has limited their application specially in resource-constrained settings. Previous approaches that employ gradual token reduction to address this challenge assume that token redundancy in one layer implies redundancy in all the following layers. We empirically demonstrate that this assumption is often not correct, i.e., tokens that are redundant in one layer can be useful in later layers. We employ this key insight to propose a novel token propagation controller (TPC) that incorporates two different token-distributions, i.e., pause probability and restart probability to control the reduction and reuse of tokens respectively, which results in more efficient token utilization. To improve the estimates of token distributions, we propose a smoothing mechanism that acts as a regularizer and helps remove noisy outliers. Furthermore, to improve the training-stability of our proposed TPC, we introduce a model stabilizer that is able to implicitly encode local image structures and minimize accuracy fluctuations during model training. We present extensive experimental results on the ImageNet-1K dataset using DeiT, LV-ViT and Swin models to demonstrate the effectiveness of our proposed method. For example, compared to baseline models, our proposed method improves the inference speed of the DeiT-S by 250% while increasing the classification accuracy by 1.0%.

Yuan Xu, Zimu Zhang, Xiaoxuan Ma et al.

Virtual and augmented reality systems increasingly demand intelligent adaptation to user behaviors for enhanced interaction experiences. Achieving this requires accurately understanding human intentions and predicting future situated behaviors - such as gaze direction and object interactions - which is vital for creating responsive VR/AR environments and applications like personalized assistants. However, accurate behavioral prediction demands modeling the underlying cognitive processes that drive human-environment interactions. In this work, we introduce a hierarchical, intention-aware framework that models human intentions and predicts detailed situated behaviors by leveraging cognitive mechanisms. Given historical human dynamics and the observation of scene contexts, our framework first identifies potential interaction targets and forecasts fine-grained future behaviors. We propose a dynamic Graph Convolutional Network (GCN) to effectively capture human-environment relationships. Extensive experiments on challenging real-world benchmarks and live VR environment demonstrate the effectiveness of our approach, achieving superior performance across all metrics and enabling practical applications for proactive VR systems that anticipate user behaviors and adapt virtual environments accordingly.

Wentao Zhu, Zhining Zhang, Yuwei Ren et al.

Mirror neurons are a class of neurons that activate both when an individual observes an action and when they perform the same action. This mechanism reveals a fundamental interplay between action understanding and embodied execution, suggesting that these two abilities are inherently connected. Nonetheless, existing machine learning methods largely overlook this interplay, treating these abilities as separate tasks. In this study, we provide a unified perspective in modeling them through the lens of representation learning. We first observe that their intermediate representations spontaneously align. Inspired by mirror neurons, we further introduce an approach that explicitly aligns the representations of observed and executed actions. Specifically, we employ two linear layers to map the representations to a shared latent space, where contrastive learning enforces the alignment of corresponding representations, effectively maximizing their mutual information. Experiments demonstrate that this simple approach fosters mutual synergy between the two tasks, effectively improving representation quality and generalization.

Sihan Zhao, Zixuan Wang, Tianyu Luan et al.

Human motion generation has found widespread applications in AR/VR, film, sports, and medical rehabilitation, offering a cost-effective alternative to traditional motion capture systems. However, evaluating the fidelity of such generated motions is a crucial, multifaceted task. Although previous approaches have attempted at motion fidelity evaluation using human perception or physical constraints, there remains an inherent gap between human-perceived fidelity and physical feasibility. Moreover, the subjective and coarse binary labeling of human perception further undermines the development of a robust data-driven metric. We address these issues by introducing a physical labeling method. This method evaluates motion fidelity by calculating the minimum modifications needed for a motion to align with physical laws. With this approach, we are able to produce fine-grained, continuous physical alignment annotations that serve as objective ground truth. With these annotations, we propose PP-Motion, a novel data-driven metric to evaluate both physical and perceptual fidelity of human motion. To effectively capture underlying physical priors, we employ Pearson's correlation loss for the training of our metric. Additionally, by incorporating a human-based perceptual fidelity loss, our metric can capture fidelity that simultaneously considers both human perception and physical alignment. Experimental results demonstrate that our metric, PP-Motion, not only aligns with physical laws but also aligns better with human perception of motion fidelity than previous work.

Yizhe Cheng, Chunxun Tian, Haoru Wang et al.

Solving Electromagnetic Inverse Scattering Problems (EISP) is fundamental in applications such as medical imaging, where the goal is to reconstruct the relative permittivity from scattered electromagnetic field. This inverse process is inherently ill-posed and highly nonlinear, making it particularly challenging, especially under sparse transmitter setups, e.g., with only one transmitter. A recent machine learning-based approach, Img-Interiors, shows promising results by leveraging continuous implicit functions. However, it requires time-consuming case-specific optimization and fails under sparse transmitter setups. To address these limitations, we revisit EISP from a data-driven perspective. The scarcity of transmitters leads to an insufficient amount of measured data, which fails to capture adequate physical information for stable inversion. Built on this insight, we propose a fully end-to-end and data-driven framework that predicts the relative permittivity of scatterers from measured fields, leveraging data distribution priors to compensate for the lack of physical information. This design enables data-driven training and feed-forward prediction of relative permittivity while maintaining strong robustness to transmitter sparsity. Extensive experiments show that our method outperforms state-of-the-art approaches in reconstruction accuracy and robustness. Notably, it achieves high-quality results even with a single transmitter, a setting where previous methods consistently fail. This work offers a fundamentally new perspective on electromagnetic inverse scattering and represents a major step toward cost-effective practical solutions for electromagnetic imaging.