Xinyu Zhang

Xinyu Zhang, Dongdong Li, Zhigang Wang et al.

Most existing unsupervised person re-identification (Re-ID) methods use clustering to generate pseudo labels for model training. Unfortunately, clustering sometimes mixes different true identities together or splits the same identity into two or more sub clusters. Training on these noisy clusters substantially hampers the Re-ID accuracy. Due to the limited samples in each identity, we suppose there may lack some underlying information to well reveal the accurate clusters. To discover these information, we propose an Implicit Sample Extension (\OurWholeMethod) method to generate what we call support samples around the cluster boundaries. Specifically, we generate support samples from actual samples and their neighbouring clusters in the embedding space through a progressive linear interpolation (PLI) strategy. PLI controls the generation with two critical factors, i.e., 1) the direction from the actual sample towards its K-nearest clusters and 2) the degree for mixing up the context information from the K-nearest clusters. Meanwhile, given the support samples, ISE further uses a label-preserving loss to pull them towards their corresponding actual samples, so as to compact each cluster. Consequently, ISE reduces the "sub and mixed" clustering errors, thus improving the Re-ID performance. Extensive experiments demonstrate that the proposed method is effective and achieves state-of-the-art performance for unsupervised person Re-ID. Code is available at: \url{https://github.com/PaddlePaddle/PaddleClas}.

Junkun Yuan, Xinyu Zhang, Hao Zhou et al. · tencent-ai

Model pre-training is essential in human-centric perception. In this paper, we first introduce masked image modeling (MIM) as a pre-training approach for this task. Upon revisiting the MIM training strategy, we reveal that human structure priors offer significant potential. Motivated by this insight, we further incorporate an intuitive human structure prior - human parts - into pre-training. Specifically, we employ this prior to guide the mask sampling process. Image patches, corresponding to human part regions, have high priority to be masked out. This encourages the model to concentrate more on body structure information during pre-training, yielding substantial benefits across a range of human-centric perception tasks. To further capture human characteristics, we propose a structure-invariant alignment loss that enforces different masked views, guided by the human part prior, to be closely aligned for the same image. We term the entire method as HAP. HAP simply uses a plain ViT as the encoder yet establishes new state-of-the-art performance on 11 human-centric benchmarks, and on-par result on one dataset. For example, HAP achieves 78.1% mAP on MSMT17 for person re-identification, 86.54% mA on PA-100K for pedestrian attribute recognition, 78.2% AP on MS COCO for 2D pose estimation, and 56.0 PA-MPJPE on 3DPW for 3D pose and shape estimation.

Bo Yang, Xinyu Zhang, Jian Zhang et al.

Single-frame infrared small target detection is considered to be a challenging task, due to the extreme imbalance between target and background, bounding box regression is extremely sensitive to infrared small target, and target information is easy to lose in the high-level semantic layer. In this article, we propose an enhancing feature learning network (EFLNet) to address these problems. First, we notice that there is an extremely imbalance between the target and the background in the infrared image, which makes the model pay more attention to the background features rather than target features. To address this problem, we propose a new adaptive threshold focal loss (ATFL) function that decouples the target and the background, and utilizes the adaptive mechanism to adjust the loss weight to force the model to allocate more attention to target features. Second, we introduce the normalized Gaussian Wasserstein distance (NWD) to alleviate the difficulty of convergence caused by the extreme sensitivity of the bounding box regression to infrared small target. Finally, we incorporate a dynamic head mechanism into the network to enable adaptive learning of the relative importance of each semantic layer. Experimental results demonstrate our method can achieve better performance in the detection performance of infrared small target compared to the state-of-the-art (SOTA) deep-learning-based methods. The source codes and bounding box annotated datasets are available at https://github.com/YangBo0411/infrared-small-target.

Xinyu Zhang, Kelechi Ogueji, Xueguang Ma et al.

Dense retrieval models using a transformer-based bi-encoder design have emerged as an active area of research. In this work, we focus on the task of monolingual retrieval in a variety of typologically diverse languages using one such design. Although recent work with multilingual transformers demonstrates that they exhibit strong cross-lingual generalization capabilities, there remain many open research questions, which we tackle here. Our study is organized as a "best practices" guide for training multilingual dense retrieval models, broken down into three main scenarios: where a multilingual transformer is available, but relevance judgments are not available in the language of interest; where both models and training data are available; and, where training data are available not but models. In considering these scenarios, we gain a better understanding of the role of multi-stage fine-tuning, the strength of cross-lingual transfer under various conditions, the usefulness of out-of-language data, and the advantages of multilingual vs. monolingual transformers. Our recommendations offer a guide for practitioners building search applications, particularly for low-resource languages, and while our work leaves open a number of research questions, we provide a solid foundation for future work.

Mingxiang Liao, Hannan Lu, Xinyu Zhang et al.

Comprehensive and constructive evaluation protocols play an important role in the development of sophisticated text-to-video (T2V) generation models. Existing evaluation protocols primarily focus on temporal consistency and content continuity, yet largely ignore the dynamics of video content. Dynamics are an essential dimension for measuring the visual vividness and the honesty of video content to text prompts. In this study, we propose an effective evaluation protocol, termed DEVIL, which centers on the dynamics dimension to evaluate T2V models. For this purpose, we establish a new benchmark comprising text prompts that fully reflect multiple dynamics grades, and define a set of dynamics scores corresponding to various temporal granularities to comprehensively evaluate the dynamics of each generated video. Based on the new benchmark and the dynamics scores, we assess T2V models with the design of three metrics: dynamics range, dynamics controllability, and dynamics-based quality. Experiments show that DEVIL achieves a Pearson correlation exceeding 90% with human ratings, demonstrating its potential to advance T2V generation models. Code is available at https://github.com/MingXiangL/DEVIL.

Li Wang, Xinyu Zhang, Wenyuan Qin et al.

3D Multi-object tracking (MOT) ensures consistency during continuous dynamic detection, conducive to subsequent motion planning and navigation tasks in autonomous driving. However, camera-based methods suffer in the case of occlusions and it can be challenging to accurately track the irregular motion of objects for LiDAR-based methods. Some fusion methods work well but do not consider the untrustworthy issue of appearance features under occlusion. At the same time, the false detection problem also significantly affects tracking. As such, we propose a novel camera-LiDAR fusion 3D MOT framework based on the Combined Appearance-Motion Optimization (CAMO-MOT), which uses both camera and LiDAR data and significantly reduces tracking failures caused by occlusion and false detection. For occlusion problems, we are the first to propose an occlusion head to select the best object appearance features multiple times effectively, reducing the influence of occlusions. To decrease the impact of false detection in tracking, we design a motion cost matrix based on confidence scores which improve the positioning and object prediction accuracy in 3D space. As existing multi-object tracking methods only consider a single category, we also propose to build a multi-category loss to implement multi-object tracking in multi-category scenes. A series of validation experiments are conducted on the KITTI and nuScenes tracking benchmarks. Our proposed method achieves state-of-the-art performance and the lowest identity switches (IDS) value (23 for Car and 137 for Pedestrian) among all multi-modal MOT methods on the KITTI test dataset. And our proposed method achieves state-of-the-art performance among all algorithms on the nuScenes test dataset with 75.3% AMOTA.

Xinyu Zhang, Vincent CS. Lee, Jia Rong et al.

Thyroid disease instances have been continuously increasing since the 1990s, and thyroid cancer has become the most rapidly rising disease among all the malignancies in recent years. Most existing studies focused on applying deep convolutional neural networks for detecting thyroid cancer. Despite their satisfactory performance on binary classification tasks, limited studies have explored multi-class classification of thyroid disease types; much less is known of the diagnosis of co-existence situation for different types of thyroid diseases. Therefore, this study proposed a novel multi-channel convolutional neural network (CNN) architecture to address the multi-class classification task of thyroid disease. The multi-channel CNN merits from computed tomography to drive a comprehensive diagnostic decision for the overall thyroid gland, emphasizing the disease co-existence circumstance. Moreover, this study also examined alternative strategies to enhance the diagnostic accuracy of CNN models through concatenation of different scales of feature maps. Benchmarking experiments demonstrate the improved performance of the proposed multi-channel CNN architecture compared with the standard single-channel CNN architecture. More specifically, the multi-channel CNN achieved an accuracy of 0.909, precision of 0.944, recall of 0.896, specificity of 0.994, and F1 of 0.917, in contrast to the single-channel CNN, which obtained 0.902, 0.892, 0.909, 0.993, 0.898, respectively. In addition, the proposed model was evaluated in different gender groups; it reached a diagnostic accuracy of 0.908 for the female group and 0.901 for the male group. Collectively, the results highlight that the proposed multi-channel CNN has excellent generalization and has the potential to be deployed to provide computational decision support in clinical settings.

Haobo Yang, Shiyan Zhang, Zhuoyi Yang et al.

With the increasing complexity of the traffic environment, the importance of safety perception in intelligent driving is growing. Conventional methods in the robust perception of intelligent driving focus on training models with anomalous data, letting the deep neural network decide how to tackle anomalies. However, these models cannot adapt smoothly to the diverse and complex real-world environment. This paper proposes a new type of metric known as Eloss and offers a novel training strategy to empower perception models from the aspect of anomaly detection. Eloss is designed based on an explanation of the perception model's information compression layers. Specifically, taking inspiration from the design of a communication system, the information transmission process of an information compression network has two expectations: the amount of information changes steadily, and the information entropy continues to decrease. Then Eloss can be obtained according to the above expectations, guiding the update of related network parameters and producing a sensitive metric to identify anomalies while maintaining the model performance. Our experiments demonstrate that Eloss can deviate from the standard value by a factor over 100 with anomalous data and produce distinctive values for similar but different types of anomalies, showing the effectiveness of the proposed method. Our code is available at: (code available after paper accepted).

Jiagang Liu, Yun Mi, Xinyu Zhang et al.

Various mobile applications that comprise dependent tasks are gaining widespread popularity and are increasingly complex. These applications often have low-latency requirements, resulting in a significant surge in demand for computing resources. With the emergence of mobile edge computing (MEC), it becomes the most significant issue to offload the application tasks onto small-scale devices deployed at the edge of the mobile network for obtaining a high-quality user experience. However, since the environment of MEC is dynamic, most existing works focusing on task graph offloading, which rely heavily on expert knowledge or accurate analytical models, fail to fully adapt to such environmental changes, resulting in the reduction of user experience. This paper investigates the task graph offloading in MEC, considering the time-varying computation capabilities of edge computing devices. To adapt to environmental changes, we model the task graph scheduling for computation offloading as a Markov Decision Process (MDP). Then, we design a deep reinforcement learning algorithm (SATA-DRL) to learn the task scheduling strategy from the interaction with the environment, to improve user experience. Extensive simulations validate that SATA-DRL is superior to existing strategies in terms of reducing average makespan and deadline violation.

Lingyan Xue, Xinyu Zhang, Weidong Jiang et al.

Machine learning and deep learning classification models are data-driven, and the model and the data jointly determine their classification performance. It is biased to evaluate the model's performance only based on the classifier accuracy while ignoring the data separability. Sometimes, the model exhibits excellent accuracy, which might be attributed to its testing on highly separable data. Most of the current studies on data separability measures are defined based on the distance between sample points, but this has been demonstrated to fail in several circumstances. In this paper, we propose a new separability measure--the rate of separability (RS), which is based on the data coding rate. We validate its effectiveness as a supplement to the separability measure by comparing it to four other distance-based measures on synthetic datasets. Then, we demonstrate the positive correlation between the proposed measure and recognition accuracy in a multi-task scenario constructed from a real dataset. Finally, we discuss the methods for evaluating the classification performance of machine learning and deep learning models considering data separability.

Sifan Zhou, Liang Li, Xinyu Zhang et al.

Due to highly constrained computing power and memory, deploying 3D lidar-based detectors on edge devices equipped in autonomous vehicles and robots poses a crucial challenge. Being a convenient and straightforward model compression approach, Post-Training Quantization (PTQ) has been widely adopted in 2D vision tasks. However, applying it directly to 3D lidar-based tasks inevitably leads to performance degradation. As a remedy, we propose an effective PTQ method called LiDAR-PTQ, which is particularly curated for 3D lidar detection (both SPConv-based and SPConv-free). Our LiDAR-PTQ features three main components, \textbf{(1)} a sparsity-based calibration method to determine the initialization of quantization parameters, \textbf{(2)} a Task-guided Global Positive Loss (TGPL) to reduce the disparity between the final predictions before and after quantization, \textbf{(3)} an adaptive rounding-to-nearest operation to minimize the layerwise reconstruction error. Extensive experiments demonstrate that our LiDAR-PTQ can achieve state-of-the-art quantization performance when applied to CenterPoint (both Pillar-based and Voxel-based). To our knowledge, for the very first time in lidar-based 3D detection tasks, the PTQ INT8 model's accuracy is almost the same as the FP32 model while enjoying $3\times$ inference speedup. Moreover, our LiDAR-PTQ is cost-effective being $30\times$ faster than the quantization-aware training method. Code will be released at \url{https://github.com/StiphyJay/LiDAR-PTQ}.

Lei Yang, Xinyu Zhang, Jun Li et al.

Modern autonomous vehicle perception systems often struggle with occlusions and limited perception range. Previous studies have demonstrated the effectiveness of cooperative perception in extending the perception range and overcoming occlusions, thereby enhancing the safety of autonomous driving. In recent years, a series of cooperative perception datasets have emerged; however, these datasets primarily focus on cameras and LiDAR, neglecting 4D Radar, a sensor used in single-vehicle autonomous driving to provide robust perception in adverse weather conditions. In this paper, to bridge the gap created by the absence of 4D Radar datasets in cooperative perception, we present V2X-Radar, the first large-scale, real-world multi-modal dataset featuring 4D Radar. V2X-Radar dataset is collected using a connected vehicle platform and an intelligent roadside unit equipped with 4D Radar, LiDAR, and multi-view cameras. The collected data encompasses sunny and rainy weather conditions, spanning daytime, dusk, and nighttime, as well as various typical challenging scenarios. The dataset consists of 20K LiDAR frames, 40K camera images, and 20K 4D Radar data, including 350K annotated boxes across five categories. To support various research domains, we have established V2X-Radar-C for cooperative perception, V2X-Radar-I for roadside perception, and V2X-Radar-V for single-vehicle perception. Furthermore, we provide comprehensive benchmarks across these three sub-datasets. We will release all datasets and benchmark codebase at https://huggingface.co/datasets/yanglei18/V2X-Radar and https://github.com/yanglei18/V2X-Radar.

Yurong Zhang, Honghao Chen, Xinyu Zhang et al.

Parameter-efficient transfer learning (PETL) is a promising task, aiming to adapt the large-scale pre-trained model to downstream tasks with a relatively modest cost. However, current PETL methods struggle in compressing computational complexity and bear a heavy inference burden due to the complete forward process. This paper presents an efficient visual recognition paradigm, called Dynamic Adapter (Dyn-Adapter), that boosts PETL efficiency by subtly disentangling features in multiple levels. Our approach is simple: first, we devise a dynamic architecture with balanced early heads for multi-level feature extraction, along with adaptive training strategy. Second, we introduce a bidirectional sparsity strategy driven by the pursuit of powerful generalization ability. These qualities enable us to fine-tune efficiently and effectively: we reduce FLOPs during inference by 50%, while maintaining or even yielding higher recognition accuracy. Extensive experiments on diverse datasets and pretrained backbones demonstrate the potential of Dyn-Adapter serving as a general efficiency booster for PETL in vision recognition tasks.

Xinyu Zhang, Changzhi Zhou, Linmei Hu et al.

Existing code large language models (LLMs) often rely on large-scale instruction data distilled from proprietary LLMs for fine-tuning, which typically incurs high costs. In this paper, we explore the potential of small-scale open-source LLMs (e.g., 7B) as synthesizers for high-quality code instruction data construction. We first observe that the data synthesis capability of small-scale LLMs can be enhanced by training on a few superior data synthesis samples from proprietary LLMs. Building on this, we propose a novel iterative self-distillation approach to bootstrap small-scale LLMs, transforming them into powerful synthesizers that reduce reliance on proprietary LLMs and minimize costs. Concretely, in each iteration, to obtain diverse and high-quality self-distilled data, we design multi-checkpoint sampling and multi-aspect scoring strategies for initial data selection. Furthermore, to identify the most influential samples, we introduce a gradient-based influence estimation method for final data filtering. Based on the code instruction datasets from the small-scale synthesizers, we develop SCoder, a family of code generation models fine-tuned from DeepSeek-Coder. SCoder models achieve state-of-the-art code generation capabilities, demonstrating the effectiveness of our method.

Bin Yan, Xinyu Zhang, Dong Wang et al.

Visual object tracking aims to precisely estimate the bounding box for the given target, which is a challenging problem due to factors such as deformation and occlusion. Many recent trackers adopt the multiple-stage tracking strategy to improve the quality of bounding box estimation. These methods first coarsely locate the target and then refine the initial prediction in the following stages. However, existing approaches still suffer from limited precision, and the coupling of different stages severely restricts the method's transferability. This work proposes a novel, flexible, and accurate refinement module called Alpha-Refine (AR), which can significantly improve the base trackers' box estimation quality. By exploring a series of design options, we conclude that the key to successful refinement is extracting and maintaining detailed spatial information as much as possible. Following this principle, Alpha-Refine adopts a pixel-wise correlation, a corner prediction head, and an auxiliary mask head as the core components. Comprehensive experiments on TrackingNet, LaSOT, GOT-10K, and VOT2020 benchmarks with multiple base trackers show that our approach significantly improves the base trackers' performance with little extra latency. The proposed Alpha-Refine method leads to a series of strengthened trackers, among which the ARSiamRPN (AR strengthened SiamRPNpp) and the ARDiMP50 (ARstrengthened DiMP50) achieve good efficiency-precision trade-off, while the ARDiMPsuper (AR strengthened DiMP-super) achieves very competitive performance at a real-time speed. Code and pretrained models are available at https://github.com/MasterBin-IIAU/AlphaRefine.

Xin Gao, Tianheng Qiu, Xinyu Zhang et al.

Coarse-to-fine schemes are widely used in traditional single-image motion deblur; however, in the context of deep learning, existing multi-scale algorithms not only require the use of complex modules for feature fusion of low-scale RGB images and deep semantics, but also manually generate low-resolution pairs of images that do not have sufficient confidence. In this work, we propose a multi-scale network based on single-input and multiple-outputs(SIMO) for motion deblurring. This simplifies the complexity of algorithms based on a coarse-to-fine scheme. To alleviate restoration defects impacting detail information brought about by using a multi-scale architecture, we combine the characteristics of real-world blurring trajectories with a learnable wavelet transform module to focus on the directional continuity and frequency features of the step-by-step transitions between blurred images to sharp images. In conclusion, we propose a multi-scale network with a learnable discrete wavelet transform (MLWNet), which exhibits state-of-the-art performance on multiple real-world deblurred datasets, in terms of both subjective and objective quality as well as computational efficiency.

Xinyu Zhang, Abdeslam Boularias

Learning a single universal policy that can perform a diverse set of manipulation tasks is a promising new direction in robotics. However, existing techniques are limited to learning policies that can only perform tasks that are encountered during training, and require a large number of demonstrations to learn new tasks. Humans, on the other hand, often can learn a new task from a single unannotated demonstration. In this work, we propose the Invariance-Matching One-shot Policy Learning (IMOP) algorithm. In contrast to the standard practice of learning the end-effector's pose directly, IMOP first learns invariant regions of the state space for a given task, and then computes the end-effector's pose through matching the invariant regions between demonstrations and test scenes. Trained on the 18 RLBench tasks, IMOP achieves a success rate that outperforms the state-of-the-art consistently, by 4.5% on average over the 18 tasks. More importantly, IMOP can learn a novel task from a single unannotated demonstration, and without any fine-tuning, and achieves an average success rate improvement of $11.5\%$ over the state-of-the-art on 22 novel tasks selected across nine categories. IMOP can also generalize to new shapes and learn to manipulate objects that are different from those in the demonstration. Further, IMOP can perform one-shot sim-to-real transfer using a single real-robot demonstration.

Lanqing Li, Hai Zhang, Xinyu Zhang et al.

As a marriage between offline RL and meta-RL, the advent of offline meta-reinforcement learning (OMRL) has shown great promise in enabling RL agents to multi-task and quickly adapt while acquiring knowledge safely. Among which, context-based OMRL (COMRL) as a popular paradigm, aims to learn a universal policy conditioned on effective task representations. In this work, by examining several key milestones in the field of COMRL, we propose to integrate these seemingly independent methodologies into a unified framework. Most importantly, we show that the pre-existing COMRL algorithms are essentially optimizing the same mutual information objective between the task variable $M$ and its latent representation $Z$ by implementing various approximate bounds. Such theoretical insight offers ample design freedom for novel algorithms. As demonstrations, we propose a supervised and a self-supervised implementation of $I(Z; M)$, and empirically show that the corresponding optimization algorithms exhibit remarkable generalization across a broad spectrum of RL benchmarks, context shift scenarios, data qualities and deep learning architectures. This work lays the information theoretic foundation for COMRL methods, leading to a better understanding of task representation learning in the context of reinforcement learning. Given its generality, we envision our framework as a promising offline pre-training paradigm of foundation models for decision making.

Changzhi Zhou, Xinyu Zhang, Dandan Song et al.

Code generation has attracted increasing attention with the rise of Large Language Models (LLMs). Many studies have developed powerful code LLMs by synthesizing code-related instruction data and applying supervised fine-tuning. However, these methods are limited by teacher model distillation and ignore the potential of iterative refinement by self-generated code. In this paper, we propose Adaptive Critique Refinement (ACR), which enables the model to refine itself by self-generated code and external critique, rather than directly imitating the code responses of the teacher model. Concretely, ACR includes a composite scoring system with LLM-as-a-Judge to evaluate the quality of code responses and a selective critique strategy with LLM-as-a-Critic to critique self-generated low-quality code responses. We develop the RefineCoder series by iteratively applying ACR, achieving continuous performance improvement on multiple code generation benchmarks. Compared to the baselines of the same size, our proposed RefineCoder series can achieve comparable or even superior performance using less data.

Xinyu Zhang, Wenjie Qiu, Yi-Chen Li et al.

Developing policies that can adjust to non-stationary environments is essential for real-world reinforcement learning applications. However, learning such adaptable policies in offline settings, with only a limited set of pre-collected trajectories, presents significant challenges. A key difficulty arises because the limited offline data makes it hard for the context encoder to differentiate between changes in the environment dynamics and shifts in the behavior policy, often leading to context misassociations. To address this issue, we introduce a novel approach called Debiased Offline Representation for fast online Adaptation (DORA). DORA incorporates an information bottleneck principle that maximizes mutual information between the dynamics encoding and the environmental data, while minimizing mutual information between the dynamics encoding and the actions of the behavior policy. We present a practical implementation of DORA, leveraging tractable bounds of the information bottleneck principle. Our experimental evaluation across six benchmark MuJoCo tasks with variable parameters demonstrates that DORA not only achieves a more precise dynamics encoding but also significantly outperforms existing baselines in terms of performance.

Haobo Yang, Shiyan Zhang, Zhuoyi Yang et al.

Deep perception networks in autonomous driving traditionally rely on data-intensive training regimes and post-hoc anomaly detection, often disregarding fundamental information-theoretic constraints governing stable information processing. We reconceptualize deep neural encoders as hierarchical communication chains that incrementally compress raw sensory inputs into task-relevant latent features. Within this framework, we establish two theoretically justified design principles for robust perception: (D1) smooth variation of mutual information between consecutive layers, and (D2) monotonic decay of latent entropy with network depth. Our analysis shows that, under realistic architectural assumptions, particularly blocks comprising repeated layers of similar capacity, enforcing smooth information flow (D1) naturally encourages entropy decay (D2), thus ensuring stable compression. Guided by these insights, we propose Eloss, a novel entropy-based regularizer designed as a lightweight, plug-and-play training objective. Rather than marginal accuracy improvements, this approach represents a conceptual shift: it unifies information-theoretic stability with standard perception tasks, enabling explicit, principled detection of anomalous sensor inputs through entropy deviations. Experimental validation on large-scale 3D object detection benchmarks (KITTI and nuScenes) demonstrates that incorporating Eloss consistently achieves competitive or improved accuracy while dramatically enhancing sensitivity to anomalies, amplifying distribution-shift signals by up to two orders of magnitude. This stable information-compression perspective not only improves interpretability but also establishes a solid theoretical foundation for safer, more robust autonomous driving perception systems.

Junjing Zheng, Chengliang Song, Weidong Jiang et al.

High-dimensional data, particularly in the form of high-order tensors, presents a major challenge in self-supervised learning. While MLP-based autoencoders (AE) are commonly employed, their dependence on flattening operations exacerbates the curse of dimensionality, leading to excessively large model sizes, high computational overhead, and challenging optimization for deep structural feature capture. Although existing tensor networks alleviate computational burdens through tensor decomposition techniques, most exhibit limited capability in learning non-linear relationships. To overcome these limitations, we introduce the Mode-Aware Non-linear Tucker Autoencoder (MA-NTAE). MA-NTAE generalized classical Tucker decomposition to a non-linear framework and employs a Pick-and-Unfold strategy, facilitating flexible per-mode encoding of high-order tensors via recursive unfold-encode-fold operations, effectively integrating tensor structural priors. Notably, MA-NTAE exhibits linear growth in computational complexity with tensor order and proportional growth with mode dimensions. Extensive experiments demonstrate MA-NTAE's performance advantages over standard AE and current tensor networks in compression and clustering tasks, which become increasingly pronounced for higher-order, higher-dimensional tensors.

Yuhan Liu, Xinyu Zhang, Haonan Chang et al.

This paper addresses the challenges of Rhythmic Insertion Tasks (RIT), where a robot must repeatedly perform high-precision insertions, such as screwing a nut into a bolt with a wrench. The inherent difficulty of RIT lies in achieving millimeter-level accuracy and maintaining consistent performance over multiple repetitions, particularly when factors like nut rotation and friction introduce additional complexity. We propose a sim-to-real framework that integrates a reinforcement learning-based insertion policy with a failure forecasting module. By representing the wrench's pose in the nut's coordinate frame rather than the robot's frame, our approach significantly enhances sim-to-real transferability. The insertion policy, trained in simulation, leverages real-time 6D pose tracking to execute precise alignment, insertion, and rotation maneuvers. Simultaneously, a neural network predicts potential execution failures, triggering a simple recovery mechanism that lifts the wrench and retries the insertion. Extensive experiments in both simulated and real-world environments demonstrate that our method not only achieves a high one-time success rate but also robustly maintains performance over long-horizon repetitive tasks.

Zhongjie Ba, Yitao Zhang, Peng Cheng et al.

Watermarking plays a key role in the provenance and detection of AI-generated content. While existing methods prioritize robustness against real-world distortions (e.g., JPEG compression and noise addition), we reveal a fundamental tradeoff: such robust watermarks inherently improve the redundancy of detectable patterns encoded into images, creating exploitable information leakage. To leverage this, we propose an attack framework that extracts leakage of watermark patterns through multi-channel feature learning using a pre-trained vision model. Unlike prior works requiring massive data or detector access, our method achieves both forgery and detection evasion with a single watermarked image. Extensive experiments demonstrate that our method achieves a 60\% success rate gain in detection evasion and 51\% improvement in forgery accuracy compared to state-of-the-art methods while maintaining visual fidelity. Our work exposes the robustness-stealthiness paradox: current "robust" watermarks sacrifice security for distortion resistance, providing insights for future watermark design.

Xinyu Zhang, Yuhan Liu, Haonan Chang et al.

Learning general-purpose models from diverse datasets has achieved great success in machine learning. In robotics, however, existing methods in multi-task learning are typically constrained to a single robot and workspace, while recent work such as RT-X requires a non-trivial action normalization procedure to manually bridge the gap between different action spaces in diverse environments. In this paper, we propose the visual kinematics chain as a precise and universal representation of quasi-static actions for robot learning over diverse environments, which requires no manual adjustment since the visual kinematic chains can be automatically obtained from the robot's model and camera parameters. We propose the Visual Kinematics Transformer (VKT), a convolution-free architecture that supports an arbitrary number of camera viewpoints, and that is trained with a single objective of forecasting kinematic structures through optimal point-set matching. We demonstrate the superior performance of VKT over BC transformers as a general agent on Calvin, RLBench, Open-X, and real robot manipulation tasks. Video demonstrations can be found at https://mlzxy.github.io/visual-kinetic-chain.

Zhenhong Zou, Xinyu Zhang, Huaping Liu et al.

There has recently been growing interest in utilizing multimodal sensors to achieve robust lane line segmentation. In this paper, we introduce a novel multimodal fusion architecture from an information theory perspective, and demonstrate its practical utility using Light Detection and Ranging (LiDAR) camera fusion networks. In particular, we develop, for the first time, a multimodal fusion network as a joint coding model, where each single node, layer, and pipeline is represented as a channel. The forward propagation is thus equal to the information transmission in the channels. Then, we can qualitatively and quantitatively analyze the effect of different fusion approaches. We argue the optimal fusion architecture is related to the essential capacity and its allocation based on the source and channel. To test this multimodal fusion hypothesis, we progressively determine a series of multimodal models based on the proposed fusion methods and evaluate them on the KITTI and the A2D2 datasets. Our optimal fusion network achieves 85%+ lane line accuracy and 98.7%+ overall. The performance gap among the models will inform continuing future research into development of optimal fusion algorithms for the deep multimodal learning community.

Sinan Tan, Huaping Liu, Di Guo et al.

Embodiment is an important characteristic for all intelligent agents (creatures and robots), while existing scene description tasks mainly focus on analyzing images passively and the semantic understanding of the scenario is separated from the interaction between the agent and the environment. In this work, we propose the Embodied Scene Description, which exploits the embodiment ability of the agent to find an optimal viewpoint in its environment for scene description tasks. A learning framework with the paradigms of imitation learning and reinforcement learning is established to teach the intelligent agent to generate corresponding sensorimotor activities. The proposed framework is tested on both the AI2Thor dataset and a real world robotic platform demonstrating the effectiveness and extendability of the developed method.

Hao Wu, Hanyuan Zhang, Xinyu Zhang et al.

Automatic map extraction is of great importance to urban computing and location-based services. Aerial image and GPS trajectory data refer to two different data sources that could be leveraged to generate the map, although they carry different types of information. Most previous works on data fusion between aerial images and data from auxiliary sensors do not fully utilize the information of both modalities and hence suffer from the issue of information loss. We propose a deep convolutional neural network called DeepDualMapper which fuses the aerial image and trajectory data in a more seamless manner to extract the digital map. We design a gated fusion module to explicitly control the information flows from both modalities in a complementary-aware manner. Moreover, we propose a novel densely supervised refinement decoder to generate the prediction in a coarse-to-fine way. Our comprehensive experiments demonstrate that DeepDualMapper can fuse the information of images and trajectories much more effectively than existing approaches, and is able to generate maps with higher accuracy.