Chenyang Li

Jun-Yan He, Zhi-Qi Cheng, Chenyang Li et al. · cmu, uw

Real-time perception, or streaming perception, is a crucial aspect of autonomous driving that has yet to be thoroughly explored in existing research. To address this gap, we present DAMO-StreamNet, an optimized framework that combines recent advances from the YOLO series with a comprehensive analysis of spatial and temporal perception mechanisms, delivering a cutting-edge solution. The key innovations of DAMO-StreamNet are (1) A robust neck structure incorporating deformable convolution, enhancing the receptive field and feature alignment capabilities (2) A dual-branch structure that integrates short-path semantic features and long-path temporal features, improving motion state prediction accuracy. (3) Logits-level distillation for efficient optimization, aligning the logits of teacher and student networks in semantic space. (4) A real-time forecasting mechanism that updates support frame features with the current frame, ensuring seamless streaming perception during inference. Our experiments demonstrate that DAMO-StreamNet surpasses existing state-of-the-art methods, achieving 37.8% (normal size (600, 960)) and 43.3% (large size (1200, 1920)) sAP without using extra data. This work not only sets a new benchmark for real-time perception but also provides valuable insights for future research. Additionally, DAMO-StreamNet can be applied to various autonomous systems, such as drones and robots, paving the way for real-time perception. The code is at https://github.com/zhiqic/DAMO-StreamNet.

Jiawen Zhu, Zhenyu Chen, Zeqi Hao et al.

Visual object tracking is a fundamental video task in computer vision. Recently, the notably increasing power of perception algorithms allows the unification of single/multiobject and box/mask-based tracking. Among them, the Segment Anything Model (SAM) attracts much attention. In this report, we propose HQTrack, a framework for High Quality Tracking anything in videos. HQTrack mainly consists of a video multi-object segmenter (VMOS) and a mask refiner (MR). Given the object to be tracked in the initial frame of a video, VMOS propagates the object masks to the current frame. The mask results at this stage are not accurate enough since VMOS is trained on several closeset video object segmentation (VOS) datasets, which has limited ability to generalize to complex and corner scenes. To further improve the quality of tracking masks, a pretrained MR model is employed to refine the tracking results. As a compelling testament to the effectiveness of our paradigm, without employing any tricks such as test-time data augmentations and model ensemble, HQTrack ranks the 2nd place in the Visual Object Tracking and Segmentation (VOTS2023) challenge. Code and models are available at https://github.com/jiawen-zhu/HQTrack.

Jin-Peng Lan, Zhi-Qi Cheng, Jun-Yan He et al. · cmu, uw

Existing Visual Object Tracking (VOT) only takes the target area in the first frame as a template. This causes tracking to inevitably fail in fast-changing and crowded scenes, as it cannot account for changes in object appearance between frames. To this end, we revamped the tracking framework with Progressive Context Encoding Transformer Tracker (ProContEXT), which coherently exploits spatial and temporal contexts to predict object motion trajectories. Specifically, ProContEXT leverages a context-aware self-attention module to encode the spatial and temporal context, refining and updating the multi-scale static and dynamic templates to progressively perform accurately tracking. It explores the complementary between spatial and temporal context, raising a new pathway to multi-context modeling for transformer-based trackers. In addition, ProContEXT revised the token pruning technique to reduce computational complexity. Extensive experiments on popular benchmark datasets such as GOT-10k and TrackingNet demonstrate that the proposed ProContEXT achieves state-of-the-art performance.

Jun-Yan He, Zhi-Qi Cheng, Chenyang Li et al. · cmu, uw

This paper introduces WordArt Designer, a user-driven framework for artistic typography synthesis, relying on the Large Language Model (LLM). The system incorporates four key modules: the LLM Engine, SemTypo, StyTypo, and TexTypo modules. 1) The LLM Engine, empowered by the LLM (e.g., GPT-3.5), interprets user inputs and generates actionable prompts for the other modules, thereby transforming abstract concepts into tangible designs. 2) The SemTypo module optimizes font designs using semantic concepts, striking a balance between artistic transformation and readability. 3) Building on the semantic layout provided by the SemTypo module, the StyTypo module creates smooth, refined images. 4) The TexTypo module further enhances the design's aesthetics through texture rendering, enabling the generation of inventive textured fonts. Notably, WordArt Designer highlights the fusion of generative AI with artistic typography. Experience its capabilities on ModelScope: https://www.modelscope.cn/studios/WordArt/WordArt.

Chenyang Li, Zhi-Qi Cheng, Jun-Yan He et al. · cmu, uw

Streaming perception is a critical task in autonomous driving that requires balancing the latency and accuracy of the autopilot system. However, current methods for streaming perception are limited as they only rely on the current and adjacent two frames to learn movement patterns. This restricts their ability to model complex scenes, often resulting in poor detection results. To address this limitation, we propose LongShortNet, a novel dual-path network that captures long-term temporal motion and integrates it with short-term spatial semantics for real-time perception. LongShortNet is notable as it is the first work to extend long-term temporal modeling to streaming perception, enabling spatiotemporal feature fusion. We evaluate LongShortNet on the challenging Argoverse-HD dataset and demonstrate that it outperforms existing state-of-the-art methods with almost no additional computational cost.

Jianfeng Huang, Chenyang Li, Yimin Lin et al.

It is hard to collect enough flaw images for training deep learning network in industrial production. Therefore, existing industrial anomaly detection methods prefer to use CNN-based unsupervised detection and localization network to achieve this task. However, these methods always fail when there are varieties happened in new signals since traditional end-to-end networks suffer barriers of fitting nonlinear model in high-dimensional space. Moreover, they have a memory library by clustering the feature of normal images essentially, which cause it is not robust to texture change. To this end, we propose the Vision Transformer based (VIT-based) unsupervised anomaly detection network. It utilizes a hierarchical task learning and human experience to enhance its interpretability. Our network consists of pattern generation and comparison networks. Pattern generation network uses two VIT-based encoder modules to extract the feature of two consecutive image patches, then uses VIT-based decoder module to learn the human designed style of these features and predict the third image patch. After this, we use the Siamese-based network to compute the similarity of the generation image patch and original image patch. Finally, we refine the anomaly localization by the bi-directional inference strategy. Comparison experiments on public dataset MVTec dataset show our method achieves 99.8% AUC, which surpasses previous state-of-the-art methods. In addition, we give a qualitative illustration on our own leather and cloth datasets. The accurate segment results strongly prove the accuracy of our method in anomaly detection.

Chenyang Li, Jihoon Chung, Mengnan Du et al.

Model compression is a crucial part of deploying neural networks (NNs), especially when the memory and storage of computing devices are limited in many applications. This paper focuses on two model compression techniques: low-rank approximation and weight pruning in neural networks, which are very popular nowadays. However, training NN with low-rank approximation and weight pruning always suffers significant accuracy loss and convergence issues. In this paper, a holistic framework is proposed for model compression from a novel perspective of nonconvex optimization by designing an appropriate objective function. Then, we introduce NN-BCD, a block coordinate descent (BCD) algorithm to solve the nonconvex optimization. One advantage of our algorithm is that an efficient iteration scheme can be derived with closed-form, which is gradient-free. Therefore, our algorithm will not suffer from vanishing/exploding gradient problems. Furthermore, with the Kurdyka-Łojasiewicz (KŁ) property of our objective function, we show that our algorithm globally converges to a critical point at the rate of O(1/k), where k denotes the number of iterations. Lastly, extensive experiments with tensor train decomposition and weight pruning demonstrate the efficiency and superior performance of the proposed framework. Our code implementation is available at https://github.com/ChenyangLi-97/NN-BCD

Zhongliang Jiang, Chenyang Li, Xuesong Li et al.

Autonomous ultrasound (US) imaging has gained increased interest recently, and it has been seen as a potential solution to overcome the limitations of free-hand US examinations, such as inter-operator variations. However, it is still challenging to accurately map planned paths from a generic atlas to individual patients, particularly for thoracic applications with high acoustic-impedance bone structures under the skin. To address this challenge, a graph-based non-rigid registration is proposed to enable transferring planned paths from the atlas to the current setup by explicitly considering subcutaneous bone surface features instead of the skin surface. To this end, the sternum and cartilage branches are segmented using a template matching to assist coarse alignment of US and CT point clouds. Afterward, a directed graph is generated based on the CT template. Then, the self-organizing map using geographical distance is successively performed twice to extract the optimal graph representations for CT and US point clouds, individually. To evaluate the proposed approach, five cartilage point clouds from distinct patients are employed. The results demonstrate that the proposed graph-based registration can effectively map trajectories from CT to the current setup for displaying US views through limited intercostal space. The non-rigid registration results in terms of Hausdorff distance (Mean$\pm$SD) is 9.48$\pm$0.27 mm and the path transferring error in terms of Euclidean distance is 2.21$\pm$1.11 mm.

Jiawen Zhu, Zhi-Qi Cheng, Jun-Yan He et al. · cmu, uw

Advances in perception modeling have significantly improved the performance of object tracking. However, the current methods for specifying the target object in the initial frame are either by 1) using a box or mask template, or by 2) providing an explicit language description. These manners are cumbersome and do not allow the tracker to have self-reasoning ability. Therefore, this work proposes a new tracking task -- Instruction Tracking, which involves providing implicit tracking instructions that require the trackers to perform tracking automatically in video frames. To achieve this, we investigate the integration of knowledge and reasoning capabilities from a Large Vision-Language Model (LVLM) for object tracking. Specifically, we propose a tracker called TrackGPT, which is capable of performing complex reasoning-based tracking. TrackGPT first uses LVLM to understand tracking instructions and condense the cues of what target to track into referring embeddings. The perception component then generates the tracking results based on the embeddings. To evaluate the performance of TrackGPT, we construct an instruction tracking benchmark called InsTrack, which contains over one thousand instruction-video pairs for instruction tuning and evaluation. Experiments show that TrackGPT achieves competitive performance on referring video object segmentation benchmarks, such as getting a new state-of the-art performance of 66.5 $\mathcal{J}\&\mathcal{F}$ on Refer-DAVIS. It also demonstrates a superior performance of instruction tracking under new evaluation protocols. The code and models are available at \href{https://github.com/jiawen-zhu/TrackGPT}{https://github.com/jiawen-zhu/TrackGPT}.

Jiayu Lin, Guanrong Chen, Bojun Jin et al.

In this paper we present the results of the AI-Debater 2023 Challenge held by the Chinese Conference on Affect Computing (CCAC 2023), and introduce the related datasets. We organize two tracks to handle the argumentative generation tasks in different scenarios, namely, Counter-Argument Generation (Track 1) and Claim-based Argument Generation (Track 2). Each track is equipped with its distinct dataset and baseline model respectively. In total, 32 competing teams register for the challenge, from which we received 11 successful submissions. In this paper, we will present the results of the challenge and a summary of the systems, highlighting commonalities and innovations among participating systems. Datasets and baseline models of the AI-Debater 2023 Challenge have been already released and can be accessed through the official website of the challenge.

Chenyang Li, Zhao Song, Weixin Wang et al.

The Deep Leakage from Gradient (DLG) attack has emerged as a prevalent and highly effective method for extracting sensitive training data by inspecting exchanged gradients. This approach poses a substantial threat to the privacy of individuals and organizations alike. This research presents a comprehensive analysis of the gradient leakage method when applied specifically to transformer-based models. Through meticulous examination, we showcase the capability to accurately recover data solely from gradients and rigorously investigate the conditions under which gradient attacks can be executed, providing compelling evidence. Furthermore, we reevaluate the approach of introducing additional noise on gradients as a protective measure against gradient attacks. To address this, we outline a theoretical proof that analyzes the associated privacy costs within the framework of differential privacy. Additionally, we affirm the convergence of the Stochastic Gradient Descent (SGD) algorithm under perturbed gradients. The primary objective of this study is to augment the understanding of gradient leakage attack and defense strategies while actively contributing to the development of privacy-preserving techniques specifically tailored for transformer-based models. By shedding light on the vulnerabilities and countermeasures associated with gradient leakage, this research aims to foster advancements in safeguarding sensitive data and upholding privacy in the context of transformer-based models.

Chenyang Li, Zhao Song, Zhaoxing Xu et al.

Leverage scores have become essential in statistics and machine learning, aiding regression analysis, randomized matrix computations, and various other tasks. This paper delves into the inverse problem, aiming to recover the intrinsic model parameters given the leverage scores gradient. This endeavor not only enriches the theoretical understanding of models trained with leverage score techniques but also has substantial implications for data privacy and adversarial security. We specifically scrutinize the inversion of the leverage score gradient, denoted as $g(x)$. An innovative iterative algorithm is introduced for the approximate resolution of the regularized least squares problem stated as $\min_{x \in \mathbb{R}^d} 0.5 \|g(x) - c\|_2^2 + 0.5\|\mathrm{diag}(w)Ax\|_2^2$. Our algorithm employs subsampled leverage score distributions to compute an approximate Hessian in each iteration, under standard assumptions, considerably mitigating the time complexity. Given that a total of $T = \log(\| x_0 - x^* \|_2/ ε)$ iterations are required, the cost per iteration is optimized to the order of $O( (\mathrm{nnz}(A) + d^ω ) \cdot \mathrm{poly}(\log(n/δ))$, where $\mathrm{nnz}(A)$ denotes the number of non-zero entries of $A$.

Reuben Docea, Rayan Younis, Yonghao Long et al.

The D4D Dataset provides paired endoscopic video and high-quality structured-light geometry for evaluating 3D reconstruction of deforming abdominal soft tissue in realistic surgical conditions. Data were acquired from six porcine cadaver sessions using a da Vinci Xi stereo endoscope and a Zivid structured-light camera, registered via optical tracking and manually curated iterative alignment methods. Three sequence types - whole deformations, incremental deformations, and moved-camera clips - probe algorithm robustness to non-rigid motion, deformation magnitude, and out-of-view updates. Each clip provides rectified stereo images, per-frame instrument masks, stereo depth, start/end structured-light point clouds, curated camera poses and camera intrinsics. In postprocessing, ICP and semi-automatic registration techniques are used to register data, and instrument masks are created. The dataset enables quantitative geometric evaluation in both visible and occluded regions, alongside photometric view-synthesis baselines. Comprising over 300,000 frames and 369 point clouds across 98 curated recordings, this resource can serve as a comprehensive benchmark for developing and evaluating non-rigid SLAM, 4D reconstruction, and depth estimation methods.

Raghav Addanki, Chenyang Li, Zhao Song et al.

Attention computation takes both the time complexity of $O(n^2)$ and the space complexity of $O(n^2)$ simultaneously, which makes deploying Large Language Models (LLMs) in streaming applications that involve long contexts requiring substantial computational resources. In recent OpenAI DevDay (Nov 6, 2023), OpenAI released a new model that is able to support a 128K-long document, in our paper, we focus on the memory-efficient issue when context length $n$ is much greater than 128K ($n \gg 2^d$). Considering a single-layer self-attention with Query, Key, and Value matrices $Q, K, V \in \mathbb{R}^{n \times d}$, the polynomial method approximates the attention output $T \in \mathbb{R}^{n \times d}$. It accomplishes this by constructing $U_1, U_2 \in \mathbb{R}^{n \times t}$ to expedite attention ${\sf Attn}(Q, K, V)$ computation within $n^{1+o(1)}$ time executions. Despite this, computing the approximated attention matrix $U_1U_2^\top \in \mathbb{R}^{n \times n}$ still necessitates $O(n^2)$ space, leading to significant memory usage. In response to these challenges, we introduce a new algorithm that only reads one pass of the data in a streaming fashion. This method employs sublinear space $o(n)$ to store three sketch matrices, alleviating the need for exact $K, V$ storage. Notably, our algorithm exhibits exceptional memory-efficient performance with super-long tokens. As the token length $n$ increases, our error guarantee diminishes while the memory usage remains nearly constant. This unique attribute underscores the potential of our technique in efficiently handling LLMs in streaming applications.

Chenyang Li, Wenxuan Liu, Guoqiang Gong et al.

Underwater object detection is critical for oceanic research and industrial safety inspections. However, the complex optical environment and the limited resources of underwater equipment pose significant challenges to achieving high accuracy and low power consumption. To address these issues, we propose Spiking Underwater YOLO (SU-YOLO), a Spiking Neural Network (SNN) model. Leveraging the lightweight and energy-efficient properties of SNNs, SU-YOLO incorporates a novel spike-based underwater image denoising method based solely on integer addition, which enhances the quality of feature maps with minimal computational overhead. In addition, we introduce Separated Batch Normalization (SeBN), a technique that normalizes feature maps independently across multiple time steps and is optimized for integration with residual structures to capture the temporal dynamics of SNNs more effectively. The redesigned spiking residual blocks integrate the Cross Stage Partial Network (CSPNet) with the YOLO architecture to mitigate spike degradation and enhance the model's feature extraction capabilities. Experimental results on URPC2019 underwater dataset demonstrate that SU-YOLO achieves mAP of 78.8% with 6.97M parameters and an energy consumption of 2.98 mJ, surpassing mainstream SNN models in both detection accuracy and computational efficiency. These results underscore the potential of SNNs for engineering applications. The code is available in https://github.com/lwxfight/snn-underwater.

Dianye Huang, Chenyang Li, Angelos Karlas et al.

Precise percutaneous needle detection is crucial for ultrasound (US)-guided interventions. However, inherent limitations such as speckles, needle-like artifacts, and low resolution make it challenging to robustly detect needles, especially when their visibility is reduced or imperceptible. To address this challenge, we propose VibNet, a learning-based framework designed to enhance the robustness and accuracy of needle detection in US images by leveraging periodic vibration applied externally to the needle shafts. VibNet integrates neural Short-Time Fourier Transform and Hough Transform modules to achieve successive sub-goals, including motion feature extraction in the spatiotemporal space, frequency feature aggregation, and needle detection in the Hough space. Due to the periodic subtle vibration, the features are more robust in the frequency domain than in the image intensity domain, making VibNet more effective than traditional intensity-based methods. To demonstrate the effectiveness of VibNet, we conducted experiments on distinct \textit{ex vivo} porcine and bovine tissue samples. The results obtained on porcine samples demonstrate that VibNet effectively detects needles even when their visibility is severely reduced, with a tip error of $1.61\pm1.56~mm$ compared to $8.15\pm9.98~mm$ for UNet and $6.63\pm7.58~mm$ for WNet, and a needle direction error of $1.64\pm1.86^{\circ}$ compared to $9.29\pm15.30^{\circ}$ for UNet and $8.54\pm17.92^{\circ}$ for WNet. Code: https://github.com/marslicy/VibNet.

Bo Shen, Marco Marena, Chenyang Li et al.

With recent missions such as advanced space-based observatories like the Solar Dynamics Observatory (SDO) and Parker Solar Probe, and ground-based telescopes like the Daniel K. Inouye Solar Telescope (DKIST), the volume, velocity, and variety of data have made solar physics enter a transformative era as solar physics big data (SPBD). With the recent advancement of deep computer vision, there are new opportunities in SPBD for tackling problems that were previously unsolvable. However, there are new challenges arising due to the inherent characteristics of SPBD and deep computer vision models. This vision paper presents an overview of the different types of SPBD, explores new opportunities in applying deep computer vision to SPBD, highlights the unique challenges, and outlines several potential future research directions.

Haoyang He, Zihua Rong, Kun Ji et al.

Reinforcement learning (RL) has recently become the dominant paradigm for strengthening the reasoning abilities of large language models (LLMs). Yet the rule-based reward functions commonly used on mathematical or programming benchmarks assess only answer format and correctness, providing no signal as to whether the induced Chain-of-Thought (CoT) actually improves the answer. Furthermore, such task-specific training offers limited control over logical depth and therefore may fail to reveal a model's genuine reasoning capacity. We propose Dynamic Reasoning Efficiency Reward (DRER) -- a plug-and-play RL reward framework that reshapes both reward and advantage signals. (i) A Reasoning Quality Reward assigns fine-grained credit to those reasoning chains that demonstrably raise the likelihood of the correct answer, directly incentivising the trajectories with beneficial CoT tokens. (ii) A Dynamic Length Advantage decays the advantage of responses whose length deviates from a validation-derived threshold, stabilising training. To facilitate rigorous assessment, we also release Logictree, a dynamically constructed deductive reasoning dataset that functions both as RL training data and as a comprehensive benchmark. Experiments confirm the effectiveness of DRER: our 7B model attains GPT-o3-mini level performance on Logictree with 400 trianing steps, while the average confidence of CoT-augmented answers rises by 30%. The model further exhibits generalisation across diverse logical-reasoning datasets, and the mathematical benchmark AIME24. These results illuminate how RL shapes CoT behaviour and chart a practical path toward enhancing formal-reasoning skills in large language models. All code and data are available in repository https://github.com/Henryhe09/DRER.

Chenyang Li, Jieyuan Liu, Bin Li et al.

Vision-Language Action (VLA) models have shown remarkable progress in robotic manipulation by leveraging the powerful perception abilities of Vision-Language Models (VLMs) to understand environments and directly output actions. However, by default, VLA models may overly attend to image tokens in the task-irrelevant region, which we describe as 'distracting tokens'. This behavior can disturb the model from the generation of the desired action tokens in each step, affecting the success rate of tasks. In this paper, we introduce a simple yet effective plug-and-play Distracting Token Pruning (DTP) framework, which dynamically detects and prunes these distracting image tokens. By correcting the model's visual attention patterns, we aim to improve the task success rate, as well as exploring the performance upper boundaries of the model without altering its original architecture or adding additional inputs. Experiments on the SIMPLER Benchmark (Li et al., 2024) show that our method consistently achieving relative improvements in task success rates across different types of novel VLA models, demonstrating generalizability to transformer-based VLAs. Further analysis reveals a negative correlation between the task success rate and the amount of attentions in the task-irrelevant region for all models tested, highlighting a common phenomenon of VLA models that could guide future research. We also publish our code at: https://anonymous.4open.science/r/CBD3.

Xingyu Li, Rongguang Wang, Yuying Wang et al.

Large language models (LLMs) remain brittle on multi-hop question answering (MHQA), where answering requires combining evidence across documents through retrieval and reasoning. Iterative retrieval systems can fail by locking onto an early low-recall trajectory and amplifying downstream errors, while planning-only approaches may produce static query sets that cannot adapt when intermediate evidence changes. We propose \textbf{Planned Active Retrieval and Reasoning RAG (PAR$^2$-RAG)}, a two-stage framework that separates \emph{coverage} from \emph{commitment}. PAR$^2$-RAG first performs breadth-first anchoring to build a high-recall evidence frontier, then applies depth-first refinement with evidence sufficiency control in an iterative loop. Across four MHQA benchmarks, PAR$^2$-RAG consistently outperforms existing state-of-the-art baselines, compared with IRCoT, PAR$^2$-RAG achieves up to \textbf{23.5\%} higher accuracy, with retrieval gains of up to \textbf{10.5\%} in NDCG.

Peng Liu, Bianca Güttner, Yutong Su et al.

Non-rigid registration is essential for Augmented Reality guided laparoscopic liver surgery by fusing preoperative information, such as tumor location and vascular structures, into the limited intraoperative view, thereby enhancing surgical navigation. A prerequisite is the accurate prediction of intraoperative liver deformation which remains highly challenging due to factors such as large deformation caused by pneumoperitoneum, respiration and tool interaction as well as noisy intraoperative data, and limited field of view due to occlusion and constrained camera movement. To address these challenges, we introduce PIVOTS, a Preoperative to Intraoperative VOlume-To-Surface registration neural network that directly takes point clouds as input for deformation prediction. The geometric feature extraction encoder allows multi-resolution feature extraction, and the decoder, comprising novel deformation aware cross attention modules, enables pre- and intraoperative information interaction and accurate multi-level displacement prediction. We train the neural network on synthetic data simulated from a biomechanical simulation pipeline and validate its performance on both synthetic and real datasets. Results demonstrate superior registration performance of our method compared to baseline methods, exhibiting strong robustness against high amounts of noise, large deformation, and various levels of intraoperative visibility. We publish the training and test sets as evaluation benchmarks and call for a fair comparison of liver registration methods with volume-to-surface data. Code and datasets are available here https://github.com/pengliu-nct/PIVOTS.

Huiyu Zhai, Xingxing Yang, Yalan Ye et al.

Facial expression recognition (FER) is a challenging task due to pervasive occlusion and dataset biases. Especially when facial information is partially occluded, existing FER models struggle to extract effective facial features, leading to inaccurate classifications. In response, we present ORSANet, which introduces the following three key contributions: First, we introduce auxiliary multi-modal semantic guidance to disambiguate facial occlusion and learn high-level semantic knowledge, which is two-fold: 1) we introduce semantic segmentation maps as dense semantics prior to generate semantics-enhanced facial representations; 2) we introduce facial landmarks as sparse geometric prior to mitigate intrinsic noises in FER, such as identity and gender biases. Second, to facilitate the effective incorporation of these two multi-modal priors, we customize a Multi-scale Cross-interaction Module (MCM) to adaptively fuse the landmark feature and semantics-enhanced representations within different scales. Third, we design a Dynamic Adversarial Repulsion Enhancement Loss (DARELoss) that dynamically adjusts the margins of ambiguous classes, further enhancing the model's ability to distinguish similar expressions. We further construct the first occlusion-oriented FER dataset to facilitate specialized robustness analysis on various real-world occlusion conditions, dubbed Occlu-FER. Extensive experiments on both public benchmarks and Occlu-FER demonstrate that our proposed ORSANet achieves SOTA recognition performance. Code is publicly available at https://github.com/Wenyuzhy/ORSANet-master.

Xu Bao, Zhi-Qi Cheng, Jun-Yan He et al.

Accurate facial landmark detection is critical for facial analysis tasks, yet prevailing heatmap and coordinate regression methods grapple with prohibitive computational costs and quantization errors. Through comprehensive theoretical analysis and experimentation, we identify and elucidate the limitations of existing techniques. To overcome these challenges, we pioneer the application of True-Range Multilateration, originally devised for GPS localization, to facial landmark detection. We propose KeyPoint Positioning System (KeyPosS) - the first framework to deduce exact landmark coordinates by triangulating distances between points of interest and anchor points predicted by a fully convolutional network. A key advantage of KeyPosS is its plug-and-play nature, enabling flexible integration into diverse decoding pipelines. Extensive experiments on four datasets demonstrate state-of-the-art performance, with KeyPosS outperforming existing methods in low-resolution settings despite minimal computational overhead. By spearheading the integration of Multilateration with facial analysis, KeyPosS marks a paradigm shift in facial landmark detection. The code is available at https://github.com/zhiqic/KeyPosS.

Ziqian Zhang, Nana Bao, Xingting Yan et al.

Various time variant non-stationary signals need to be pre-processed properly in hydrological time series forecasting in real world, for example, predictions of water level. Decomposition method is a good candidate and widely used in such a pre-processing problem. However, decomposition methods with an inappropriate sampling technique may introduce future data which is not available in practical applications, and result in incorrect decomposition-based forecasting models. In this work, a novel Fully Stepwise Decomposition-Based (FSDB) sampling technique is well designed for the decomposition-based forecasting model, strictly avoiding introducing future information. This sampling technique with decomposition methods, such as Variational Mode Decomposition (VMD) and Singular spectrum analysis (SSA), is applied to predict water level time series in three different stations of Guoyang and Chaohu basins in China. Results of VMD-based hybrid model using FSDB sampling technique show that Nash-Sutcliffe Efficiency (NSE) coefficient is increased by 6.4%, 28.8% and 7.0% in three stations respectively, compared with those obtained from the currently most advanced sampling technique. In the meantime, for series of SSA-based experiments, NSE is increased by 3.2%, 3.1% and 1.1% respectively. We conclude that the newly developed FSDB sampling technique can be used to enhance the performance of decomposition-based hybrid model in water level time series forecasting in real world.

Ruizhong Miao, Yuying Wang, Rongguang Wang et al.

Semantic search in retrieval-augmented generation (RAG) systems is often insufficient for complex information needs, particularly when relevant evidence is scattered across multiple sources. Prior approaches to this problem include agentic retrieval strategies, which expand the semantic search space by generating additional queries. However, these methods do not fully leverage the organizational structure of the data and instead rely on iterative exploration, which can lead to inefficient retrieval. Another class of approaches employs knowledge graphs to model non-semantic relationships through graph edges. Although effective in capturing richer proximities, such methods incur significant maintenance costs and are often incompatible with the vector stores used in most production systems. To address these limitations, we propose GraphER, a graph-based enrichment and reranking method that captures multiple forms of proximity beyond semantic similarity. GraphER independently enriches data objects during offline indexing and performs graph-based reranking over candidate objects at query time. This design does not require a knowledge graph, allowing GraphER to integrate seamlessly with standard vector stores. In addition, GraphER is retriever-agnostic and introduces negligible latency overhead. Experiments on multiple retrieval benchmarks demonstrate the effectiveness of the proposed approach.

Chenyang Li, Yingyu Liang, Zhenmei Shi et al.

In the evolving landscape of machine learning, a pivotal challenge lies in deciphering the internal representations harnessed by neural networks and Transformers. Building on recent progress toward comprehending how networks execute distinct target functions, our study embarks on an exploration of the underlying reasons behind networks adopting specific computational strategies. We direct our focus to the complex algebraic learning task of modular addition involving $k$ inputs. Our research presents a thorough analytical characterization of the features learned by stylized one-hidden layer neural networks and one-layer Transformers in addressing this task. A cornerstone of our theoretical framework is the elucidation of how the principle of margin maximization shapes the features adopted by one-hidden layer neural networks. Let $p$ denote the modulus, $D_p$ denote the dataset of modular arithmetic with $k$ inputs and $m$ denote the network width. We demonstrate that a neuron count of $ m \geq 2^{2k-2} \cdot (p-1) $, these networks attain a maximum $ L_{2,k+1} $-margin on the dataset $ D_p $. Furthermore, we establish that each hidden-layer neuron aligns with a specific Fourier spectrum, integral to solving modular addition problems. By correlating our findings with the empirical observations of similar studies, we contribute to a deeper comprehension of the intrinsic computational mechanisms of neural networks. Furthermore, we observe similar computational mechanisms in attention matrices of one-layer Transformers. Our work stands as a significant stride in unraveling their operation complexities, particularly in the realm of complex algebraic tasks.

Jerry Yao-Chieh Hu, Wei-Po Wang, Ammar Gilani et al.

We investigate the statistical and computational limits of prompt tuning for transformer-based foundation models. Our key contributions are prompt tuning on \emph{single-head} transformers with only a \emph{single} self-attention layer: (i) is universal, and (ii) supports efficient (even almost-linear time) algorithms under the Strong Exponential Time Hypothesis (SETH). Statistically, we prove that prompt tuning on such simplest possible transformers are universal approximators for sequence-to-sequence Lipschitz functions. In addition, we provide an exponential-in-$dL$ and -in-$(1/ε)$ lower bound on the required soft-prompt tokens for prompt tuning to memorize any dataset with 1-layer, 1-head transformers. Computationally, we identify a phase transition in the efficiency of prompt tuning, determined by the norm of the \emph{soft-prompt-induced} keys and queries, and provide an upper bound criterion. Beyond this criterion, no sub-quadratic (efficient) algorithm for prompt tuning exists under SETH. Within this criterion, we showcase our theory by proving the existence of almost-linear time prompt tuning inference algorithms. These fundamental limits provide important necessary conditions for designing expressive and efficient prompt tuning methods for practitioners.

Jiawen Zhu, Xin Chen, Haiwen Diao et al.

The speed-precision trade-off is a critical problem for visual object tracking which usually requires low latency and deployment on constrained resources. Existing solutions for efficient tracking mainly focus on adopting light-weight backbones or modules, which nevertheless come at the cost of a sacrifice in precision. In this paper, inspired by dynamic network routing, we propose DyTrack, a dynamic transformer framework for efficient tracking. Real-world tracking scenarios exhibit diverse levels of complexity. We argue that a simple network is sufficient for easy frames in video sequences, while more computation could be assigned to difficult ones. DyTrack automatically learns to configure proper reasoning routes for various inputs, gaining better utilization of the available computational budget. Thus, it can achieve higher performance with the same running speed. We formulate instance-specific tracking as a sequential decision problem and attach terminating branches to intermediate layers of the entire model. Especially, to fully utilize the computations, we introduce the feature recycling mechanism to reuse the outputs of predecessors. Furthermore, a target-aware self-distillation strategy is designed to enhance the discriminating capabilities of early predictions by effectively mimicking the representation pattern of the deep model. Extensive experiments on multiple benchmarks demonstrate that DyTrack achieves promising speed-precision trade-offs with only a single model. For instance, DyTrack obtains 64.9% AUC on LaSOT with a speed of 256 fps.

Dapeng Zhang, Dayu Chen, Peng Zhi et al.

Constructing online High-Definition (HD) maps is crucial for the static environment perception of autonomous driving systems (ADS). Existing solutions typically attempt to detect vectorized HD map elements with unified models; however, these methods often overlook the distinct characteristics of different non-cubic map elements, making accurate distinction challenging. To address these issues, we introduce an expert-based online HD map method, termed MapExpert. MapExpert utilizes sparse experts, distributed by our routers, to describe various non-cubic map elements accurately. Additionally, we propose an auxiliary balance loss function to distribute the load evenly across experts. Furthermore, we theoretically analyze the limitations of prevalent bird's-eye view (BEV) feature temporal fusion methods and introduce an efficient temporal fusion module called Learnable Weighted Moving Descentage. This module effectively integrates relevant historical information into the final BEV features. Combined with an enhanced slice head branch, the proposed MapExpert achieves state-of-the-art performance and maintains good efficiency on both nuScenes and Argoverse2 datasets.

Chenyang Li, Yingyu Liang, Zhenmei Shi et al.

The weighted low-rank approximation problem is a fundamental numerical linear algebra problem and has many applications in machine learning. Given a $n \times n$ weight matrix $W$ and a $n \times n$ matrix $A$, the goal is to find two low-rank matrices $U, V \in \mathbb{R}^{n \times k}$ such that the cost of $\| W \circ (U V^\top - A) \|_F^2$ is minimized. Previous work has to pay $Ω(n^2)$ time when matrices $A$ and $W$ are dense, e.g., having $Ω(n^2)$ non-zero entries. In this work, we show that there is a certain regime, even if $A$ and $W$ are dense, we can still hope to solve the weighted low-rank approximation problem in almost linear $n^{1+o(1)}$ time.

Jun-Yan He, Zhi-Qi Cheng, Chenyang Li et al. · cmu, uw

This paper introduces the WordArt Designer API, a novel framework for user-driven artistic typography synthesis utilizing Large Language Models (LLMs) on ModelScope. We address the challenge of simplifying artistic typography for non-professionals by offering a dynamic, adaptive, and computationally efficient alternative to traditional rigid templates. Our approach leverages the power of LLMs to understand and interpret user input, facilitating a more intuitive design process. We demonstrate through various case studies how users can articulate their aesthetic preferences and functional requirements, which the system then translates into unique and creative typographic designs. Our evaluations indicate significant improvements in user satisfaction, design flexibility, and creative expression over existing systems. The WordArt Designer API not only democratizes the art of typography but also opens up new possibilities for personalized digital communication and design.

Chenyang Li, Tanmay Sunil Kapure, Prokash Chandra Roy et al.

Fatigue life characterizes the duration a material can function before failure under specific environmental conditions, and is traditionally assessed using stress-life (S-N) curves. While machine learning and deep learning offer promising results for fatigue life prediction, they face the overfitting challenge because of the small size of fatigue experimental data in specific materials. To address this challenge, we propose, DeepOFormer, by formulating S-N curve prediction as an operator learning problem. DeepOFormer improves the deep operator learning framework with a transformer-based encoder and a mean L2 relative error loss function. We also consider Stussi, Weibull, and Pascual and Meeker (PM) features as domain-informed features. These features are motivated by empirical fatigue models. To evaluate the performance of our DeepOFormer, we compare it with different deep learning models and XGBoost on a dataset with 54 S-N curves of aluminum alloys. With seven different aluminum alloys selected for testing, our DeepOFormer achieves an R2 of 0.9515, a mean absolute error of 0.2080, and a mean relative error of 0.5077, significantly outperforming state-of-the-art deep/machine learning methods including DeepONet, TabTransformer, and XGBoost, etc. The results highlight that our Deep0Former integrating with domain-informed features substantially improves prediction accuracy and generalization capabilities for fatigue life prediction in aluminum alloys.

Mingyu Liu, Nana Bao, Xingting Yan et al.

Understanding the combined influences of meteorological and hydrological factors on water level and flood events is essential, particularly in today's changing climate environments. Transformer, as one kind of the cutting-edge deep learning methods, offers an effective approach to model intricate nonlinear processes, enables the extraction of key features and water level predictions. EXplainable Artificial Intelligence (XAI) methods play important roles in enhancing the understandings of how different factors impact water level. In this study, we propose a Transformer variant by integrating sparse attention mechanism and introducing nonlinear output layer for the decoder module. The variant model is utilized for multi-step forecasting of water level, by considering meteorological and hydrological factors simultaneously. It is shown that the variant model outperforms traditional Transformer across different lead times with respect to various evaluation metrics. The sensitivity analyses based on XAI technology demonstrate the significant influence of meteorological factors on water level evolution, in which temperature is shown to be the most dominant meteorological factor. Therefore, incorporating both meteorological and hydrological factors is necessary for reliable hydrological prediction and flood prevention. In the meantime, XAI technology provides insights into certain predictions, which is beneficial for understanding the prediction results and evaluating the reasonability.

Jinsong Chen, Chenyang Li, GaiChao Li et al.

Node tokenized graph Transformers (GTs) have shown promising performance in node classification. The generation of token sequences is the key module in existing tokenized GTs which transforms the input graph into token sequences, facilitating the node representation learning via Transformer. In this paper, we observe that the generations of token sequences in existing GTs only focus on the first-order neighbors on the constructed similarity graphs, which leads to the limited usage of nodes to generate diverse token sequences, further restricting the potential of tokenized GTs for node classification. To this end, we propose a new method termed SwapGT. SwapGT first introduces a novel token swapping operation based on the characteristics of token sequences that fully leverages the semantic relevance of nodes to generate more informative token sequences. Then, SwapGT leverages a Transformer-based backbone to learn node representations from the generated token sequences. Moreover, SwapGT develops a center alignment loss to constrain the representation learning from multiple token sequences, further enhancing the model performance. Extensive empirical results on various datasets showcase the superiority of SwapGT for node classification.

Junwen He, Yifan Wang, Lijun Wang et al.

Text logo design heavily relies on the creativity and expertise of professional designers, in which arranging element layouts is one of the most important procedures. However, this specific task has received limited attention, often overshadowed by broader layout generation tasks such as document or poster design. In this paper, we propose a Vision-Language Model (VLM)-based framework that generates content-aware text logo layouts by integrating multi-modal inputs with user-defined constraints, enabling more flexible and robust layout generation for real-world applications. We introduce two model techniques that reduce the computational cost for processing multiple glyph images simultaneously, without compromising performance. To support instruction tuning of our model, we construct two extensive text logo datasets that are five times larger than existing public datasets. In addition to geometric annotations (\textit{e.g.}, text masks and character recognition), our datasets include detailed layout descriptions in natural language, enabling the model to reason more effectively in handling complex designs and custom user inputs. Experimental results demonstrate the effectiveness of our proposed framework and datasets, outperforming existing methods on various benchmarks that assess geometric aesthetics and human preferences.

Xiang Huang, Zhi-Qi Cheng, Jun-Yan He et al. · cmu, uw

The advancement of autonomous driving systems hinges on the ability to achieve low-latency and high-accuracy perception. To address this critical need, this paper introduces Dynamic Routing Network (DyRoNet), a low-rank enhanced dynamic routing framework designed for streaming perception in autonomous driving systems. DyRoNet integrates a suite of pre-trained branch networks, each meticulously fine-tuned to function under distinct environmental conditions. At its core, the framework offers a speed router module, developed to assess and route input data to the most suitable branch for processing. This approach not only addresses the inherent limitations of conventional models in adapting to diverse driving conditions but also ensures the balance between performance and efficiency. Extensive experimental evaluations demonstrate the adaptability of DyRoNet to diverse branch selection strategies, resulting in significant performance enhancements across different scenarios. This work establishes a new benchmark for streaming perception and provides valuable engineering insights for future work.

Yida Cai, Ranjuexiao Hu, Huiyuan Xie et al.

Legal relations form a highly consequential analytical framework of civil law system, serving as a crucial foundation for resolving disputes and realizing values of the rule of law in judicial practice. However, legal relations in Chinese civil cases remain underexplored in the field of legal artificial intelligence (legal AI), largely due to the absence of comprehensive schemas. In this work, we firstly introduce a comprehensive schema, which contains a hierarchical taxonomy and definitions of arguments, for AI systems to capture legal relations in Chinese civil cases. Based on this schema, we then formulate legal relation extraction task and present LexRel, an expert-annotated benchmark for legal relation extraction in Chinese civil law. We use LexRel to evaluate state-of-the-art large language models (LLMs) on legal relation extractions, showing that current LLMs exhibit significant limitations in accurately identifying civil legal relations. Furthermore, we demonstrate that incorporating legal relations information leads to consistent performance gains on other downstream legal AI tasks.

Xin Zhang, Xingyu Li, Rongguang Wang et al.

Accurate chart comprehension represents a critical challenge in advancing multimodal learning systems, as extensive information is compressed into structured visual representations. However, existing vision-language models (VLMs) frequently struggle to generalize on unseen charts because it requires abstract, symbolic, and quantitative reasoning over structured visual representations. In this work, we introduce Chart-RL, an effective reinforcement learning (RL) method that employs mathematically verifiable rewards to enhance chart question answering in VLMs. Our experiments demonstrate that Chart-RL consistently outperforms supervised fine-tuning (SFT) across different chart understanding benchmarks, achieving relative improvements of 16.7% on MutlChartQA, and 11.5% on ChartInsights. We conduct robustness analysis, where Chart-RL achieves enhanced performance in 18 of 25 perturbed chart categories, demonstrating strong consistency and reasoning capability across visual variations. Furthermore, we demonstrate that task difficulty and inherent complexity are more critical than data quantity in RL training. For instance, Chart-RL trained on merely 10 complex chart-query examples significantly outperforms models trained on over 6,000 simple examples. Additionally, training on challenging reasoning tasks not only improves in-domain generalization relative to simpler tasks, but also facilitate strong transfer to out-of-domain visual mathematical problems.

Haolong Yan, Jia Wang, Xin Huang et al.

Recent advances in multimodal large language models unlock unprecedented opportunities for GUI automation. However, a fundamental challenge remains: how to efficiently acquire high-quality training data while maintaining annotation reliability? We introduce a self-evolving training pipeline powered by the Calibrated Step Reward System, which converts model-generated trajectories into reliable training signals through trajectory-level calibration, achieving >90% annotation accuracy with 10-100x lower cost. Leveraging this pipeline, we introduce Step-GUI, a family of models (4B/8B) that achieves state-of-the-art GUI performance (8B: 80.2% AndroidWorld, 48.5% OSWorld, 62.6% ScreenShot-Pro) while maintaining robust general capabilities. As GUI agent capabilities improve, practical deployment demands standardized interfaces across heterogeneous devices while protecting user privacy. To this end, we propose GUI-MCP, the first Model Context Protocol for GUI automation with hierarchical architecture that combines low-level atomic operations and high-level task delegation to local specialist models, enabling high-privacy execution where sensitive data stays on-device. Finally, to assess whether agents can handle authentic everyday usage, we introduce AndroidDaily, a benchmark grounded in real-world mobile usage patterns with 3146 static actions and 235 end-to-end tasks across high-frequency daily scenarios (8B: static 89.91%, end-to-end 52.50%). Our work advances the development of practical GUI agents and demonstrates strong potential for real-world deployment in everyday digital interactions.

Chenyang Li, Wenbing Tang, Yihao Huang et al.

Vision-and-Language Navigation (VLN) agents have made remarkable progress, but their robustness remains insufficiently studied. Existing adversarial evaluations often rely on perturbations that manifest as unusual textures rarely encountered in everyday indoor environments. Errors under such contrived conditions have limited practical relevance, as real-world agents are unlikely to encounter such artificial patterns. In this work, we focus on indoor lighting, an intrinsic yet largely overlooked scene attribute that strongly influences navigation. We propose Indoor Lighting-based Adversarial Attack (ILA), a black-box framework that manipulates global illumination to disrupt VLN agents. Motivated by typical household lighting usage, we design two attack modes: Static Indoor Lighting-based Attack (SILA), where the lighting intensity remains constant throughout an episode, and Dynamic Indoor Lighting-based Attack (DILA), where lights are switched on or off at critical moments to induce abrupt illumination changes. We evaluate ILA on two state-of-the-art VLN models across three navigation tasks. Results show that ILA significantly increases failure rates while reducing trajectory efficiency, revealing previously unrecognized vulnerabilities of VLN agents to realistic indoor lighting variations.

Huiyuan Xie, Chenyang Li, Huining Zhu et al.

Legal reasoning is a fundamental component of legal analysis and decision-making. Existing computational approaches to legal reasoning predominantly rely on generic reasoning frameworks such as syllogism and IRAC, which do not comprehensively examine the nuanced processes that underpin legal reasoning. Moreover, current research has largely focused on criminal cases, with insufficient modeling for civil cases. In this work, we present a novel framework for explicitly modeling legal reasoning in the analysis of Chinese tort-related civil cases. We first operationalize the legal reasoning processes used in tort analysis into the LawChain framework. LawChain is a three-module reasoning framework, with each module consisting of multiple finer-grained sub-steps. Informed by the LawChain framework, we introduce the task of tort legal reasoning and construct an evaluation benchmark, LawChain$_{eval}$, to systematically assess the critical steps within analytical reasoning chains for tort analysis. Leveraging this benchmark, we evaluate state-of-the-art large language models for their legal reasoning ability in civil tort contexts. Our results indicate that current models still fall short in accurately handling crucial elements of tort legal reasoning. Furthermore, we introduce several baseline approaches that explicitly incorporate LawChain-style reasoning through prompting or post-training. We conduct further experiments on additional legal analysis tasks, such as Legal Named-Entity Recognition and Criminal Damages Calculation, to verify the generalizability of these baselines. The proposed baseline approaches achieve significant improvements in tort-related legal reasoning and generalize well to related legal analysis tasks, thus demonstrating the value of explicitly modeling legal reasoning chains to enhance the reasoning capabilities of language models.

Chenyang Li, Qin Li, Haimin Wang et al.

High-resolution (HR) solar imaging is crucial for capturing fine-scale dynamic features such as filaments and fibrils. However, the spatial resolution of the full-disk H$α$ images is limited and insufficient to resolve these small-scale structures. To address this, we propose a GAN-based superresolution approach to enhance low-resolution (LR) full-disk H$α$ images from the Global Oscillation Network Group (GONG) to a quality comparable with HR observations from the Big Bear Solar Observatory/Goode Solar Telescope (BBSO/GST). We employ Real-ESRGAN with Residual-in-Residual Dense Blocks and a relativistic discriminator. We carefully aligned GONG-GST pairs. The model effectively recovers fine details within sunspot penumbrae and resolves fine details in filaments and fibrils, achieving an average mean squared error (MSE) of 467.15, root mean squared error (RMSE) of 21.59, and cross-correlation (CC) of 0.7794. Slight misalignments between image pairs limit quantitative performance, which we plan to address in future work alongside dataset expansion to further improve reconstruction quality.

Zhongyu Chen, Chenyang Li, Xuesong Li et al.

Precise needle alignment is essential for percutaneous needle insertion in robotic ultrasound-guided procedures. However, inherent challenges such as speckle noise, needle-like artifacts, and low image resolution make robust needle detection difficult, particularly when visibility is reduced or lost. In this paper, we propose a method to restore needle alignment when the ultrasound imaging plane and the needle insertion plane are misaligned. Unlike many existing approaches that rely heavily on needle visibility in ultrasound images, our method uses a more robust feature by periodically vibrating the needle using a mechanical system. Specifically, we propose a vibration-based energy metric that remains effective even when the needle is fully out of plane. Using this metric, we develop a control strategy to reposition the ultrasound probe in response to misalignments between the imaging plane and the needle insertion plane in both translation and rotation. Experiments conducted on ex-vivo porcine tissue samples using a dual-arm robotic ultrasound-guided needle insertion system demonstrate the effectiveness of the proposed approach. The experimental results show the translational error of 0.41$\pm$0.27 mm and the rotational error of 0.51$\pm$0.19 degrees.

Chenyang Li, Jinsong Chen, John E. Hopcroft et al.

Neighborhood-aware tokenized graph Transformers have recently shown great potential for node classification tasks. Despite their effectiveness, our in-depth analysis of neighborhood tokens reveals two critical limitations in the existing paradigm. First, current neighborhood token generation methods fail to adequately capture attribute correlations within a neighborhood. Second, the conventional self-attention mechanism suffers from attention diversion when processing neighborhood tokens, where high-hop neighborhoods receive disproportionate focus, severely disrupting information interactions between the target node and its neighborhood tokens. To address these challenges, we propose DAM-GT, Dual positional encoding-based Attention Masking graph Transformer. DAM-GT introduces a novel dual positional encoding scheme that incorporates attribute-aware encoding via an attribute clustering strategy, effectively preserving node correlations in both topological and attribute spaces. In addition, DAM-GT formulates a new attention mechanism with a simple yet effective masking strategy to guide interactions between target nodes and their neighborhood tokens, overcoming the issue of attention diversion. Extensive experiments on various graphs with different homophily levels as well as different scales demonstrate that DAM-GT consistently outperforms state-of-the-art methods in node classification tasks.

Jun-Yan He, Zhi-Qi Cheng, Chenyang Li et al.

MetaDesigner introduces a transformative framework for artistic typography synthesis, powered by Large Language Models (LLMs) and grounded in a user-centric design paradigm. Its foundation is a multi-agent system comprising the Pipeline, Glyph, and Texture agents, which collectively orchestrate the creation of customizable WordArt, ranging from semantic enhancements to intricate textural elements. A central feedback mechanism leverages insights from both multimodal models and user evaluations, enabling iterative refinement of design parameters. Through this iterative process, MetaDesigner dynamically adjusts hyperparameters to align with user-defined stylistic and thematic preferences, consistently delivering WordArt that excels in visual quality and contextual resonance. Empirical evaluations underscore the system's versatility and effectiveness across diverse WordArt applications, yielding outputs that are both aesthetically compelling and context-sensitive.

Zhongliang Jiang, Yunfeng Kang, Yuan Bi et al.

Ultrasound imaging has been widely used in clinical examinations owing to the advantages of being portable, real-time, and radiation-free. Considering the potential of extensive deployment of autonomous examination systems in hospitals, robotic US imaging has attracted increased attention. However, due to the inter-patient variations, it is still challenging to have an optimal path for each patient, particularly for thoracic applications with limited acoustic windows, e.g., intercostal liver imaging. To address this problem, a class-aware cartilage bone segmentation network with geometry-constraint post-processing is presented to capture patient-specific rib skeletons. Then, a dense skeleton graph-based non-rigid registration is presented to map the intercostal scanning path from a generic template to individual patients. By explicitly considering the high-acoustic impedance bone structures, the transferred scanning path can be precisely located in the intercostal space, enhancing the visibility of internal organs by reducing the acoustic shadow. To evaluate the proposed approach, the final path mapping performance is validated on five distinct CTs and two volunteer US data, resulting in ten pairs of CT-US combinations. Results demonstrate that the proposed graph-based registration method can robustly and precisely map the path from CT template to individual patients (Euclidean error: $2.21\pm1.11~mm$).

Jiuxiang Gu, Chenyang Li, Yingyu Liang et al.

The softmax activation function plays a crucial role in the success of large language models (LLMs), particularly in the self-attention mechanism of the widely adopted Transformer architecture. However, the underlying learning dynamics that contribute to the effectiveness of softmax remain largely unexplored. As a step towards better understanding, this paper provides a theoretical study of the optimization and generalization properties of two-layer softmax neural networks, providing theoretical insights into their superior performance as other activation functions, such as ReLU and exponential. Leveraging the Neural Tangent Kernel (NTK) framework, our analysis reveals that the normalization effect of the softmax function leads to a good perturbation property of the induced NTK matrix, resulting in a good convex region of the loss landscape. Consequently, softmax neural networks can learn the target function in the over-parametrization regime. To demonstrate the broad applicability of our theoretical findings, we apply them to the task of learning score estimation functions in diffusion models, a promising approach for generative modeling. Our analysis shows that gradient-based algorithms can learn the score function with a provable accuracy. Our work provides a deeper understanding of the effectiveness of softmax neural networks and their potential in various domains, paving the way for further advancements in natural language processing and beyond.

Chenyang Li, Gongxu Luo

Although the multilingual Neural Machine Translation(NMT), which extends Google's multilingual NMT, has ability to perform zero-shot translation and the iterative self-learning algorithm can improve the quality of zero-shot translation, it confronts with two problems: the multilingual NMT model is prone to generate wrong target language when implementing zero-shot translation; the self-learning algorithm, which uses beam search to generate synthetic parallel data, demolishes the diversity of the generated source language and amplifies the impact of the same noise during the iterative learning process. In this paper, we propose the tagged-multilingual NMT model and improve the self-learning algorithm to handle these two problems. Firstly, we extend the Google's multilingual NMT model and add target tokens to the target languages, which associates the start tag with the target language to ensure that the source language can be translated to the required target language. Secondly, we improve the self-learning algorithm by replacing beam search with random sample to increases the diversity of the generated data and makes it properly cover the true data distribution. Experimental results on IWSLT show that the adjusted tagged-multilingual NMT separately obtains 9.41 and 7.85 BLEU scores over the multilingual NMT on 2010 and 2017 Romanian-Italian test sets. Similarly, it obtains 9.08 and 7.99 BLEU scores on Italian-Romanian zero-shot translation. Furthermore, the improved self-learning algorithm shows its superiorities over the conventional self-learning algorithm on zero-shot translations.

Jingchao Su, Xu Chen, Ya Zhang et al.

Relational learning aims to make relation inference by exploiting the correlations among different types of entities. Exploring relational learning on multiple bipartite graphs has been receiving attention because of its popular applications such as recommendations. How to make efficient relation inference with few observed links is the main problem on multiple bipartite graphs. Most existing approaches attempt to solve the sparsity problem via learning shared representations to integrate knowledge from multi-source data for shared entities. However, they merely model the correlations from one aspect (e.g. distribution, representation), and cannot impose sufficient constraints on different relations of the shared entities. One effective way of modeling the multi-domain data is to learn the joint distribution of the shared entities across domains.In this paper, we propose Collaborative Adversarial Learning (CAL) that explicitly models the joint distribution of the shared entities across multiple bipartite graphs. The objective of CAL is formulated from a variational lower bound that maximizes the joint log-likelihoods of the observations. In particular, CAL consists of distribution-level and feature-level alignments for knowledge from multiple bipartite graphs. The two-level alignment acts as two different constraints on different relations of the shared entities and facilitates better knowledge transfer for relational learning on multiple bipartite graphs. Extensive experiments on two real-world datasets have shown that the proposed model outperforms the existing methods.

Chenyang Li, Xu Chen, Ya Zhang et al.

Knowledge graphs (KGs) composed of users, objects, and tags are widely used in web applications ranging from E-commerce, social media sites to news portals. This paper concentrates on an attractive application which aims to predict the object-tag links in the KG for better tag recommendation and object explanation. When predicting the object-tag links, both the first-order and high-order proximities between entities in the KG propagate essential similarity information for better prediction. Most existing methods focus on preserving the first-order proximity between entities in the KG. However, they cannot capture the high-order proximities in an explicit way, and the adopted margin-based criterion cannot measure the first-order proximity on the global structure accurately. In this paper, we propose a novel approach named Dual Graph Embedding (DGE) that models both the first-order and high-order proximities in the KG via an auto-encoding architecture to facilitate better object-tag relation inference. Here the dual graphs contain an object graph and a tag graph that explicitly depict the high-order object-object and tag-tag proximities in the KG. The dual graph encoder in DGE then encodes these high-order proximities in the dual graphs into entity embeddings. The decoder formulates a skip-gram objective that maximizes the first-order proximity between observed object-tag pairs over the global proximity structure. With the supervision of the decoder, the embeddings derived by the encoder will be refined to capture both the first-order and high-order proximities in the KG for better link prediction. Extensive experiments on three real-world datasets demonstrate that DGE outperforms the state-of-the-art methods.