SearcharxivSearch

arXiv subjects

Xinru Zhang

Publications and source records attributed to Xinru Zhang.

13 recordsLinked to original sources

Notes on gamma invariants of finite dimensional Hopf algebras

Let $H$ be a finite-dimensional, non-semisimple Hopf algebra over an algebraically closed field $\mathbf{k}$. This paper investigates the asymptotic behavior of the core of left $H$-modules through the lens of the gamma invariant $γ_{\mathfrak{X}}$ relative to a representation ideal $I_{\mathfrak{X}}$. We establish an equivalent characterization for the quotient of the Green ring $R_{\mathfrak{X}}$ to be a transitive fusion ring, demonstrating that transitivity is synonymous with the non-degeneracy of a naturally induced bilinear form and the collapse of the ideals $P_{+}$, $P_{-}$ and $I_{\operatorname{max}}$ into a single ideal. Furthermore, we prove that the Green ring exhibits the structure of a representation ring in the sense of Benson, provided that the square of the antipode is an inner automorphism and the equality $I_{\operatorname{max}}=I_{\operatorname{proj}}$ holds. As an explicit application of these frameworks, we analyze the Drinfeld double $D(H_4)$ of the Sweedler algebra, identifying an infinite family of distinct representation ideals and proving that the maximal gamma invariant $γ_{\operatorname{max}}$ induces a genuine ring homomorphism. Finally, for Hopf algebras of finite representation type under the assumption $P_{+} = P_{-} = I_{\operatorname{max}}$, we show that $γ_{\operatorname{max}}$ coincides precisely with the Frobenius--Perron dimension, and we explicitly compute the gamma invariants for the standard basis elements of the Green ring of the Taft algebra $H_n(q)$.

math.QA

Native-Domain Cross-Attention for Camera-LiDAR Extrinsic Calibration Under Large Initial Perturbations

Accurate camera-LiDAR fusion relies on precise extrinsic calibration, which fundamentally depends on establishing reliable cross-modal correspondences under potentially large misalignments. Existing learning-based methods typically project LiDAR points into depth maps for feature fusion, which distorts 3D geometry and degrades performance when the extrinsic initialization is far from the ground truth. To address this issue, we propose an extrinsic-aware cross-attention framework that directly aligns image patches and LiDAR point groups in their native domains. The proposed attention mechanism explicitly injects extrinsic parameter hypotheses into the correspondence modeling process, enabling geometry-consistent cross-modal interaction without relying on projected 2D depth maps. Extensive experiments on the KITTI and nuScenes benchmarks demonstrate that our method consistently outperforms state-of-the-art approaches in both accuracy and robustness. Under large extrinsic perturbations, our approach achieves accurate calibration in 88% of KITTI cases and 99% of nuScenes cases, substantially surpassing the second-best baseline. We have open sourced our code on https://github.com/gitouni/ProjFusion to benefit the community.

cs.CV

A Knowledge Noise Mitigation Framework for Knowledge-based Visual Question Answering

Knowledge-based visual question answering (KB-VQA) requires a model to understand images and utilize external knowledge to provide accurate answers. Existing approaches often directly augment models with retrieved information from knowledge sources while ignoring substantial knowledge redundancy, which introduces noise into the answering process. To address this, we propose a training-free framework with knowledge focusing for KB-VQA, that mitigates the impact of noise by enhancing knowledge relevance and reducing redundancy. First, for knowledge retrieval, our framework concludes essential parts from the image-question pairs, creating low-noise queries that enhance the retrieval of highly relevant knowledge. Considering that redundancy still persists in the retrieved knowledge, we then prompt large models to identify and extract answer-beneficial segments from knowledge. In addition, we introduce a selective knowledge integration strategy, allowing the model to incorporate knowledge only when it lacks confidence in answering the question, thereby mitigating the influence of redundant information. Our framework enables the acquisition of accurate and critical knowledge, and extensive experiments demonstrate that it outperforms state-of-the-art methods.

cs.CV

Iterative Camera-LiDAR Extrinsic Optimization via Surrogate Diffusion

Cameras and LiDAR are essential sensors for autonomous vehicles. Camera-LiDAR data fusion compensate for deficiencies of stand-alone sensors but relies on precise extrinsic calibration. Many learning-based calibration methods predict extrinsic parameters in a single step. Driven by the growing demand for higher accuracy, a few approaches utilize multi-range models or integrate multiple methods to improve extrinsic parameter predictions, but these strategies incur extended training times and require additional storage for separate models. To address these issues, we propose a single-model iterative approach based on surrogate diffusion to significantly enhance the capacity of individual calibration methods. By applying a buffering technique proposed by us, the inference time of our surrogate diffusion is 43.7% less than that of multi-range models. Additionally, we create a calibration network as our denoiser, featuring both projection-first and encoding-first branches for effective point feature extraction. Extensive experiments demonstrate that our diffusion model outperforms other single-model iterative methods and delivers competitive results compared to multi-range models. Our denoiser exceeds state-of-the-art calibration methods, reducing the rotation error by 24.5% compared to the second-best method. Furthermore, with the proposed diffusion applied, it achieves 20.4% less rotation error and 9.6% less translation error.

cs.CV

Iterative Camera-LiDAR Extrinsic Optimization via Surrogate Diffusion

Cameras and LiDAR are essential sensors for autonomous vehicles. The fusion of camera and LiDAR data addresses the limitations of individual sensors but relies on precise extrinsic calibration. Recently, numerous end-to-end calibration methods have been proposed; however, most predict extrinsic parameters in a single step and lack iterative optimization capabilities. To address the increasing demand for higher accuracy, we propose a versatile iterative framework based on surrogate diffusion. This framework can enhance the performance of any calibration method without requiring architectural modifications. Specifically, the initial extrinsic parameters undergo iterative refinement through a denoising process, in which the original calibration method serves as a surrogate denoiser to estimate the final extrinsics at each step. For comparative analysis, we selected four state-of-the-art calibration methods as surrogate denoisers and compared the results of our diffusion process with those of two other iterative approaches. Extensive experiments demonstrate that when integrated with our diffusion model, all calibration methods achieve higher accuracy, improved robustness, and greater stability compared to other iterative techniques and their single-step counterparts.

cs.CV

Controllability concepts for mean-field dynamics with reduced-rank coefficients

In this paper we explore several novel notions of exact controllability for mean-field linear controlled stochastic differential equations (SDEs). A key feature of our study is that the noise coefficient is not required to be of full rank. We begin by demonstrating that classical exact controllability with $\mathbb{L}^2$-controls necessarily requires both rank conditions on the noise introduced in [8] and subsequent works. When these rank conditions are not satisfied, we introduce alternative rank requirements on the drift, which enable exact controllability by relaxing the regularity of the controls. In cases where both the aforementioned rank conditions fail, we propose and characterize a new notion of exact terminal controllability to normal laws (ETCNL). Additionally, we investigate a new class of Wasserstein-set-valued backward SDEs that arise naturally associated to ETCNL.

math.OC

A Stochastic Porous Media Schr{ö}dinger Equation: Feynman-type Motivation, Well-Posedness and Control Interpretation

This paper's aim is threefold. First, using Feynman's path approach to the derivation of theclassical Schr{ö}dinger's equation in [6] and by introducing a slight path (or wave) dependency ofthe action, we derive a new class of equations of Schr{ö}dinger type where the driving operatoris no longer the Laplace one but rather of complex porous media-type. Second, using suitableconcepts of monotonicity in the complex setting and on appropriate functional spaces, we showthe existence and uniqueness of the solution to this type of equation. In the formulation of ourequation, we adjoin possible measurement absolute errors translating in an additive Brownianperturbation and interactions between different waves translating in a mean-field (or McKean-Vlasov) dependency of drift coefficient. Finally, using Fitzpatrick's characterization of maximalmonotone operators (cf. [7]), we propose a Br{é}zis-Ekeland type characterization of the solutionof the deterministic equation via a control problem. This is envisaged as a possible way toovercome strict monotonicity requirements in the complex setting.

math.AP

SegHeD: Segmentation of Heterogeneous Data for Multiple Sclerosis Lesions with Anatomical Constraints

Assessment of lesions and their longitudinal progression from brain magnetic resonance (MR) images plays a crucial role in diagnosing and monitoring multiple sclerosis (MS). Machine learning models have demonstrated a great potential for automated MS lesion segmentation. Training such models typically requires large-scale high-quality datasets that are consistently annotated. However, MS imaging datasets are often small, segregated across multiple sites, with different formats (cross-sectional or longitudinal), and diverse annotation styles. This poses a significant challenge to train a unified MS lesion segmentation model. To tackle this challenge, we present SegHeD, a novel multi-dataset multi-task segmentation model that can incorporate heterogeneous data as input and perform all-lesion, new-lesion, as well as vanishing-lesion segmentation. Furthermore, we account for domain knowledge about MS lesions, incorporating longitudinal, spatial, and volumetric constraints into the segmentation model. SegHeD is assessed on five MS datasets and achieves a high performance in all, new, and vanishing-lesion segmentation, outperforming several state-of-the-art methods in this field.

eess.IV

A Foundation Model for Brain Lesion Segmentation with Mixture of Modality Experts

Brain lesion segmentation plays an essential role in neurological research and diagnosis. As brain lesions can be caused by various pathological alterations, different types of brain lesions tend to manifest with different characteristics on different imaging modalities. Due to this complexity, brain lesion segmentation methods are often developed in a task-specific manner. A specific segmentation model is developed for a particular lesion type and imaging modality. However, the use of task-specific models requires predetermination of the lesion type and imaging modality, which complicates their deployment in real-world scenarios. In this work, we propose a universal foundation model for 3D brain lesion segmentation, which can automatically segment different types of brain lesions for input data of various imaging modalities. We formulate a novel Mixture of Modality Experts (MoME) framework with multiple expert networks attending to different imaging modalities. A hierarchical gating network combines the expert predictions and fosters expertise collaboration. Furthermore, we introduce a curriculum learning strategy during training to avoid the degeneration of each expert network and preserve their specialization. We evaluated the proposed method on nine brain lesion datasets, encompassing five imaging modalities and eight lesion types. The results show that our model outperforms state-of-the-art universal models and provides promising generalization to unseen datasets.

eess.IV

Advancing Brain Tumor Inpainting with Generative Models

Synthesizing healthy brain scans from diseased brain scans offers a potential solution to address the limitations of general-purpose algorithms, such as tissue segmentation and brain extraction algorithms, which may not effectively handle diseased images. We consider this a 3D inpainting task and investigate the adaptation of 2D inpainting methods to meet the requirements of 3D magnetic resonance imaging(MRI) data. Our contributions encompass potential modifications tailored to MRI-specific needs, and we conducted evaluations of multiple inpainting techniques using the BraTS2023 Inpainting datasets to assess their efficacy and limitations.

eess.IV

Unsupervised Brain Tumor Segmentation with Image-based Prompts

Automated brain tumor segmentation based on deep learning (DL) has achieved promising performance. However, it generally relies on annotated images for model training, which is not always feasible in clinical settings. Therefore, the development of unsupervised DL-based brain tumor segmentation approaches without expert annotations is desired. Motivated by the success of prompt learning (PL) in natural language processing, we propose an approach to unsupervised brain tumor segmentation by designing image-based prompts that allow indication of brain tumors, and this approach is dubbed as PL-based Brain Tumor Segmentation (PL-BTS). Specifically, instead of directly training a model for brain tumor segmentation with a large amount of annotated data, we seek to train a model that can answer the question: is a voxel in the input image associated with tumor-like hyper-/hypo-intensity? Such a model can be trained by artificially generating tumor-like hyper-/hypo-intensity on images without tumors with hand-crafted designs. Since the hand-crafted designs may be too simplistic to represent all kinds of real tumors, the trained model may overfit the simplistic hand-crafted task rather than actually answer the question of abnormality. To address this problem, we propose the use of a validation task, where we generate a different hand-crafted task to monitor overfitting. In addition, we propose PL-BTS+ that further improves PL-BTS by exploiting unannotated images with brain tumors. Compared with competing unsupervised methods, the proposed method has achieved marked improvements on both public and in-house datasets, and we have also demonstrated its possible extension to other brain lesion segmentation tasks.

cs.CV

Knowledge Graph Completion Method Combined With Adaptive Enhanced Semantic Information

Translation models tend to ignore the rich semantic information in triads in the process of knowledge graph complementation. To remedy this shortcoming, this paper constructs a knowledge graph complementation method that incorporates adaptively enhanced semantic information. The hidden semantic information inherent in the triad is obtained by fine-tuning the BERT model, and the attention feature embedding method is used to calculate the semantic attention scores between relations and entities in positive and negative triads and incorporate them into the structural information to form a soft constraint rule for semantic information. The rule is added to the original translation model to realize the adaptive enhancement of semantic information. In addition, the method takes into account the effect of high-dimensional vectors on the effect, and uses the BERT-whitening method to reduce the dimensionality and generate a more efficient semantic vector representation. After experimental comparison, the proposed method performs better on both FB15K and WIN18 datasets, with a numerical improvement of about 2.6% compared with the original translation model, which verifies the reasonableness and effectiveness of the method.

cs.CL

CarveMix: A Simple Data Augmentation Method for Brain Lesion Segmentation

Brain lesion segmentation provides a valuable tool for clinical diagnosis, and convolutional neural networks (CNNs) have achieved unprecedented success in the task. Data augmentation is a widely used strategy that improves the training of CNNs, and the design of the augmentation method for brain lesion segmentation is still an open problem. In this work, we propose a simple data augmentation approach, dubbed as CarveMix, for CNN-based brain lesion segmentation. Like other "mix"-based methods, such as Mixup and CutMix, CarveMix stochastically combines two existing labeled images to generate new labeled samples. Yet, unlike these augmentation strategies based on image combination, CarveMix is lesion-aware, where the combination is performed with an attention on the lesions and a proper annotation is created for the generated image. Specifically, from one labeled image we carve a region of interest (ROI) according to the lesion location and geometry, and the size of the ROI is sampled from a probability distribution. The carved ROI then replaces the corresponding voxels in a second labeled image, and the annotation of the second image is replaced accordingly as well. In this way, we generate new labeled images for network training and the lesion information is preserved. To evaluate the proposed method, experiments were performed on two brain lesion datasets. The results show that our method improves the segmentation accuracy compared with other simple data augmentation approaches.

eess.IV