arXiv subjects

Weidong Cai

Publications and source records attributed to Weidong Cai.

At least 73 records · Page 4Linked to original sources

arXiv2025

Breaking the Modality Barrier: Universal Embedding Learning with Multimodal LLMs

The Contrastive Language-Image Pre-training (CLIP) framework has become a widely used approach for multimodal representation learning, particularly in image-text retrieval and clustering. However, its efficacy is constrained by three key limitations: (1) text token truncation, (2) isolated image-text encoding, and (3) deficient compositionality due to bag-of-words behavior. While recent Multimodal Large Language Models (MLLMs) have demonstrated significant advances in generalized vision-language understanding, their potential for learning transferable multimodal representations remains underexplored.In this work, we present UniME (Universal Multimodal Embedding), a novel two-stage framework that leverages MLLMs to learn discriminative representations for diverse downstream tasks. In the first stage, we perform textual discriminative knowledge distillation from a powerful LLM-based teacher model to enhance the embedding capability of the MLLMś language component. In the second stage, we introduce hard negative enhanced instruction tuning to further advance discriminative representation learning. Specifically, we initially mitigate false negative contamination and then sample multiple hard negatives per instance within each batch, forcing the model to focus on challenging samples. This approach not only improves discriminative power but also enhances instruction-following ability in downstream tasks. We conduct extensive experiments on the MMEB benchmark and multiple retrieval tasks, including short and long caption retrieval and compositional retrieval. Results demonstrate that UniME achieves consistent performance improvement across all tasks, exhibiting superior discriminative and compositional capabilities.

cs.CV↗

arXiv2025

UniME-V2: MLLM-as-a-Judge for Universal Multimodal Embedding Learning

Universal multimodal embedding models are foundational to various tasks. Existing approaches typically employ in-batch negative mining by measuring the similarity of query-candidate pairs. However, these methods often struggle to capture subtle semantic differences among candidates and lack diversity in negative samples. Moreover, the embeddings exhibit limited discriminative ability in distinguishing false and hard negatives. In this paper, we leverage the advanced understanding capabilities of MLLMs to enhance representation learning and present a novel Universal Multimodal Embedding (UniME-V2) model. Our approach first constructs a potential hard negative set through global retrieval. We then introduce the MLLM-as-a-Judge mechanism, which utilizes MLLMs to assess the semantic alignment of query-candidate pairs and generate soft semantic matching scores. These scores serve as a foundation for hard negative mining, mitigating the impact of false negatives and enabling the identification of diverse, high-quality hard negatives. Furthermore, the semantic matching scores are used as soft labels to mitigate the rigid one-to-one mapping constraint. By aligning the similarity matrix with the soft semantic matching score matrix, the model learns semantic distinctions among candidates, significantly enhancing its discriminative capacity. To further improve performance, we propose UniME-V2-Reranker, a reranking model trained on our mined hard negatives through a joint pairwise and listwise optimization approach. We conduct comprehensive experiments on the MMEB benchmark and multiple retrieval tasks, demonstrating that our method achieves state-of-the-art performance on average across all tasks.

cs.CV↗

arXiv2025

Gene-DML: Dual-Pathway Multi-Level Discrimination for Gene Expression Prediction from Histopathology Images

Accurately predicting gene expression from histopathology images offers a scalable and non-invasive approach to molecular profiling, with significant implications for precision medicine and computational pathology. However, existing methods often underutilize the cross-modal representation alignment between histopathology images and gene expression profiles across multiple representational levels, thereby limiting their prediction performance. To address this, we propose Gene-DML, a unified framework that structures latent space through Dual-pathway Multi-Level discrimination to enhance correspondence between morphological and transcriptional modalities. The multi-scale instance-level discrimination pathway aligns hierarchical histopathology representations extracted at local, neighbor, and global levels with gene expression profiles, capturing scale-aware morphological-transcriptional relationships. In parallel, the cross-level instance-group discrimination pathway enforces structural consistency between individual (image/gene) instances and modality-crossed (gene/image, respectively) groups, strengthening the alignment across modalities. By jointly modeling fine-grained and structural-level discrimination, Gene-DML is able to learn robust cross-modal representations, enhancing both predictive accuracy and generalization across diverse biological contexts. Extensive experiments on public spatial transcriptomics datasets demonstrate that Gene-DML achieves state-of-the-art performance in gene expression prediction. The code and processed datasets are available at https://github.com/YXSong000/Gene-DML.

cs.CV↗

arXiv2025

ART-ASyn: Anatomy-aware Realistic Texture-based Anomaly Synthesis Framework for Chest X-Rays

Unsupervised anomaly detection aims to identify anomalies without pixel-level annotations. Synthetic anomaly-based methods exhibit a unique capacity to introduce controllable irregularities with known masks, enabling explicit supervision during training. However, existing methods often produce synthetic anomalies that are visually distinct from real pathological patterns and ignore anatomical structure. This paper presents a novel Anatomy-aware Realistic Texture-based Anomaly Synthesis framework (ART-ASyn) for chest X-rays that generates realistic and anatomically consistent lung opacity related anomalies using texture-based augmentation guided by our proposed Progressive Binary Thresholding Segmentation method (PBTSeg) for lung segmentation. The generated paired samples of synthetic anomalies and their corresponding precise pixel-level anomaly mask for each normal sample enable explicit segmentation supervision. In contrast to prior work limited to one-class classification, ART-ASyn is further evaluated for zero-shot anomaly segmentation, demonstrating generalizability on an unseen dataset without target-domain annotations. Code availability is available at https://github.com/angelacao-hub/ART-ASyn.

cs.CV↗

arXiv2025

HiFusion: Hierarchical Intra-Spot Alignment and Regional Context Fusion for Spatial Gene Expression Prediction from Histopathology

Spatial transcriptomics (ST) bridges gene expression and tissue morphology but faces clinical adoption barriers due to technical complexity and prohibitive costs. While computational methods predict gene expression from H&E-stained whole-slide images (WSIs), existing approaches often fail to capture the intricate biological heterogeneity within spots and are susceptible to morphological noise when integrating contextual information from surrounding tissue. To overcome these limitations, we propose HiFusion, a novel deep learning framework that integrates two complementary components. First, we introduce the Hierarchical Intra-Spot Modeling module that extracts fine-grained morphological representations through multi-resolution sub-patch decomposition, guided by a feature alignment loss to ensure semantic consistency across scales. Concurrently, we present the Context-aware Cross-scale Fusion module, which employs cross-attention to selectively incorporate biologically relevant regional context, thereby enhancing representational capacity. This architecture enables comprehensive modeling of both cellular-level features and tissue microenvironmental cues, which are essential for accurate gene expression prediction. Extensive experiments on two benchmark ST datasets demonstrate that HiFusion achieves state-of-the-art performance across both 2D slide-wise cross-validation and more challenging 3D sample-specific scenarios. These results underscore HiFusion's potential as a robust, accurate, and scalable solution for ST inference from routine histopathology.

cs.CV↗

arXiv2025

Multi-source-free Domain Adaptation via Uncertainty-aware Adaptive Distillation

Source-free domain adaptation (SFDA) alleviates the domain discrepancy among data obtained from domains without accessing the data for the awareness of data privacy. However, existing conventional SFDA methods face inherent limitations in medical contexts, where medical data are typically collected from multiple institutions using various equipment. To address this problem, we propose a simple yet effective method, named Uncertainty-aware Adaptive Distillation (UAD) for the multi-source-free unsupervised domain adaptation (MSFDA) setting. UAD aims to perform well-calibrated knowledge distillation from (i) model level to deliver coordinated and reliable base model initialisation and (ii) instance level via model adaptation guided by high-quality pseudo-labels, thereby obtaining a high-performance target domain model. To verify its general applicability, we evaluate UAD on two image-based diagnosis benchmarks among two multi-centre datasets, where our method shows a significant performance gain compared with existing works. The code is available at https://github.com/YXSong000/UAD.

cs.CV↗

arXiv2025

MIRROR: Multi-Modal Pathological Self-Supervised Representation Learning via Modality Alignment and Retention

Histopathology and transcriptomics are fundamental modalities in oncology, encapsulating the morphological and molecular aspects of the disease. Multi-modal self-supervised learning has demonstrated remarkable potential in learning pathological representations by integrating diverse data sources. Conventional multi-modal integration methods primarily emphasize modality alignment, while paying insufficient attention to retaining the modality-specific structures. However, unlike conventional scenarios where multi-modal inputs share highly overlapping features, histopathology and transcriptomics exhibit pronounced heterogeneity, offering orthogonal yet complementary insights. Histopathology provides morphological and spatial context, elucidating tissue architecture and cellular topology, whereas transcriptomics delineates molecular signatures through gene expression patterns. This inherent disparity introduces a major challenge in aligning them while maintaining modality-specific fidelity. To address these challenges, we present MIRROR, a novel multi-modal representation learning method designed to foster both modality alignment and retention. MIRROR employs dedicated encoders to extract comprehensive features for each modality, which is further complemented by a modality alignment module to achieve seamless integration between phenotype patterns and molecular profiles. Furthermore, a modality retention module safeguards unique attributes from each modality, while a style clustering module mitigates redundancy and enhances disease-relevant information by modeling and aligning consistent pathological signatures within a clustering space. Extensive evaluations on TCGA cohorts for cancer subtyping and survival analysis highlight MIRROR's superior performance, demonstrating its effectiveness in constructing comprehensive oncological feature representations and benefiting the cancer diagnosis.

cs.CV↗

arXiv2025

NVS-SQA: Exploring Self-Supervised Quality Representation Learning for Neurally Synthesized Scenes without References

Neural View Synthesis (NVS), such as NeRF and 3D Gaussian Splatting, effectively creates photorealistic scenes from sparse viewpoints, typically evaluated by quality assessment methods like PSNR, SSIM, and LPIPS. However, these full-reference methods, which compare synthesized views to reference views, may not fully capture the perceptual quality of neurally synthesized scenes (NSS), particularly due to the limited availability of dense reference views. Furthermore, the challenges in acquiring human perceptual labels hinder the creation of extensive labeled datasets, risking model overfitting and reduced generalizability. To address these issues, we propose NVS-SQA, a NSS quality assessment method to learn no-reference quality representations through self-supervision without reliance on human labels. Traditional self-supervised learning predominantly relies on the "same instance, similar representation" assumption and extensive datasets. However, given that these conditions do not apply in NSS quality assessment, we employ heuristic cues and quality scores as learning objectives, along with a specialized contrastive pair preparation process to improve the effectiveness and efficiency of learning. The results show that NVS-SQA outperforms 17 no-reference methods by a large margin (i.e., on average 109.5% in SRCC, 98.6% in PLCC, and 91.5% in KRCC over the second best) and even exceeds 16 full-reference methods across all evaluation metrics (i.e., 22.9% in SRCC, 19.1% in PLCC, and 18.6% in KRCC over the second best).

cs.CV↗

arXiv2025

A Multimodal Deep Learning Approach for White Matter Shape Prediction in Diffusion MRI Tractography

Shape measures have emerged as promising descriptors of white matter tractography, offering complementary insights into anatomical variability and associations with cognitive and clinical phenotypes. However, conventional methods for computing shape measures are computationally expensive and time-consuming for large-scale datasets due to reliance on voxel-based representations. We propose Tract2Shape, a novel multimodal deep learning framework that leverages geometric (point cloud) and scalar (tabular) features to predict ten white matter tractography shape measures. To enhance model efficiency, we utilize a dimensionality reduction algorithm for the model to predict five primary shape components. The model is trained and evaluated on two independently acquired datasets, the HCP-YA dataset, and the PPMI dataset. We evaluate the performance of Tract2Shape by training and testing it on the HCP-YA dataset and comparing the results with state-of-the-art models. To further assess its robustness and generalization ability, we also test Tract2Shape on the unseen PPMI dataset. Tract2Shape outperforms SOTA deep learning models across all ten shape measures, achieving the highest average Pearson's r and the lowest nMSE on the HCP-YA dataset. The ablation study shows that both multimodal input and PCA contribute to performance gains. On the unseen testing PPMI dataset, Tract2Shape maintains a high Pearson's r and low nMSE, demonstrating strong generalizability in cross-dataset evaluation. Tract2Shape enables fast, accurate, and generalizable prediction of white matter shape measures from tractography data, supporting scalable analysis across datasets. This framework lays a promising foundation for future large-scale white matter shape analysis.

eess.IV↗

arXiv2025

Beyond Random Masking: A Dual-Stream Approach for Rotation-Invariant Point Cloud Masked Autoencoders

Existing rotation-invariant point cloud masked autoencoders (MAE) rely on random masking strategies that overlook geometric structure and semantic coherence. Random masking treats patches independently, failing to capture spatial relationships consistent across orientations and overlooking semantic object parts that maintain identity regardless of rotation. We propose a dual-stream masking approach combining 3D Spatial Grid Masking and Progressive Semantic Masking to address these fundamental limitations. Grid masking creates structured patterns through coordinate sorting to capture geometric relationships that persist across different orientations, while semantic masking uses attention-driven clustering to discover semantically meaningful parts and maintain their coherence during masking. These complementary streams are orchestrated via curriculum learning with dynamic weighting, progressing from geometric understanding to semantic discovery. Designed as plug-and-play components, our strategies integrate into existing rotation-invariant frameworks without architectural changes, ensuring broad compatibility across different approaches. Comprehensive experiments on ModelNet40, ScanObjectNN, and OmniObject3D demonstrate consistent improvements across various rotation scenarios, showing substantial performance gains over the baseline rotation-invariant methods.

cs.CV↗

arXiv2025

TractGraphFormer: Anatomically Informed Hybrid Graph CNN-Transformer Network for Interpretable Sex and Age Prediction from Diffusion MRI Tractography

The relationship between brain connections and non-imaging phenotypes is increasingly studied using deep neural networks. However, the local and global properties of brain white matter networks are often overlooked in convolutional network design. We introduce TractGraphFormer, a hybrid Graph CNN-Transformer deep learning framework tailored for diffusion MRI tractography. This model leverages local anatomical characteristics and global feature dependencies of white matter structures. The Graph CNN module captures white matter geometry and grey matter connectivity to aggregate local features from anatomically similar white matter connections, while the Transformer module uses self-attention to enhance global information learning. Additionally, TractGraphFormer includes an attention module for interpreting predictive white matter connections. We apply TractGraphFormer to tasks of sex and age prediction. TractGraphFormer shows strong performance in large datasets of children (n=9345) and young adults (n=1065). Overall, our approach suggests that widespread connections in the WM are predictive of the sex and age of an individual. For each prediction task, consistent predictive anatomical tracts are identified across the two datasets. The proposed approach highlights the potential of integrating local anatomical information and global feature dependencies to improve prediction performance in machine learning with diffusion MRI tractography.

cs.CV↗

arXiv2025

From Promise to Practical Reality: Transforming Diffusion MRI Analysis with Fast Deep Learning Enhancement

Fiber orientation distribution (FOD) is an advanced diffusion MRI modeling technique that represents complex white matter fiber configurations, and a key step for subsequent brain tractography and connectome analysis. Its reliability and accuracy, however, heavily rely on the quality of the MRI acquisition and the subsequent estimation of the FODs at each voxel. Generating reliable FODs from widely available clinical protocols with single-shell and low-angular-resolution acquisitions remains challenging but could potentially be addressed with recent advances in deep learning-based enhancement techniques. Despite advancements, existing methods have predominantly been assessed on healthy subjects, which have proved to be a major hurdle for their clinical adoption. In this work, we validate a newly optimized enhancement framework, FastFOD-Net, across healthy controls and six neurological disorders. This accelerated end-to-end deep learning framework enhancing FODs with superior performance and delivering training/inference efficiency for clinical use ($60\times$ faster comparing to its predecessor). With the most comprehensive clinical evaluation to date, our work demonstrates the potential of FastFOD-Net in accelerating clinical neuroscience research, empowering diffusion MRI analysis for disease differentiation, improving interpretability in connectome applications, and reducing measurement errors to lower sample size requirements. Critically, this work will facilitate the more widespread adoption of, and build clinical trust in, deep learning based methods for diffusion MRI enhancement. Specifically, FastFOD-Net enables robust analysis of real-world, clinical diffusion MRI data, comparable to that achievable with high-quality research acquisitions.

cs.CV↗

arXiv2025

Gotta Hear Them All: Towards Sound Source Aware Audio Generation

Audio synthesis has broad applications in multimedia. Recent advancements have made it possible to generate relevant audios from inputs describing an audio scene, such as images or texts. However, the immersiveness and expressiveness of the generation are limited. One possible problem is that existing methods solely rely on the global scene and overlook details of local sounding objects (i.e., sound sources). To address this issue, we propose a Sound Source-Aware Audio (SS2A) generator. SS2A is able to locally perceive multimodal sound sources from a scene with visual detection and cross-modality translation. It then contrastively learns a Cross-Modal Sound Source (CMSS) Manifold to semantically disambiguate each source. Finally, we attentively mix their CMSS semantics into a rich audio representation, from which a pretrained audio generator outputs the sound. To model the CMSS manifold, we curate a novel single-sound-source visual-audio dataset VGGS3 from VGGSound. We also design a Sound Source Matching Score to clearly measure localized audio relevance. With the effectiveness of explicit sound source modeling, SS2A achieves state-of-the-art performance in extensive image-to-audio tasks. We also qualitatively demonstrate SS2A's ability to achieve intuitive synthesis control by compositing vision, text, and audio conditions. Furthermore, we show that our sound source modeling can achieve competitive video-to-audio performance with a straightforward temporal aggregation mechanism.

cs.MM↗

arXiv2025

RealSyn: An Effective and Scalable Multimodal Interleaved Document Transformation Paradigm

After pre-training on extensive image-text pairs, Contrastive Language-Image Pre-training (CLIP) demonstrates promising performance on a wide variety of benchmarks. However, a substantial volume of multimodal interleaved documents remains underutilized for contrastive vision-language representation learning. To fully leverage these unpaired documents, we initially establish a Real-World Data Extraction pipeline to extract high-quality images and texts. Then we design a hierarchical retrieval method to efficiently associate each image with multiple semantically relevant realistic texts. To further enhance fine-grained visual information, we propose an image semantic augmented generation module for synthetic text production. Furthermore, we employ a semantic balance sampling strategy to improve dataset diversity, enabling better learning of long-tail concepts. Based on these innovations, we construct RealSyn, a dataset combining realistic and synthetic texts, available in three scales: 15M, 30M, and 100M. We compare our dataset with other widely used datasets of equivalent scale for CLIP training. Models pre-trained on RealSyn consistently achieve state-of-the-art performance across various downstream tasks, including linear probe, zero-shot transfer, zero-shot robustness, and zero-shot retrieval. Furthermore, extensive experiments confirm that RealSyn significantly enhances contrastive vision-language representation learning and demonstrates robust scalability. To facilitate future research, the RealSyn dataset and pretrained model weights are released at https://github.com/deepglint/RealSyn.

cs.CV↗

arXiv2025

CA-W3D: Leveraging Context-Aware Knowledge for Weakly Supervised Monocular 3D Detection

Weakly supervised monocular 3D detection, while less annotation-intensive, often struggles to capture the global context required for reliable 3D reasoning. Conventional label-efficient methods focus on object-centric features, neglecting contextual semantic relationships that are critical in complex scenes. In this work, we propose a Context-Aware Weak Supervision for Monocular 3D object detection, namely CA-W3D, to address this limitation in a two-stage training paradigm. Specifically, we first introduce a pre-training stage employing Region-wise Object Contrastive Matching (ROCM), which aligns regional object embeddings derived from a trainable monocular 3D encoder and a frozen open-vocabulary 2D visual grounding model. This alignment encourages the monocular encoder to discriminate scene-specific attributes and acquire richer contextual knowledge. In the second stage, we incorporate a pseudo-label training process with a Dual-to-One Distillation (D2OD) mechanism, which effectively transfers contextual priors into the monocular encoder while preserving spatial fidelity and maintaining computational efficiency during inference. Extensive experiments conducted on the public KITTI benchmark demonstrate the effectiveness of our approach, surpassing the SoTA method over all metrics, highlighting the importance of contextual-aware knowledge in weakly-supervised monocular 3D detection.

cs.CV↗

arXiv2025

VRM: Knowledge Distillation via Virtual Relation Matching

Knowledge distillation (KD) aims to transfer the knowledge of a more capable yet cumbersome teacher model to a lightweight student model. In recent years, relation-based KD methods have fallen behind, as their instance-matching counterparts dominate in performance. In this paper, we revive relational KD by identifying and tackling several key issues in relation-based methods, including their susceptibility to overfitting and spurious responses. Specifically, we transfer novelly constructed affinity graphs that compactly encapsulate a wealth of beneficial inter-sample, inter-class, and inter-view correlations by exploiting virtual views and relations as a new kind of knowledge. As a result, the student has access to richer guidance signals and stronger regularisation throughout the distillation process. To further mitigate the adverse impact of spurious responses, we prune the affinity graphs by dynamically detaching redundant and unreliable edges. Extensive experiments on CIFAR-100, ImageNet, and MS-COCO datasets demonstrate the superior performance of the proposed virtual relation matching (VRM) method, where it consistently sets new state-of-the-art records over a range of models, architectures, tasks, and set-ups. For instance, VRM for the first time hits 74.0% accuracy for ResNet50-to-MobileNetV2 distillation on ImageNet, and improves DeiT-T by 14.44% on CIFAR-100 with a ResNet56 teacher.

cs.CV↗

arXiv2025

ChoreoMuse: Robust Music-to-Dance Video Generation with Style Transfer and Beat-Adherent Motion

Modern artistic productions increasingly demand automated choreography generation that adapts to diverse musical styles and individual dancer characteristics. Existing approaches often fail to produce high-quality dance videos that harmonize with both musical rhythm and user-defined choreography styles, limiting their applicability in real-world creative contexts. To address this gap, we introduce ChoreoMuse, a diffusion-based framework that uses SMPL format parameters and their variation version as intermediaries between music and video generation, thereby overcoming the usual constraints imposed by video resolution. Critically, ChoreoMuse supports style-controllable, high-fidelity dance video generation across diverse musical genres and individual dancer characteristics, including the flexibility to handle any reference individual at any resolution. Our method employs a novel music encoder MotionTune to capture motion cues from audio, ensuring that the generated choreography closely follows the beat and expressive qualities of the input music. To quantitatively evaluate how well the generated dances match both musical and choreographic styles, we introduce two new metrics that measure alignment with the intended stylistic cues. Extensive experiments confirm that ChoreoMuse achieves state-of-the-art performance across multiple dimensions, including video quality, beat alignment, dance diversity, and style adherence, demonstrating its potential as a robust solution for a wide range of creative applications. Video results can be found on our project page: https://choreomuse.github.io.

cs.GR↗

arXiv2025

ScSAM: Debiasing Morphology and Distributional Variability in Subcellular Semantic Segmentation

The significant morphological and distributional variability among subcellular components poses a long-standing challenge for learning-based organelle segmentation models, significantly increasing the risk of biased feature learning. Existing methods often rely on single mapping relationships, overlooking feature diversity and thereby inducing biased training. Although the Segment Anything Model (SAM) provides rich feature representations, its application to subcellular scenarios is hindered by two key challenges: (1) The variability in subcellular morphology and distribution creates gaps in the label space, leading the model to learn spurious or biased features. (2) SAM focuses on global contextual understanding and often ignores fine-grained spatial details, making it challenging to capture subtle structural alterations and cope with skewed data distributions. To address these challenges, we introduce ScSAM, a method that enhances feature robustness by fusing pre-trained SAM with Masked Autoencoder (MAE)-guided cellular prior knowledge to alleviate training bias from data imbalance. Specifically, we design a feature alignment and fusion module to align pre-trained embeddings to the same feature space and efficiently combine different representations. Moreover, we present a cosine similarity matrix-based class prompt encoder to activate class-specific features to recognize subcellular categories. Extensive experiments on diverse subcellular image datasets demonstrate that ScSAM outperforms state-of-the-art methods.

cs.CV↗