Vision Foundation Models (VFMs) have become a de facto choice for many downstream vision tasks, like image classification, image segmentation, and object localization. However, they can also provide significant utility for downstream 3D tasks that can leverage the cross-modal information (e.g., from paired image data). In our work, we further explore the utility of VFMs for adapting from a labeled source to unlabeled target data for the task of LiDAR-based 3D semantic segmentation. Our method consumes paired 2D-3D (image and point cloud) data and relies on the robust (cross-domain) features from a VFM to train a 3D backbone on a mix of labeled source and unlabeled target data. At the heart of our method lies a fusion network that is guided by both the image and point cloud streams, with their relative contributions adjusted based on the target domain. We extensively compare our proposed methodology with different state-of-the-art methods in several settings and achieve strong performance gains. For example, achieving an average improvement of 6.5 mIoU (over all tasks), when compared with the previous state-of-the-art.
Deep neural networks face several challenges in hyperspectral image classification, including high-dimensional data, sparse distribution of ground objects, and spectral redundancy, which often lead to classification overfitting and limited generalization capability. To more efficiently adapt to ground object distributions while extracting image features without introducing excessive parameters and skipping redundant information, this paper proposes KANet based on an improved 3D-DenseNet model, consisting of 3D KAN Conv and an adaptive grid update mechanism. By introducing learnable univariate B-spline functions on network edges, specifically by flattening three-dimensional neighborhoods into vectors and applying B-spline-parameterized nonlinear activation functions to replace the fixed linear weights of traditional 3D convolutional kernels, we precisely capture complex spectral-spatial nonlinear relationships in hyperspectral data. Simultaneously, through a dynamic grid adjustment mechanism, we adaptively update the grid point positions of B-splines based on the statistical characteristics of input data, optimizing the resolution of spline functions to match the non-uniform distribution of spectral features, significantly improving the model's accuracy in high-dimensional data modeling and parameter efficiency, effectively alleviating the curse of dimensionality. This characteristic demonstrates superior neural scaling laws compared to traditional convolutional neural networks and reduces overfitting risks in small-sample and high-noise scenarios. KANet enhances model representation capability through a 3D dynamic expert convolution system without increasing network depth or width. The proposed method demonstrates superior performance on IN, UP, and KSC datasets, outperforming mainstream hyperspectral image classification approaches.
In-context learning (ICL) enables Large Vision-Language Models (LVLMs) to adapt to new tasks without parameter updates, using a few demonstrations from a large support set. However, selecting informative demonstrations leads to high computational and memory costs. While some methods explore selecting a small and representative coreset in the text classification, evaluating all support set samples remains costly, and discarded samples lead to unnecessary information loss. These methods may also be less effective for image classification due to differences in feature spaces. Given these limitations, we propose Key-based Coreset Optimization (KeCO), a novel framework that leverages untapped data to construct a compact and informative coreset. We introduce visual features as keys within the coreset, which serve as the anchor for identifying samples to be updated through different selection strategies. By leveraging untapped samples from the support set, we update the keys of selected coreset samples, enabling the randomly initialized coreset to evolve into a more informative coreset under low computational cost. Through extensive experiments on coarse-grained and fine-grained image classification benchmarks, we demonstrate that KeCO effectively enhances ICL performance for image classification task, achieving an average improvement of more than 20\%. Notably, we evaluate KeCO under a simulated online scenario, and the strong performance in this scenario highlights the practical value of our framework for resource-constrained real-world scenarios.
Background: Automated classification of thyroid fine needle aspiration biopsy (FNAB) images faces challenges in limited data, inter-observer variability, and computational cost. Efficient, interpretable models are crucial for clinical support. Objective: To develop and externally validate a deep learning system for the multi-class classification of thyroid FNAB images into three key categories that directly guide post-biopsy treatment decisions in Vietnam: benign (B2), suspicious for malignancy (B5), and malignant (B6), while achieving high diagnostic accuracy with low computational overhead. Methods: Our framework features: (1) YOLOv10-based cell cluster detection for informative sub-region extraction and noise reduction; (2) a curriculum learning-inspired protocol sequencing localized crops to full images for multi-scale feature capture; (3) adaptive lightweight EfficientNetB0 (4 millions parameters) selection balancing performance and efficiency; and (4) a Transformer-inspired module for multi-scale, multi-region analysis. External validation used 1,015 independent FNAB images. Results: ThyroidEffi Basic achieved a macro F1 of 89.19\% and AUCs of 0.98 (B2), 0.95 (B5), and 0.96 (B6) on the internal test set. External validation yielded AUCs of 0.9495 (B2), 0.7436 (B5), and 0.8396 (B6). ThyroidEffi Premium improved macro F1 to 89.77\%. Grad-CAM highlighted key diagnostic regions, confirming interpretability. The system processed 1000 cases in 30 seconds, demonstrating feasibility on widely accessible hardware like a 12-core CPU. Conclusions: This work demonstrates that high-accuracy, interpretable thyroid FNAB image classification is achievable with minimal computational demands.
Tree tensor networks (TTNs) offer powerful models for image classification. While these TTN image classifiers already show excellent performance on classical hardware, embedding them into quantum neural networks (QNNs) may further improve the performance by leveraging quantum resources. However, embedding TTN classifiers into QNNs for multiclass classification remains challenging. Key obstacles are the highorder gate operations required for large bond dimensions and the mid-circuit postselection with exponentially low success rates necessary for the exact embedding. In this work, to address these challenges, we propose forest tensor network (FTN)-classifiers, which aggregate multiple small-bond-dimension TTNs. This allows us to handle multiclass classification without requiring large gates in the embedded circuits. We then remove the overhead of mid-circuit postselection by extending the adiabatic encoding framework to our setting and smoothly encode the FTN-classifiers into a quantum forest tensor network (qFTN)- classifiers. Numerical experiments on MNIST and CIFAR-10 demonstrate that we can successfully train FTN-classifiers and encode them into qFTN-classifiers, while maintaining or even improving the performance of the pre-trained FTN-classifiers. These results suggest that synergy between TTN classification models and QNNs can provide a robust and scalable framework for multiclass quantum-enhanced image classification.
Vision Transformers (ViTs) have revolutionized computer vision by leveraging self-attention to model long-range dependencies. However, ViTs face challenges such as high computational costs due to the quadratic scaling of self-attention and the requirement of a large amount of training data. To address these limitations, we propose the Efficient Convolutional Vision Transformer (ECViT), a hybrid architecture that effectively combines the strengths of CNNs and Transformers. ECViT introduces inductive biases such as locality and translation invariance, inherent to Convolutional Neural Networks (CNNs) into the Transformer framework by extracting patches from low-level features and enhancing the encoder with convolutional operations. Additionally, it incorporates local-attention and a pyramid structure to enable efficient multi-scale feature extraction and representation. Experimental results demonstrate that ECViT achieves an optimal balance between performance and efficiency, outperforming state-of-the-art models on various image classification tasks while maintaining low computational and storage requirements. ECViT offers an ideal solution for applications that prioritize high efficiency without compromising performance.
Diffusion models (DMs) have revolutionized text-to-image generation, enabling the creation of highly realistic and customized images from text prompts. With the rise of parameter-efficient fine-tuning (PEFT) techniques like LoRA, users can now customize powerful pre-trained models using minimal computational resources. However, the widespread sharing of fine-tuned DMs on open platforms raises growing ethical and legal concerns, as these models may inadvertently or deliberately generate sensitive or unauthorized content, such as copyrighted material, private individuals, or harmful content. Despite the increasing regulatory attention on generative AI, there are currently no practical tools for systematically auditing these models before deployment. In this paper, we address the problem of concept auditing: determining whether a fine-tuned DM has learned to generate a specific target concept. Existing approaches typically rely on prompt-based input crafting and output-based image classification but suffer from critical limitations, including prompt uncertainty, concept drift, and poor scalability. To overcome these challenges, we introduce Prompt-Agnostic Image-Free Auditing (PAIA), a novel, model-centric concept auditing framework. By treating the DM as the object of inspection, PAIA enables direct analysis of internal model behavior, bypassing the need for optimized prompts or generated images. We evaluate PAIA on 320 controlled model and 690 real-world community models sourced from a public DM sharing platform. PAIA achieves over 90% detection accuracy while reducing auditing time by 18-40x compared to existing baselines. To our knowledge, PAIA is the first scalable and practical solution for pre-deployment concept auditing of diffusion models, providing a practical foundation for safer and more transparent diffusion model sharing.
Knowledge distillation (KD) is a technique for transferring knowledge from complex teacher models to simpler student models, significantly enhancing model efficiency and accuracy. It has demonstrated substantial advancements in various applications including image classification, object detection, language modeling, text classification, and sentiment analysis. Recent innovations in KD methods, such as attention-based approaches, block-wise logit distillation, and decoupling distillation, have notably improved student model performance. These techniques focus on stimulus complexity, attention mechanisms, and global information capture to optimize knowledge transfer. In addition, KD has proven effective in compressing large language models while preserving accuracy, reducing computational overhead, and improving inference speed. This survey synthesizes the latest literature, highlighting key findings, contributions, and future directions in knowledge distillation to provide insights for researchers and practitioners on its evolving role in artificial intelligence and machine learning.
Biological synapses effortlessly balance memory retention and flexibility, yet artificial neural networks still struggle with the extremes of catastrophic forgetting and catastrophic remembering. Here, we introduce Metaplasticity from Synaptic Uncertainty (MESU), a Bayesian framework that updates network parameters according their uncertainty. This approach allows a principled combination of learning and forgetting that ensures that critical knowledge is preserved while unused or outdated information is gradually released. Unlike standard Bayesian approaches -- which risk becoming overly constrained, and popular continual-learning methods that rely on explicit task boundaries, MESU seamlessly adapts to streaming data. It further provides reliable epistemic uncertainty estimates, allowing out-of-distribution detection, the only computational cost being to sample the weights multiple times to provide proper output statistics. Experiments on image-classification benchmarks demonstrate that MESU mitigates catastrophic forgetting, while maintaining plasticity for new tasks. When training 200 sequential permuted MNIST tasks, MESU outperforms established continual learning techniques in terms of accuracy, capability to learn additional tasks, and out-of-distribution data detection. Additionally, due to its non-reliance on task boundaries, MESU outperforms conventional learning techniques on the incremental training of CIFAR-100 tasks consistently in a wide range of scenarios. Our results unify ideas from metaplasticity, Bayesian inference, and Hessian-based regularization, offering a biologically-inspired pathway to robust, perpetual learning.
Although Mamba models significantly improve hyperspectral image (HSI) classification, one critical challenge is the difficulty in building the sequence of Mamba tokens efficiently. This paper presents a Sparse Deformable Mamba (SDMamba) approach for enhanced HSI classification, with the following contributions. First, to enhance Mamba sequence, an efficient Sparse Deformable Sequencing (SDS) approach is designed to adaptively learn the ''optimal" sequence, leading to sparse and deformable Mamba sequence with increased detail preservation and decreased computations. Second, to boost spatial-spectral feature learning, based on SDS, a Sparse Deformable Spatial Mamba Module (SDSpaM) and a Sparse Deformable Spectral Mamba Module (SDSpeM) are designed for tailored modeling of the spatial information spectral information. Last, to improve the fusion of SDSpaM and SDSpeM, an attention based feature fusion approach is designed to integrate the outputs of the SDSpaM and SDSpeM. The proposed method is tested on several benchmark datasets with many state-of-the-art approaches, demonstrating that the proposed approach can achieve higher accuracy with less computation, and better detail small-class preservation capability.
Hyperspectral image (HSI) classification remains a challenging task due to the intricate spatial-spectral correlations. Existing transformer models excel in capturing long-range dependencies but often suffer from information redundancy and attention inefficiencies, limiting their ability to model fine-grained relationships crucial for HSI classification. To overcome these limitations, this work proposes MemFormer, a lightweight and memory-enhanced transformer. MemFormer introduces a memory-enhanced multi-head attention mechanism that iteratively refines a dynamic memory module, enhancing feature extraction while reducing redundancy across layers. Additionally, a dynamic memory enrichment strategy progressively captures complex spatial and spectral dependencies, leading to more expressive feature representations. To further improve structural consistency, we incorporate a spatial-spectral positional encoding (SSPE) tailored for HSI data, ensuring continuity without the computational burden of convolution-based approaches. Extensive experiments on benchmark datasets demonstrate that MemFormer achieves superior classification accuracy, outperforming state-of-the-art methods.
Computer-aided Whole Slide Image (WSI) classification has the potential to enhance the accuracy and efficiency of clinical pathological diagnosis. It is commonly formulated as a Multiple Instance Learning (MIL) problem, where each WSI is treated as a bag and the small patches extracted from the WSI are considered instances within that bag. However, obtaining labels for a large number of bags is a costly and time-consuming process, particularly when utilizing existing WSIs for new classification tasks. This limitation renders most existing WSI classification methods ineffective. To address this issue, we propose a novel WSI classification problem setting, more aligned with clinical practice, termed Weakly Semi-supervised Whole slide image Classification (WSWC). In WSWC, a small number of bags are labeled, while a significant number of bags remain unlabeled. The MIL nature of the WSWC problem, coupled with the absence of patch labels, distinguishes it from typical semi-supervised image classification problems, making existing algorithms for natural images unsuitable for directly solving the WSWC problem. In this paper, we present a concise and efficient framework, named CroCo, to tackle the WSWC problem through two-level Cross Consistency supervision. CroCo comprises two heterogeneous classifier branches capable of performing both instance classification and bag classification. The fundamental idea is to establish cross-consistency supervision at both the bag-level and instance-level between the two branches during training. Extensive experiments conducted on four datasets demonstrate that CroCo achieves superior bag classification and instance classification performance compared to other comparative methods when limited WSIs with bag labels are available. To the best of our knowledge, this paper presents for the first time the WSWC problem and gives a successful resolution.
Fine-Grained Visual Classification (FGVC) aims to categorize closely related subclasses, a task complicated by minimal inter-class differences and significant intra-class variance. Existing methods often rely on additional annotations for image classification, overlooking the valuable information embedded in Tree Hierarchies that depict hierarchical label relationships. To leverage this knowledge to improve classification accuracy and consistency, we propose a novel Cross-Hierarchical Bidirectional Consistency Learning (CHBC) framework. The CHBC framework extracts discriminative features across various hierarchies using a specially designed module to decompose and enhance attention masks and features. We employ bidirectional consistency loss to regulate the classification outcomes across different hierarchies, ensuring label prediction consistency and reducing misclassification. Experiments on three widely used FGVC datasets validate the effectiveness of the CHBC framework. Ablation studies further investigate the application strategies of feature enhancement and consistency constraints, underscoring the significant contributions of the proposed modules.
The demand for lightweight models in image classification tasks under resource-constrained environments necessitates a balance between computational efficiency and robust feature representation. Traditional attention mechanisms, despite their strong feature modeling capability, often struggle with high computational complexity and structural rigidity, limiting their applicability in scenarios with limited computational resources (e.g., edge devices or real-time systems). To address this, we propose the Multi-Agent Aggregation Module (MAAM), a lightweight attention architecture integrated with the MindSpore framework. MAAM employs three parallel agent branches with independently parameterized operations to extract heterogeneous features, adaptively fused via learnable scalar weights, and refined through a convolutional compression layer. Leveraging MindSpore's dynamic computational graph and operator fusion, MAAM achieves 87.0% accuracy on the CIFAR-10 dataset, significantly outperforming conventional CNN (58.3%) and MLP (49.6%) models, while improving training efficiency by 30%. Ablation studies confirm the critical role of agent attention (accuracy drops to 32.0% if removed) and compression modules (25.5% if omitted), validating their necessity for maintaining discriminative feature learning. The framework's hardware acceleration capabilities and minimal memory footprint further demonstrate its practicality, offering a deployable solution for image classification in resource-constrained scenarios without compromising accuracy.
Over the past years, advances in artificial intelligence (AI) have demonstrated how AI can solve many perception and generation tasks, such as image classification and text writing, yet reasoning remains a challenge. This paper introduces the FLIP dataset, a benchmark for evaluating AI reasoning capabilities based on human verification tasks on the Idena blockchain. FLIP challenges present users with two orderings of 4 images, requiring them to identify the logically coherent one. By emphasizing sequential reasoning, visual storytelling, and common sense, FLIP provides a unique testbed for multimodal AI systems. Our experiments evaluate state-of-the-art models, leveraging both vision-language models (VLMs) and large language models (LLMs). Results reveal that even the best open-sourced and closed-sourced models achieve maximum accuracies of 75.5% and 77.9%, respectively, in zero-shot settings, compared to human performance of 95.3%. Captioning models aid reasoning models by providing text descriptions of images, yielding better results than when using the raw images directly, 69.6% vs. 75.2% for Gemini 1.5 Pro. Combining the predictions from 15 models in an ensemble increases the accuracy to 85.2%. These findings highlight the limitations of existing reasoning models and the need for robust multimodal benchmarks like FLIP. The full codebase and dataset will be available at https://github.com/aplesner/FLIP-Reasoning-Challenge.
Humans can develop internal world models that encode common sense knowledge, telling them how the world works and predicting the consequences of their actions. This concept has emerged as a promising direction for establishing general-purpose machine-learning models in recent preliminary works, e.g., for visual representation learning. In this paper, we present CheXWorld, the first effort towards a self-supervised world model for radiographic images. Specifically, our work develops a unified framework that simultaneously models three aspects of medical knowledge essential for qualified radiologists, including 1) local anatomical structures describing the fine-grained characteristics of local tissues (e.g., architectures, shapes, and textures); 2) global anatomical layouts describing the global organization of the human body (e.g., layouts of organs and skeletons); and 3) domain variations that encourage CheXWorld to model the transitions across different appearance domains of radiographs (e.g., varying clarity, contrast, and exposure caused by collecting radiographs from different hospitals, devices, or patients). Empirically, we design tailored qualitative and quantitative analyses, revealing that CheXWorld successfully captures these three dimensions of medical knowledge. Furthermore, transfer learning experiments across eight medical image classification and segmentation benchmarks showcase that CheXWorld significantly outperforms existing SSL methods and large-scale medical foundation models. Code & pre-trained models are available at https://github.com/LeapLabTHU/CheXWorld.
Neuroblastoma, adrenal-derived, is among the most common pediatric solid malignancies, characterized by significant clinical heterogeneity. Timely and accurate pathological diagnosis from hematoxylin and eosin-stained whole-slide images is critical for patient prognosis. However, current diagnostic practices primarily rely on subjective manual examination by pathologists, leading to inconsistent accuracy. Existing automated whole-slide image classification methods encounter challenges such as poor interpretability, limited feature extraction capabilities, and high computational costs, restricting their practical clinical deployment. To overcome these limitations, we propose CMSwinKAN, a contrastive-learning-based multi-scale feature fusion model tailored for pathological image classification, which enhances the Swin Transformer architecture by integrating a Kernel Activation Network within its multilayer perceptron and classification head modules, significantly improving both interpretability and accuracy. By fusing multi-scale features and leveraging contrastive learning strategies, CMSwinKAN mimics clinicians' comprehensive approach, effectively capturing global and local tissue characteristics. Additionally, we introduce a heuristic soft voting mechanism guided by clinical insights to bridge patch-level predictions to whole-slide image-level classifications seamlessly. We verified the CMSwinKAN on the publicly available BreakHis dataset and the PpNTs dataset, which was established by our hospital. Results demonstrate that CMSwinKAN performs better than existing state-of-the-art pathology-specific models pre-trained on large datasets. Our source code is available at https://github.com/JSLiam94/CMSwinKAN.
This work aligns deep learning (DL) with human reasoning capabilities and needs to enable more efficient, interpretable, and robust image classification. We approach this from three perspectives: explainability, causality, and biological vision. Introduction and background open this work before diving into operative chapters. First, we assess neural networks' visualization techniques for medical images and validate an explainable-by-design method for breast mass classification. A comprehensive review at the intersection of XAI and causality follows, where we introduce a general scaffold to organize past and future research, laying the groundwork for our second perspective. In the causality direction, we propose novel modules that exploit feature co-occurrence in medical images, leading to more effective and explainable predictions. We further introduce CROCODILE, a general framework that integrates causal concepts, contrastive learning, feature disentanglement, and prior knowledge to enhance generalization. Lastly, we explore biological vision, examining how humans recognize objects, and propose CoCoReco, a connectivity-inspired network with context-aware attention mechanisms. Overall, our key findings include: (i) simple activation maximization lacks insight for medical imaging DL models; (ii) prototypical-part learning is effective and radiologically aligned; (iii) XAI and causal ML are deeply connected; (iv) weak causal signals can be leveraged without a priori information to improve performance and interpretability; (v) our framework generalizes across medical domains and out-of-distribution data; (vi) incorporating biological circuit motifs improves human-aligned recognition. This work contributes toward human-aligned DL and highlights pathways to bridge the gap between research and clinical adoption, with implications for improved trust, diagnostic accuracy, and safe deployment.
Deep learning has achieved significant breakthroughs in medical imaging, but these advancements are often dependent on large, well-annotated datasets. However, obtaining such datasets poses a significant challenge, as it requires time-consuming and labor-intensive annotations from medical experts. Consequently, there is growing interest in learning paradigms such as incomplete, inexact, and absent supervision, which are designed to operate under limited, inexact, or missing labels. This survey categorizes and reviews the evolving research in these areas, analyzing around 600 notable contributions since 2018. It covers tasks such as image classification, segmentation, and detection across various medical application areas, including but not limited to brain, chest, and cardiac imaging. We attempt to establish the relationships among existing research studies in related areas. We provide formal definitions of different learning paradigms and offer a comprehensive summary and interpretation of various learning mechanisms and strategies, aiding readers in better understanding the current research landscape and ideas. We also discuss potential future research challenges.
Deep neural networks face several challenges in hyperspectral image classification, including high-dimensional data, sparse distribution of ground objects, and spectral redundancy, which often lead to classification overfitting and limited generalization capability. To more efficiently adapt to ground object distributions while extracting image features without introducing excessive parameters and skipping redundant information, this paper proposes EKGNet based on an improved 3D-DenseNet model, consisting of a context-aware mapping network and a dynamic kernel generation module. The context-aware mapping module translates global contextual information of hyperspectral inputs into instructions for combining base convolutional kernels, while the dynamic kernels are composed of K groups of base convolutions, analogous to K different types of experts specializing in fundamental patterns across various dimensions. The mapping module and dynamic kernel generation mechanism form a tightly coupled system - the former generates meaningful combination weights based on inputs, while the latter constructs an adaptive expert convolution system using these weights. This dynamic approach enables the model to focus more flexibly on key spatial structures when processing different regions, rather than relying on the fixed receptive field of a single static convolutional kernel. EKGNet enhances model representation capability through a 3D dynamic expert convolution system without increasing network depth or width. The proposed method demonstrates superior performance on IN, UP, and KSC datasets, outperforming mainstream hyperspectral image classification approaches.