Distill CLIP (DCLIP): Enhancing Image-Text Retrieval via Cross-Modal Transformer Distillation

We present Distill CLIP (DCLIP), a fine-tuned variant of the CLIP model that enhances multimodal image-text retrieval while preserving the original model's strong zero-shot classification capabilities. CLIP models are typically constrained by fixed image resolutions and limited context, which can hinder their effectiveness in retrieval tasks that require fine-grained cross-modal understanding. DCLIP addresses these challenges through a meta teacher-student distillation framework, where a cross-modal transformer teacher is fine-tuned to produce enriched embeddings via bidirectional cross-attention between YOLO-extracted image regions and corresponding textual spans. These semantically and spatially aligned global representations guide the training of a lightweight student model using a hybrid loss that combines contrastive learning and cosine similarity objectives. Despite being trained on only ~67,500 samples curated from MSCOCO, Flickr30k, and Conceptual Captions-just a fraction of CLIP's original dataset-DCLIP significantly improves image-text retrieval metrics (Recall@K, MAP), while retaining approximately 94% of CLIP's zero-shot classification performance. These results demonstrate that DCLIP effectively mitigates the trade-off between task specialization and generalization, offering a resource-efficient, domain-adaptive, and detail-sensitive solution for advanced vision-language tasks. Code available at https://anonymous.4open.science/r/DCLIP-B772/README.md.

DSAGL: Dual-Stream Attention-Guided Learning for Weakly Supervised Whole Slide Image Classification

Whole-slide images (WSIs) are critical for cancer diagnosis due to their ultra-high resolution and rich semantic content. However, their massive size and the limited availability of fine-grained annotations pose substantial challenges for conventional supervised learning. We propose DSAGL (Dual-Stream Attention-Guided Learning), a novel weakly supervised classification framework that combines a teacher-student architecture with a dual-stream design. DSAGL explicitly addresses instance-level ambiguity and bag-level semantic consistency by generating multi-scale attention-based pseudo labels and guiding instance-level learning. A shared lightweight encoder (VSSMamba) enables efficient long-range dependency modeling, while a fusion-attentive module (FASA) enhances focus on sparse but diagnostically relevant regions. We further introduce a hybrid loss to enforce mutual consistency between the two streams. Experiments on CIFAR-10, NCT-CRC, and TCGA-Lung datasets demonstrate that DSAGL consistently outperforms state-of-the-art MIL baselines, achieving superior discriminative performance and robustness under weak supervision.

MAMBO: High-Resolution Generative Approach for Mammography Images

Mammography is the gold standard for the detection and diagnosis of breast cancer. This procedure can be significantly enhanced with Artificial Intelligence (AI)-based software, which assists radiologists in identifying abnormalities. However, training AI systems requires large and diverse datasets, which are often difficult to obtain due to privacy and ethical constraints. To address this issue, the paper introduces MAMmography ensemBle mOdel (MAMBO), a novel patch-based diffusion approach designed to generate full-resolution mammograms. Diffusion models have shown breakthrough results in realistic image generation, yet few studies have focused on mammograms, and none have successfully generated high-resolution outputs required to capture fine-grained features of small lesions. To achieve this, MAMBO integrates separate diffusion models to capture both local and global (image-level) contexts. The contextual information is then fed into the final patch-based model, significantly aiding the noise removal process. This thoughtful design enables MAMBO to generate highly realistic mammograms of up to 3840x3840 pixels. Importantly, this approach can be used to enhance the training of classification models and extended to anomaly detection. Experiments, both numerical and radiologist validation, assess MAMBO's capabilities in image generation, super-resolution, and anomaly detection, highlighting its potential to enhance mammography analysis for more accurate diagnoses and earlier lesion detection.

PiPViT: Patch-based Visual Interpretable Prototypes for Retinal Image Analysis

Background and Objective: Prototype-based methods improve interpretability by learning fine-grained part-prototypes; however, their visualization in the input pixel space is not always consistent with human-understandable biomarkers. In addition, well-known prototype-based approaches typically learn extremely granular prototypes that are less interpretable in medical imaging, where both the presence and extent of biomarkers and lesions are critical. Methods: To address these challenges, we propose PiPViT (Patch-based Visual Interpretable Prototypes), an inherently interpretable prototypical model for image recognition. Leveraging a vision transformer (ViT), PiPViT captures long-range dependencies among patches to learn robust, human-interpretable prototypes that approximate lesion extent only using image-level labels. Additionally, PiPViT benefits from contrastive learning and multi-resolution input processing, which enables effective localization of biomarkers across scales. Results: We evaluated PiPViT on retinal OCT image classification across four datasets, where it achieved competitive quantitative performance compared to state-of-the-art methods while delivering more meaningful explanations. Moreover, quantitative evaluation on a hold-out test set confirms that the learned prototypes are semantically and clinically relevant. We believe PiPViT can transparently explain its decisions and assist clinicians in understanding diagnostic outcomes. Github page: https://github.com/marziehoghbaie/PiPViT

Variational Supervised Contrastive Learning

Contrastive learning has proven to be highly efficient and adaptable in shaping representation spaces across diverse modalities by pulling similar samples together and pushing dissimilar ones apart. However, two key limitations persist: (1) Without explicit regulation of the embedding distribution, semantically related instances can inadvertently be pushed apart unless complementary signals guide pair selection, and (2) excessive reliance on large in-batch negatives and tailored augmentations hinders generalization. To address these limitations, we propose Variational Supervised Contrastive Learning (VarCon), which reformulates supervised contrastive learning as variational inference over latent class variables and maximizes a posterior-weighted evidence lower bound (ELBO) that replaces exhaustive pair-wise comparisons for efficient class-aware matching and grants fine-grained control over intra-class dispersion in the embedding space. Trained exclusively on image data, our experiments on CIFAR-10, CIFAR-100, ImageNet-100, and ImageNet-1K show that VarCon (1) achieves state-of-the-art performance for contrastive learning frameworks, reaching 79.36% Top-1 accuracy on ImageNet-1K and 78.29% on CIFAR-100 with a ResNet-50 encoder while converging in just 200 epochs; (2) yields substantially clearer decision boundaries and semantic organization in the embedding space, as evidenced by KNN classification, hierarchical clustering results, and transfer-learning assessments; and (3) demonstrates superior performance in few-shot learning than supervised baseline and superior robustness across various augmentation strategies.

seq-JEPA: Autoregressive Predictive Learning of Invariant-Equivariant World Models

Current self-supervised algorithms mostly rely on transformations such as data augmentation and masking to learn visual representations. This is achieved by inducing invariance or equivariance with respect to these transformations after encoding two views of an image. This dominant two-view paradigm can limit the flexibility of learned representations for downstream adaptation by creating performance trade-offs between invariance-related tasks such as image classification and more fine-grained equivariance-related tasks. In this work, we introduce \emph{seq-JEPA}, a world modeling paradigm based on joint-embedding predictive architecture that leverages architectural inductive biases to resolve this trade-off. Without requiring an additional equivariance predictor or loss term, seq-JEPA simultaneously learns two architecturally segregated representations: one equivariant to the specified transformations and another invariant to them and suited for tasks such as classification. To do so, our model processes a short sequence of different views (observations) of an input image. Each encoded view is concatenated with embeddings corresponding to the relative transformation (action) producing the next observation in the sequence. A transformer encoder outputs an aggregate representation of this sequence, which is subsequently conditioned on the action leading to the next observation to predict its representation. Empirically, seq-JEPA achieves strong performance on equivariant benchmarks and image classification without sacrificing one for the other. Additionally, our framework excels at tasks that inherently require aggregating a sequence of observations, such as path integration across actions and predictive learning across eye movements.

Cross-Modal Clustering-Guided Negative Sampling for Self-Supervised Joint Learning from Medical Images and Reports

Learning medical visual representations directly from paired images and reports through multimodal self-supervised learning has emerged as a novel and efficient approach to digital diagnosis in recent years. However, existing models suffer from several severe limitations. 1) neglecting the selection of negative samples, resulting in the scarcity of hard negatives and the inclusion of false negatives; 2) focusing on global feature extraction, but overlooking the fine-grained local details that are crucial for medical image recognition tasks; and 3) contrastive learning primarily targets high-level features but ignoring low-level details which are essential for accurate medical analysis. Motivated by these critical issues, this paper presents a Cross-Modal Cluster-Guided Negative Sampling (CM-CGNS) method with two-fold ideas. First, it extends the k-means clustering used for local text features in the single-modal domain to the multimodal domain through cross-modal attention. This improvement increases the number of negative samples and boosts the model representation capability. Second, it introduces a Cross-Modal Masked Image Reconstruction (CM-MIR) module that leverages local text-to-image features obtained via cross-modal attention to reconstruct masked local image regions. This module significantly strengthens the model's cross-modal information interaction capabilities and retains low-level image features essential for downstream tasks. By well handling the aforementioned limitations, the proposed CM-CGNS can learn effective and robust medical visual representations suitable for various recognition tasks. Extensive experimental results on classification, detection, and segmentation tasks across five downstream datasets show that our method outperforms state-of-the-art approaches on multiple metrics, verifying its superior performance.

FG-CLIP: Fine-Grained Visual and Textual Alignment

Contrastive Language-Image Pre-training (CLIP) excels in multimodal tasks such as image-text retrieval and zero-shot classification but struggles with fine-grained understanding due to its focus on coarse-grained short captions. To address this, we propose Fine-Grained CLIP (FG-CLIP), which enhances fine-grained understanding through three key innovations. First, we leverage large multimodal models to generate 1.6 billion long caption-image pairs for capturing global-level semantic details. Second, a high-quality dataset is constructed with 12 million images and 40 million region-specific bounding boxes aligned with detailed captions to ensure precise, context-rich representations. Third, 10 million hard fine-grained negative samples are incorporated to improve the model's ability to distinguish subtle semantic differences. Corresponding training methods are meticulously designed for these data. Extensive experiments demonstrate that FG-CLIP outperforms the original CLIP and other state-of-the-art methods across various downstream tasks, including fine-grained understanding, open-vocabulary object detection, image-text retrieval, and general multimodal benchmarks. These results highlight FG-CLIP's effectiveness in capturing fine-grained image details and improving overall model performance. The related data, code, and models are available at https://github.com/360CVGroup/FG-CLIP.

Enhancing Multimodal In-Context Learning for Image Classification through Coreset Optimization

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.

Holistic White-light Polyp Classification via Alignment-free Dense Distillation of Auxiliary Optical Chromoendoscopy

White Light Imaging (WLI) and Narrow Band Imaging (NBI) are the two main colonoscopic modalities for polyp classification. While NBI, as optical chromoendoscopy, offers valuable vascular details, WLI remains the most common and often the only available modality in resource-limited settings. However, WLI-based methods typically underperform, limiting their clinical applicability. Existing approaches transfer knowledge from NBI to WLI through global feature alignment but often rely on cropped lesion regions, which are susceptible to detection errors and neglect contextual and subtle diagnostic cues. To address this, this paper proposes a novel holistic classification framework that leverages full-image diagnosis without requiring polyp localization. The key innovation lies in the Alignment-free Dense Distillation (ADD) module, which enables fine-grained cross-domain knowledge distillation regardless of misalignment between WLI and NBI images. Without resorting to explicit image alignment, ADD learns pixel-wise cross-domain affinities to establish correspondences between feature maps, guiding the distillation along the most relevant pixel connections. To further enhance distillation reliability, ADD incorporates Class Activation Mapping (CAM) to filter cross-domain affinities, ensuring the distillation path connects only those semantically consistent regions with equal contributions to polyp diagnosis. Extensive results on public and in-house datasets show that our method achieves state-of-the-art performance, relatively outperforming the other approaches by at least 2.5% and 16.2% in AUC, respectively. Code is available at: https://github.com/Huster-Hq/ADD.