Cross-modal image-text retrieval is challenging because of the diverse possible associations between content from different modalities. Traditional methods learn a single-vector embedding to represent semantics of each sample, but struggle to capture nuanced and diverse relationships that can exist across modalities. Set-based approaches, which represent each sample with multiple embeddings, offer a promising alternative, as they can capture richer and more diverse relationships. In this paper, we show that, despite their promise, these set-based representations continue to face issues including sparse supervision and set collapse, which limits their effectiveness. To address these challenges, we propose Maximal Pair Assignment Similarity to optimize one-to-one matching between embedding sets which preserve semantic diversity within the set. We also introduce two loss functions to further enhance the representations: Global Discriminative Loss to enhance distinction among embeddings, and Intra-Set Divergence Loss to prevent collapse within each set. Our method achieves state-of-the-art performance on MS-COCO and Flickr30k without relying on external data.
We present FlagEvalMM, an open-source evaluation framework designed to comprehensively assess multimodal models across a diverse range of vision-language understanding and generation tasks, such as visual question answering, text-to-image/video generation, and image-text retrieval. We decouple model inference from evaluation through an independent evaluation service, thus enabling flexible resource allocation and seamless integration of new tasks and models. Moreover, FlagEvalMM utilizes advanced inference acceleration tools (e.g., vLLM, SGLang) and asynchronous data loading to significantly enhance evaluation efficiency. Extensive experiments show that FlagEvalMM offers accurate and efficient insights into model strengths and limitations, making it a valuable tool for advancing multimodal research. The framework is publicly accessible athttps://github.com/flageval-baai/FlagEvalMM.
Dual encoder Vision-Language Models (VLM) such as CLIP are widely used for image-text retrieval tasks. However, those models struggle with compositionality, showing a bag-of-words-like behavior that limits their retrieval performance. Many different training approaches have been proposed to improve the vision-language compositionality capabilities of those models. In comparison, inference-time techniques have received little attention. In this paper, we propose to add simple structure at inference, where, given an image and a caption: i) we divide the image into different smaller crops, ii) we extract text segments, capturing objects, attributes and relations, iii) using a VLM, we find the image crops that better align with text segments obtaining matches, and iv) we compute the final image-text similarity aggregating the individual similarities of the matches. Based on various popular dual encoder VLMs, we evaluate our approach in controlled and natural datasets for VL compositionality. We find that our approach consistently improves the performance of evaluated VLMs without any training, which shows the potential of inference-time techniques. The results are especially good for attribute-object binding as shown in the controlled dataset. As a result of an extensive analysis: i) we show that processing image crops is actually essential for the observed gains in performance, and ii) we identify specific areas to further improve inference-time approaches.
Foundation models represent the most prominent and recent paradigm shift in artificial intelligence. Foundation models are large models, trained on broad data that deliver high accuracy in many downstream tasks, often without fine-tuning. For this reason, models such as CLIP , DINO or Vision Transfomers (ViT), are becoming the bedrock of many industrial AI-powered applications. However, the reliance on pre-trained foundation models also introduces significant security concerns, as these models are vulnerable to adversarial attacks. Such attacks involve deliberately crafted inputs designed to deceive AI systems, jeopardizing their reliability. This paper studies the vulnerabilities of vision foundation models, focusing specifically on CLIP and ViTs, and explores the transferability of adversarial attacks to downstream tasks. We introduce a novel attack, targeting the structure of transformer-based architectures in a task-agnostic fashion. We demonstrate the effectiveness of our attack on several downstream tasks: classification, captioning, image/text retrieval, segmentation and depth estimation. Code available at:https://github.com/HondamunigePrasannaSilva/attack-attention
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.
Vision-language retrieval (VLR) has attracted significant attention in both academia and industry, which involves using text (or images) as queries to retrieve corresponding images (or text). However, existing methods often neglect the rich visual semantics knowledge of entities, thus leading to incorrect retrieval results. To address this problem, we propose the Entity Visual Description enhanced CLIP (EvdCLIP), designed to leverage the visual knowledge of entities to enrich queries. Specifically, since humans recognize entities through visual cues, we employ a large language model (LLM) to generate Entity Visual Descriptions (EVDs) as alignment cues to complement textual data. These EVDs are then integrated into raw queries to create visually-rich, EVD-enhanced queries. Furthermore, recognizing that EVD-enhanced queries may introduce noise or low-quality expansions, we develop a novel, trainable EVD-aware Rewriter (EaRW) for vision-language retrieval tasks. EaRW utilizes EVD knowledge and the generative capabilities of the language model to effectively rewrite queries. With our specialized training strategy, EaRW can generate high-quality and low-noise EVD-enhanced queries. Extensive quantitative and qualitative experiments on image-text retrieval benchmarks validate the superiority of EvdCLIP on vision-language retrieval tasks.
Remote Sensing Image-Text Retrieval (RSITR) plays a critical role in geographic information interpretation, disaster monitoring, and urban planning by establishing semantic associations between image and textual descriptions. Existing Parameter-Efficient Fine-Tuning (PEFT) methods for Vision-and-Language Pre-training (VLP) models typically adopt symmetric adapter structures for exploring cross-modal correlations. However, the strong discriminative nature of text modality may dominate the optimization process and inhibits image representation learning. The nonnegligible imbalanced cross-modal optimization remains a bottleneck to enhancing the model performance. To address this issue, this study proposes a Representation Discrepancy Bridging (RDB) method for the RSITR task. On the one hand, a Cross-Modal Asymmetric Adapter (CMAA) is designed to enable modality-specific optimization and improve feature alignment. The CMAA comprises a Visual Enhancement Adapter (VEA) and a Text Semantic Adapter (TSA). VEA mines fine-grained image features by Differential Attention (DA) mechanism, while TSA identifies key textual semantics through Hierarchical Attention (HA) mechanism. On the other hand, this study extends the traditional single-task retrieval framework to a dual-task optimization framework and develops a Dual-Task Consistency Loss (DTCL). The DTCL improves cross-modal alignment robustness through an adaptive weighted combination of cross-modal, classification, and exponential moving average consistency constraints. Experiments on RSICD and RSITMD datasets show that the proposed RDB method achieves a 6%-11% improvement in mR metrics compared to state-of-the-art PEFT methods and a 1.15%-2% improvement over the full fine-tuned GeoRSCLIP model.
Multilingual vision-language models promise universal image-text retrieval, yet their social biases remain under-explored. We present the first systematic audit of three public multilingual CLIP checkpoints -- M-CLIP, NLLB-CLIP, and CAPIVARA-CLIP -- across ten languages that vary in resource availability and grammatical gender. Using balanced subsets of \textsc{FairFace} and the \textsc{PATA} stereotype suite in a zero-shot setting, we quantify race and gender bias and measure stereotype amplification. Contrary to the assumption that multilinguality mitigates bias, every model exhibits stronger gender bias than its English-only baseline. CAPIVARA-CLIP shows its largest biases precisely in the low-resource languages it targets, while the shared cross-lingual encoder of NLLB-CLIP transports English gender stereotypes into gender-neutral languages; loosely coupled encoders largely avoid this transfer. Highly gendered languages consistently magnify all measured bias types, but even gender-neutral languages remain vulnerable when cross-lingual weight sharing imports foreign stereotypes. Aggregated metrics conceal language-specific ``hot spots,'' underscoring the need for fine-grained, language-aware bias evaluation in future multilingual vision-language research.
Leaf disease identification plays a pivotal role in smart agriculture. However, many existing studies still struggle to integrate image and textual modalities to compensate for each other's limitations. Furthermore, many of these approaches rely on pretraining with constrained datasets such as ImageNet, which lack domain-specific information. We propose SCOLD (Soft-target COntrastive learning for Leaf Disease identification), a context-aware vision-language foundation model tailored to address these challenges for agricultural tasks. SCOLD is developed using a diverse corpus of plant leaf images and corresponding symptom descriptions, comprising over 186,000 image-caption pairs aligned with 97 unique concepts. Through task-agnostic pretraining, SCOLD leverages contextual soft targets to mitigate overconfidence in contrastive learning by smoothing labels, thereby improving model generalization and robustness on fine-grained classification tasks. Experimental results demonstrate that SCOLD outperforms existing vision-language models such as OpenAI-CLIP-L, BioCLIP, and SigLIP2 across several benchmarks, including zero-shot and few-shot classification, image-text retrieval, and image classification, while maintaining a competitive parameter footprint. Ablation studies further highlight SCOLD's effectiveness in contrast to its counterparts. The proposed approach significantly advances the agricultural vision-language foundation model, offering strong performance with minimal or no supervised fine-tuning. This work lays a solid groundwork for future research on models trained with long-form and simplified contexts, tasks involving class ambiguity, and multi-modal systems for intelligent plant disease diagnostics. The code for this study is available at https://huggingface.co/enalis/scold
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. We construct a comprehensive dataset, termed FineHARD, by integrating high-quality region-specific annotations with hard fine-grained negative samples. 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 data, code, and models are available at https://github.com/360CVGroup/FG-CLIP.
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 MLLM\'s 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.
Recent advancement in large-scale Artificial Intelligence (AI) models offering multimodal services have become foundational in AI systems, making them prime targets for model theft. Existing methods select Out-of-Distribution (OoD) data as backdoor watermarks and retrain the original model for copyright protection. However, existing methods are susceptible to malicious detection and forgery by adversaries, resulting in watermark evasion. In this work, we propose Model-\underline{ag}nostic Black-box Backdoor W\underline{ate}rmarking Framework (AGATE) to address stealthiness and robustness challenges in multimodal model copyright protection. Specifically, we propose an adversarial trigger generation method to generate stealthy adversarial triggers from ordinary dataset, providing visual fidelity while inducing semantic shifts. To alleviate the issue of anomaly detection among model outputs, we propose a post-transform module to correct the model output by narrowing the distance between adversarial trigger image embedding and text embedding. Subsequently, a two-phase watermark verification is proposed to judge whether the current model infringes by comparing the two results with and without the transform module. Consequently, we consistently outperform state-of-the-art methods across five datasets in the downstream tasks of multimodal image-text retrieval and image classification. Additionally, we validated the robustness of AGATE under two adversarial attack scenarios.
Visual understanding is inherently contextual -- what we focus on in an image depends on the task at hand. For instance, given an image of a person holding a bouquet of flowers, we may focus on either the person such as their clothing, or the type of flowers, depending on the context of interest. Yet, most existing image encoding paradigms represent an image as a fixed, generic feature vector, overlooking the potential needs of prioritizing varying visual information for different downstream use cases. In this work, we introduce FocalLens, a conditional visual encoding method that produces different representations for the same image based on the context of interest, expressed flexibly through natural language. We leverage vision instruction tuning data and contrastively finetune a pretrained vision encoder to take natural language instructions as additional inputs for producing conditional image representations. Extensive experiments validate that conditional image representation from FocalLens better pronounce the visual features of interest compared to generic features produced by standard vision encoders like CLIP. In addition, we show FocalLens further leads to performance improvements on a range of downstream tasks including image-image retrieval, image classification, and image-text retrieval, with an average gain of 5 and 10 points on the challenging SugarCrepe and MMVP-VLM benchmarks, respectively.
Despite that deep learning (DL) methods have presented tremendous potential in many medical image analysis tasks, the practical applications of medical DL models are limited due to the lack of enough data samples with manual annotations. By noting that the clinical radiology examinations are associated with radiology reports that describe the images, we propose to develop a foundation model for multi-model head MRI by using contrastive learning on the images and the corresponding radiology findings. In particular, a contrastive learning framework is proposed, where a mixed syntax and semantic similarity matching metric is integrated to reduce the thirst of extreme large dataset in conventional contrastive learning framework. Our proposed similarity enhanced contrastive language image pretraining (SeLIP) is able to effectively extract more useful features. Experiments revealed that our proposed SeLIP performs well in many downstream tasks including image-text retrieval task, classification task, and image segmentation, which highlights the importance of considering the similarities among texts describing different images in developing medical image foundation models.
Vision-Language Models (VLMs) have recently emerged as powerful tools, excelling in tasks that integrate visual and textual comprehension, such as image captioning, visual question answering, and image-text retrieval. However, existing benchmarks for VLMs include spatial components, which often fail to isolate spatial reasoning from related tasks such as object detection or semantic comprehension. In this paper, we address these deficiencies with a multi-faceted approach towards understanding spatial reasoning. Informed by the diverse and multi-dimensional nature of human spatial reasoning abilities, we present a detailed analysis that first delineates the core elements of spatial reasoning: spatial relations, orientation and navigation, mental rotation, and spatial visualization, and then assesses the performance of these models in both synthetic and real-world images, bridging controlled and naturalistic contexts. We analyze 13 state-of-the-art Vision-Language Models, uncovering pivotal insights into their spatial reasoning performance. Our results reveal profound shortcomings in current VLMs, with average accuracy across the 13 models approximating random chance, highlighting spatial reasoning as a persistent obstacle. This work not only exposes the pressing need to advance spatial reasoning within VLMs but also establishes a solid platform for future exploration. Code available on GitHub (https://github.com/stogiannidis/srbench) and dataset available on HuggingFace (https://huggingface.co/datasets/stogiannidis/srbench).
Vision-language models (VLMs) have shown promise in 2D medical image analysis, but extending them to 3D remains challenging due to the high computational demands of volumetric data and the difficulty of aligning 3D spatial features with clinical text. We present Med3DVLM, a 3D VLM designed to address these challenges through three key innovations: (1) DCFormer, an efficient encoder that uses decomposed 3D convolutions to capture fine-grained spatial features at scale; (2) SigLIP, a contrastive learning strategy with pairwise sigmoid loss that improves image-text alignment without relying on large negative batches; and (3) a dual-stream MLP-Mixer projector that fuses low- and high-level image features with text embeddings for richer multi-modal representations. We evaluate our model on the M3D dataset, which includes radiology reports and VQA data for 120,084 3D medical images. Results show that Med3DVLM achieves superior performance across multiple benchmarks. For image-text retrieval, it reaches 61.00% R@1 on 2,000 samples, significantly outperforming the current state-of-the-art M3D model (19.10%). For report generation, it achieves a METEOR score of 36.42% (vs. 14.38%). In open-ended visual question answering (VQA), it scores 36.76% METEOR (vs. 33.58%), and in closed-ended VQA, it achieves 79.95% accuracy (vs. 75.78%). These results highlight Med3DVLM's ability to bridge the gap between 3D imaging and language, enabling scalable, multi-task reasoning across clinical applications. Our code is publicly available at https://github.com/mirthAI/Med3DVLM.
Universal multimodal embedding models play a critical role in tasks such as interleaved image-text retrieval, multimodal RAG, and multimodal clustering. However, our empirical results indicate that existing LMM-based embedding models trained with the standard InfoNCE loss exhibit a high degree of overlap in similarity distribution between positive and negative pairs, making it challenging to distinguish hard negative pairs effectively. To deal with this issue, we propose a simple yet effective framework that dynamically improves the embedding model's representation learning for negative pairs based on their discriminative difficulty. Within this framework, we train a series of models, named LLaVE, and evaluate them on the MMEB benchmark, which covers 4 meta-tasks and 36 datasets. Experimental results show that LLaVE establishes stronger baselines that achieve state-of-the-art (SOTA) performance while demonstrating strong scalability and efficiency. Specifically, LLaVE-2B surpasses the previous SOTA 7B models, while LLaVE-7B achieves a further performance improvement of 6.2 points. Although LLaVE is trained on image-text data, it can generalize to text-video retrieval tasks in a zero-shot manner and achieve strong performance, demonstrating its remarkable potential for transfer to other embedding tasks.
Image-Text Retrieval (ITR) finds broad applications in healthcare, aiding clinicians and radiologists by automatically retrieving relevant patient cases in the database given the query image and/or report, for more efficient clinical diagnosis and treatment, especially for rare diseases. However conventional ITR systems typically only rely on global image or text representations for measuring patient image/report similarities, which overlook local distinctiveness across patient cases. This often results in suboptimal retrieval performance. In this paper, we propose an Anatomical Location-Conditioned Image-Text Retrieval (ALC-ITR) framework, which, given a query image and the associated suspicious anatomical region(s), aims to retrieve similar patient cases exhibiting the same disease or symptoms in the same anatomical region. To perform location-conditioned multimodal retrieval, we learn a medical Relevance-Region-Aligned Vision Language (RRA-VL) model with semantic global-level and region-/word-level alignment to produce generalizable, well-aligned multi-modal representations. Additionally, we perform location-conditioned contrastive learning to further utilize cross-pair region-level contrastiveness for improved multi-modal retrieval. We show that our proposed RRA-VL achieves state-of-the-art localization performance in phase-grounding tasks, and satisfying multi-modal retrieval performance with or without location conditioning. Finally, we thoroughly investigate the generalizability and explainability of our proposed ALC-ITR system in providing explanations and preliminary diagnosis reports given retrieved patient cases (conditioned on anatomical regions), with proper off-the-shelf LLM prompts.
Training vision-language models for image-text alignment typically requires large datasets to achieve robust performance. In low-data scenarios, standard contrastive learning can struggle to align modalities effectively due to overfitting and unstable training dynamics. In this paper, we propose a variance-aware loss scheduling approach that dynamically adjusts the weighting of the contrastive loss based on the statistical variability (uncertainty) in the model's alignment predictions. Using a subset of the Flickr8k image-caption dataset to simulate limited data conditions, we demonstrate that our approach improves image-text retrieval accuracy compared to a fixed-weight baseline. We also compare against other adaptive weighting strategies (using output entropy and cosine similarity spread) and find that variance-aware scheduling provides the best overall trade-off. Qualitatively, our method yields more distinct multimodal embeddings as shown by t-SNE visualizations. Moreover, in a stress test with noise-injected captions and images, the variance-guided loss proves more robust, maintaining higher recall when random perturbations are introduced. These results highlight the benefit of adaptive loss weighting for multimodal alignment in low-data regimes.
Despite significant progress in Vision-Language Pre-training (VLP), current approaches predominantly emphasize feature extraction and cross-modal comprehension, with limited attention to generating or transforming visual content. This gap hinders the model's ability to synthesize coherent and novel visual representations from textual prompts, thereby reducing the effectiveness of multi-modal learning. In this work, we propose MedUnifier, a unified VLP framework tailored for medical data. MedUnifier seamlessly integrates text-grounded image generation capabilities with multi-modal learning strategies, including image-text contrastive alignment, image-text matching and image-grounded text generation. Unlike traditional methods that reply on continuous visual representations, our approach employs visual vector quantization, which not only facilitates a more cohesive learning strategy for cross-modal understanding but also enhances multi-modal generation quality by effectively leveraging discrete representations. Our framework's effectiveness is evidenced by the experiments on established benchmarks, including uni-modal tasks (supervised fine-tuning), cross-modal tasks (image-text retrieval and zero-shot image classification), and multi-modal tasks (medical report generation, image synthesis), where it achieves state-of-the-art performance across various tasks. MedUnifier also offers a highly adaptable tool for a wide range of language and vision tasks in healthcare, marking advancement toward the development of a generalizable AI model for medical applications.