Early detection of eye diseases like glaucoma, macular degeneration, and diabetic retinopathy is crucial for preventing vision loss. While artificial intelligence (AI) foundation models hold significant promise for addressing these challenges, existing ophthalmic foundation models primarily focus on a single modality, whereas diagnosing eye diseases requires multiple modalities. A critical yet often overlooked aspect is harnessing the multi-view information across various modalities for the same patient. Additionally, due to the long-tail nature of ophthalmic diseases, standard fully supervised or unsupervised learning approaches often struggle. Therefore, it is essential to integrate clinical text to capture a broader spectrum of diseases. We propose EyeCLIP, a visual-language foundation model developed using over 2.77 million multi-modal ophthalmology images with partial text data. To fully leverage the large multi-modal unlabeled and labeled data, we introduced a pretraining strategy that combines self-supervised reconstructions, multi-modal image contrastive learning, and image-text contrastive learning to learn a shared representation of multiple modalities. Through evaluation using 14 benchmark datasets, EyeCLIP can be transferred to a wide range of downstream tasks involving ocular and systemic diseases, achieving state-of-the-art performance in disease classification, visual question answering, and cross-modal retrieval. EyeCLIP represents a significant advancement over previous methods, especially showcasing few-shot, even zero-shot capabilities in real-world long-tail scenarios.
Heterogeneous graph neural networks (HGNNs) have significantly propelled the information retrieval (IR) field. Still, the effectiveness of HGNNs heavily relies on high-quality labels, which are often expensive to acquire. This challenge has shifted attention towards Heterogeneous Graph Contrastive Learning (HGCL), which usually requires pre-defined meta-paths. However, our findings reveal that meta-path combinations significantly affect performance in unsupervised settings, an aspect often overlooked in current literature. Existing HGCL methods have considerable variability in outcomes across different meta-path combinations, thereby challenging the optimization process to achieve consistent and high performance. In response, we introduce \textsf{LAMP} (\underline{\textbf{L}}earn\underline{\textbf{A}}ble \underline{\textbf{M}}eta-\underline{\textbf{P}}ath), a novel adversarial contrastive learning approach that integrates various meta-path sub-graphs into a unified and stable structure, leveraging the overlap among these sub-graphs. To address the denseness of this integrated sub-graph, we propose an adversarial training strategy for edge pruning, maintaining sparsity to enhance model performance and robustness. \textsf{LAMP} aims to maximize the difference between meta-path and network schema views for guiding contrastive learning to capture the most meaningful information. Our extensive experimental study conducted on four diverse datasets from the Heterogeneous Graph Benchmark (HGB) demonstrates that \textsf{LAMP} significantly outperforms existing state-of-the-art unsupervised models in terms of accuracy and robustness.
The increasing size of screening libraries poses a significant challenge for the development of virtual screening methods for drug discovery, necessitating a re-evaluation of traditional approaches in the era of big data. Although 3D pharmacophore screening remains a prevalent technique, its application to very large datasets is limited by the computational cost associated with matching query pharmacophores to database molecules. In this study, we introduce PharmacoMatch, a novel contrastive learning approach based on neural subgraph matching. Our method reinterprets pharmacophore screening as an approximate subgraph matching problem and enables efficient querying of conformational databases by encoding query-target relationships in the embedding space. We conduct comprehensive investigations of the learned representations and evaluate PharmacoMatch as pre-screening tool in a zero-shot setting. We demonstrate significantly shorter runtimes and comparable performance metrics to existing solutions, providing a promising speed-up for screening very large datasets.
Graph neural network(GNN) has been a powerful approach in collaborative filtering(CF) due to its ability to model high-order user-item relationships. Recently, to alleviate the data sparsity and enhance representation learning, many efforts have been conducted to integrate contrastive learning(CL) with GNNs. Despite the promising improvements, the contrastive view generation based on structure and representation perturbations in existing methods potentially disrupts the collaborative information in contrastive views, resulting in limited effectiveness of positive alignment. To overcome this issue, we propose CoGCL, a novel framework that aims to enhance graph contrastive learning by constructing contrastive views with stronger collaborative information via discrete codes. The core idea is to map users and items into discrete codes rich in collaborative information for reliable and informative contrastive view generation. To this end, we initially introduce a multi-level vector quantizer in an end-to-end manner to quantize user and item representations into discrete codes. Based on these discrete codes, we enhance the collaborative information of contrastive views by considering neighborhood structure and semantic relevance respectively. For neighborhood structure, we propose virtual neighbor augmentation by treating discrete codes as virtual neighbors, which expands an observed user-item interaction into multiple edges involving discrete codes. Regarding semantic relevance, we identify similar users/items based on shared discrete codes and interaction targets to generate the semantically relevant view. Through these strategies, we construct contrastive views with stronger collaborative information and develop a triple-view graph contrastive learning approach. Extensive experiments on four public datasets demonstrate the effectiveness of our proposed approach.
Robust frame-wise embeddings are essential to perform video analysis and understanding tasks. We present a self-supervised method for representation learning based on aligning temporal video sequences. Our framework uses a transformer-based encoder to extract frame-level features and leverages them to find the optimal alignment path between video sequences. We introduce the novel Local-Alignment Contrastive (LAC) loss, which combines a differentiable local alignment loss to capture local temporal dependencies with a contrastive loss to enhance discriminative learning. Prior works on video alignment have focused on using global temporal ordering across sequence pairs, whereas our loss encourages identifying the best-scoring subsequence alignment. LAC uses the differentiable Smith-Waterman (SW) affine method, which features a flexible parameterization learned through the training phase, enabling the model to adjust the temporal gap penalty length dynamically. Evaluations show that our learned representations outperform existing state-of-the-art approaches on action recognition tasks.
Subject-driven text-to-image (T2I) customization has drawn significant interest in academia and industry. This task enables pre-trained models to generate novel images based on unique subjects. Existing studies adopt a self-reconstructive perspective, focusing on capturing all details of a single image, which will misconstrue the specific image's irrelevant attributes (e.g., view, pose, and background) as the subject intrinsic attributes. This misconstruction leads to both overfitting or underfitting of irrelevant and intrinsic attributes of the subject, i.e., these attributes are over-represented or under-represented simultaneously, causing a trade-off between similarity and controllability. In this study, we argue an ideal subject representation can be achieved by a cross-differential perspective, i.e., decoupling subject intrinsic attributes from irrelevant attributes via contrastive learning, which allows the model to focus more on intrinsic attributes through intra-consistency (features of the same subject are spatially closer) and inter-distinctiveness (features of different subjects have distinguished differences). Specifically, we propose CustomContrast, a novel framework, which includes a Multilevel Contrastive Learning (MCL) paradigm and a Multimodal Feature Injection (MFI) Encoder. The MCL paradigm is used to extract intrinsic features of subjects from high-level semantics to low-level appearance through crossmodal semantic contrastive learning and multiscale appearance contrastive learning. To facilitate contrastive learning, we introduce the MFI encoder to capture cross-modal representations. Extensive experiments show the effectiveness of CustomContrast in subject similarity and text controllability.
Self-supervised learning of point cloud aims to leverage unlabeled 3D data to learn meaningful representations without reliance on manual annotations. However, current approaches face challenges such as limited data diversity and inadequate augmentation for effective feature learning. To address these challenges, we propose GS-PT, which integrates 3D Gaussian Splatting (3DGS) into point cloud self-supervised learning for the first time. Our pipeline utilizes transformers as the backbone for self-supervised pre-training and introduces novel contrastive learning tasks through 3DGS. Specifically, the transformers aim to reconstruct the masked point cloud. 3DGS utilizes multi-view rendered images as input to generate enhanced point cloud distributions and novel view images, facilitating data augmentation and cross-modal contrastive learning. Additionally, we incorporate features from depth maps. By optimizing these tasks collectively, our method enriches the tri-modal self-supervised learning process, enabling the model to leverage the correlation across 3D point clouds and 2D images from various modalities. We freeze the encoder after pre-training and test the model's performance on multiple downstream tasks. Experimental results indicate that GS-PT outperforms the off-the-shelf self-supervised learning methods on various downstream tasks including 3D object classification, real-world classifications, and few-shot learning and segmentation.
Recent advances in unsupervised representation learning often rely on knowing the number of classes to improve feature extraction and clustering. However, this assumption raises an important question: is the number of classes always necessary, and do class labels fully capture the fine-grained features within the data? In this paper, we propose Contrastive Disentangling (CD), a framework designed to learn representations without relying on class priors. CD leverages a multi-level contrastive learning strategy, integrating instance-level and feature-level contrastive losses with a normalized entropy loss to capture semantically rich and fine-grained representations. Specifically, (1) the instance-level contrastive loss separates feature representations across samples; (2) the feature-level contrastive loss promotes independence among feature heads; and (3) the normalized entropy loss ensures feature diversity and prevents feature collapse. Extensive experiments on CIFAR-10, CIFAR-100, STL-10, and ImageNet-10 demonstrate that CD outperforms existing methods in scenarios where class information is unavailable or ambiguous. The code is available at https://github.com/Hoper-J/Contrastive-Disentangling.
We introduce VidLPRO, a novel video-language (VL) pre-training framework designed specifically for robotic and laparoscopic surgery. While existing surgical VL models primarily rely on contrastive learning, we propose a more comprehensive approach to capture the intricate temporal dynamics and align video with language. VidLPRO integrates video-text contrastive learning, video-text matching, and masked language modeling objectives to learn rich VL representations. To support this framework, we present GenSurg+, a carefully curated dataset derived from GenSurgery, comprising 17k surgical video clips paired with captions generated by GPT-4 using transcripts extracted by the Whisper model. This dataset addresses the need for large-scale, high-quality VL data in the surgical domain. Extensive experiments on benchmark datasets, including Cholec80 and AutoLaparo, demonstrate the efficacy of our approach. VidLPRO achieves state-of-the-art performance in zero-shot surgical phase recognition, significantly outperforming existing surgical VL models such as SurgVLP and HecVL. Our model demonstrates improvements of up to 21.5\% in accuracy and 15.7% in F1 score, setting a new benchmark in the field. Notably, VidLPRO exhibits robust performance even with single-frame inference, while effectively scaling with increased temporal context. Ablation studies reveal the impact of frame sampling strategies on model performance and computational efficiency. These results underscore VidLPRO's potential as a foundation model for surgical video understanding.
Domain generalization (DG) task aims to learn a robust model from source domains that could handle the out-of-distribution (OOD) issue. In order to improve the generalization ability of the model in unseen domains, increasing the diversity of training samples is an effective solution. However, existing augmentation approaches always have some limitations. On the one hand, the augmentation manner in most DG methods is not enough as the model may not see the perturbed features in approximate the worst case due to the randomness, thus the transferability in features could not be fully explored. On the other hand, the causality in discriminative features is not involved in these methods, which harms the generalization ability of model due to the spurious correlations. To address these issues, we propose a Dual-stream Feature Augmentation~(DFA) method by constructing some hard features from two perspectives. Firstly, to improve the transferability, we construct some targeted features with domain related augmentation manner. Through the guidance of uncertainty, some hard cross-domain fictitious features are generated to simulate domain shift. Secondly, to take the causality into consideration, the spurious correlated non-causal information is disentangled by an adversarial mask, then the more discriminative features can be extracted through these hard causal related information. Different from previous fixed synthesizing strategy, the two augmentations are integrated into a unified learnable feature disentangle model. Based on these hard features, contrastive learning is employed to keep the semantic consistency and improve the robustness of the model. Extensive experiments on several datasets demonstrated that our approach could achieve state-of-the-art performance for domain generalization. Our code is available at: https://github.com/alusi123/DFA.
The development of LLMs has greatly enhanced the intelligence and fluency of question answering, while the emergence of retrieval enhancement has enabled models to better utilize external information. However, the presence of noise and errors in retrieved information poses challenges to the robustness of LLMs. In this work, to evaluate the model's performance under multiple interferences, we first construct a dataset based on machine reading comprehension datasets simulating various scenarios, including critical information absence, noise, and conflicts. To address the issue of model accuracy decline caused by noisy external information, we propose a data augmentation-based fine-tuning method to enhance LLM's robustness against noise. Additionally, contrastive learning approach is utilized to preserve the model's discrimination capability of external information. We have conducted experiments on both existing LLMs and our approach, the results are evaluated by GPT-4, which indicates that our proposed methods improve model robustness while strengthening the model's discrimination capability.
Image clustering, which involves grouping images into different clusters without labels, is a key task in unsupervised learning. Although previous deep clustering methods have achieved remarkable results, they only explore the intrinsic information of the image itself but overlook external supervision knowledge to improve the semantic understanding of images. Recently, visual-language pre-trained model on large-scale datasets have been used in various downstream tasks and have achieved great results. However, there is a gap between visual representation learning and textual semantic learning, and how to properly utilize the representation of two different modalities for clustering is still a big challenge. To tackle the challenges, we propose a novel image clustering framwork, named Dual-level Cross-Modal Contrastive Clustering (DXMC). Firstly, external textual information is introduced for constructing a semantic space which is adopted to generate image-text pairs. Secondly, the image-text pairs are respectively sent to pre-trained image and text encoder to obtain image and text embeddings which subsquently are fed into four well-designed networks. Thirdly, dual-level cross-modal contrastive learning is conducted between discriminative representations of different modalities and distinct level. Extensive experimental results on five benchmark datasets demonstrate the superiority of our proposed method.
Cross-domain recommendation (CDR) aims to improve recommendation accuracy in sparse domains by transferring knowledge from data-rich domains. However, existing CDR approaches often assume that user-item interaction data across domains is publicly available, neglecting user privacy concerns. Additionally, they experience performance degradation with sparse overlapping users due to their reliance on a large number of fully shared users for knowledge transfer. To address these challenges, we propose a Federated Prototype-based Contrastive Learning (CL) framework for Privacy Preserving CDR, called FedPCL-CDR. This approach utilizes non-overlapping user information and differential prototypes to improve model performance within a federated learning framework. FedPCL-CDR comprises two key modules: local domain (client) learning and global server aggregation. In the local domain, FedPCL-CDR first clusters all user data and utilizes local differential privacy (LDP) to learn differential prototypes, effectively utilizing non-overlapping user information and protecting user privacy. It then conducts knowledge transfer by employing both local and global prototypes returned from the server in a CL manner. Meanwhile, the global server aggregates differential prototypes sent from local domains to learn both local and global prototypes. Extensive experiments on four CDR tasks across Amazon and Douban datasets demonstrate that FedPCL-CDR surpasses SOTA baselines. We release our code at https://github.com/Lili1013/FedPCL CDR
The application of activity recognition in the "AI + Education" field is gaining increasing attention. However, current work mainly focuses on the recognition of activities in manually captured videos and a limited number of activity types, with little attention given to recognizing activities in surveillance images from real classrooms. In real classroom settings, normal teaching activities such as reading, account for a large proportion of samples, while rare non-teaching activities such as eating, continue to appear. This requires a model that can learn non-teaching activities from few samples without forgetting the normal teaching activities, which necessitates fewshot continual learning (FSCL) capability. To address this gap, we constructed a continual learning dataset focused on classroom surveillance image activity recognition called ARIC (Activity Recognition in Classroom). The dataset has advantages such as multiple perspectives, a wide variety of activities, and real-world scenarios, but it also presents challenges like similar activities and imbalanced sample distribution. To overcome these challenges, we designed a few-shot continual learning method that combines supervised contrastive learning (SCL) and an adaptive covariance classifier (ACC). During the base phase, we proposed a SCL approach based on feature augmentation to enhance the model's generalization ability. In the incremental phase, we employed an ACC to more accurately describe the distribution of new classes. Experimental results demonstrate that our method outperforms other existing methods on the ARIC dataset.
This paper examines the Code-Switching (CS) phenomenon where two languages intertwine within a single utterance. There exists a noticeable need for research on the CS between English and Korean. We highlight that the current Equivalence Constraint (EC) theory for CS in other languages may only partially capture English-Korean CS complexities due to the intrinsic grammatical differences between the languages. We introduce a novel Koglish dataset tailored for English-Korean CS scenarios to mitigate such challenges. First, we constructed the Koglish-GLUE dataset to demonstrate the importance and need for CS datasets in various tasks. We found the differential outcomes of various foundation multilingual language models when trained on a monolingual versus a CS dataset. Motivated by this, we hypothesized that SimCSE, which has shown strengths in monolingual sentence embedding, would have limitations in CS scenarios. We construct a novel Koglish-NLI (Natural Language Inference) dataset using a CS augmentation-based approach to verify this. From this CS-augmented dataset Koglish-NLI, we propose a unified contrastive learning and augmentation method for code-switched embeddings, ConCSE, highlighting the semantics of CS sentences. Experimental results validate the proposed ConCSE with an average performance enhancement of 1.77\% on the Koglish-STS(Semantic Textual Similarity) tasks.
Hierarchical Text Classification (HTC) aims to categorize text data based on a structured label hierarchy, resulting in predicted labels forming a sub-hierarchy tree. The semantics of the text should align with the semantics of the labels in this sub-hierarchy. With the sub-hierarchy changing for each sample, the dynamic nature of text-label alignment poses challenges for existing methods, which typically process text and labels independently. To overcome this limitation, we propose a Text-Label Alignment (TLA) loss specifically designed to model the alignment between text and labels. We obtain a set of negative labels for a given text and its positive label set. By leveraging contrastive learning, the TLA loss pulls the text closer to its positive label and pushes it away from its negative label in the embedding space. This process aligns text representations with related labels while distancing them from unrelated ones. Building upon this framework, we introduce the Hierarchical Text-Label Alignment (HTLA) model, which leverages BERT as the text encoder and GPTrans as the graph encoder and integrates text-label embeddings to generate hierarchy-aware representations. Experimental results on benchmark datasets and comparison with existing baselines demonstrate the effectiveness of HTLA for HTC.
In visual Reinforcement Learning (RL), learning from pixel-based observations poses significant challenges on sample efficiency, primarily due to the complexity of extracting informative state representations from high-dimensional data. Previous methods such as contrastive-based approaches have made strides in improving sample efficiency but fall short in modeling the nuanced evolution of states. To address this, we introduce MOOSS, a novel framework that leverages a temporal contrastive objective with the help of graph-based spatial-temporal masking to explicitly model state evolution in visual RL. Specifically, we propose a self-supervised dual-component strategy that integrates (1) a graph construction of pixel-based observations for spatial-temporal masking, coupled with (2) a multi-level contrastive learning mechanism that enriches state representations by emphasizing temporal continuity and change of states. MOOSS advances the understanding of state dynamics by disrupting and learning from spatial-temporal correlations, which facilitates policy learning. Our comprehensive evaluation on multiple continuous and discrete control benchmarks shows that MOOSS outperforms previous state-of-the-art visual RL methods in terms of sample efficiency, demonstrating the effectiveness of our method. Our code is released at https://github.com/jsun57/MOOSS.
Although foundational vision-language models (VLMs) have proven to be very successful for various semantic discrimination tasks, they still struggle to perform faithfully for fine-grained categorization. Moreover, foundational models trained on one domain do not generalize well on a different domain without fine-tuning. We attribute these to the limitations of the VLM's semantic representations and attempt to improve their fine-grained visual awareness using generative modeling. Specifically, we propose two novel methods: Generative Class Prompt Learning (GCPL) and Contrastive Multi-class Prompt Learning (CoMPLe). Utilizing text-to-image diffusion models, GCPL significantly improves the visio-linguistic synergy in class embeddings by conditioning on few-shot exemplars with learnable class prompts. CoMPLe builds on this foundation by introducing a contrastive learning component that encourages inter-class separation during the generative optimization process. Our empirical results demonstrate that such a generative class prompt learning approach substantially outperform existing methods, offering a better alternative to few shot image recognition challenges. The source code will be made available at: https://github.com/soumitri2001/GCPL.
Sparse and noisy images (SNIs), like those in spatial gene expression data, pose significant challenges for effective representation learning and clustering, which are essential for thorough data analysis and interpretation. In response to these challenges, we propose Dual Advancement of Representation Learning and Clustering (DARLC), an innovative framework that leverages contrastive learning to enhance the representations derived from masked image modeling. Simultaneously, DARLC integrates cluster assignments in a cohesive, end-to-end approach. This integrated clustering strategy addresses the "class collision problem" inherent in contrastive learning, thus improving the quality of the resulting representations. To generate more plausible positive views for contrastive learning, we employ a graph attention network-based technique that produces denoised images as augmented data. As such, our framework offers a comprehensive approach that improves the learning of representations by enhancing their local perceptibility, distinctiveness, and the understanding of relational semantics. Furthermore, we utilize a Student's t mixture model to achieve more robust and adaptable clustering of SNIs. Extensive experiments, conducted across 12 different types of datasets consisting of SNIs, demonstrate that DARLC surpasses the state-of-the-art methods in both image clustering and generating image representations that accurately capture gene interactions. Code is available at https://github.com/zipging/DARLC.
Recognizing actions from a limited set of labeled videos remains a challenge as annotating visual data is not only tedious but also can be expensive due to classified nature. Moreover, handling spatio-temporal data using deep $3$D transformers for this can introduce significant computational complexity. In this paper, our objective is to address video action recognition in a semi-supervised setting by leveraging only a handful of labeled videos along with a collection of unlabeled videos in a compute efficient manner. Specifically, we rearrange multiple frames from the input videos in row-column form to construct super images. Subsequently, we capitalize on the vast pool of unlabeled samples and employ contrastive learning on the encoded super images. Our proposed approach employs two pathways to generate representations for temporally augmented super images originating from the same video. Specifically, we utilize a 2D image-transformer to generate representations and apply a contrastive loss function to minimize the similarity between representations from different videos while maximizing the representations of identical videos. Our method demonstrates superior performance compared to existing state-of-the-art approaches for semi-supervised action recognition across various benchmark datasets, all while significantly reducing computational costs.