Forecasting future stock trends remains challenging for academia and industry due to stochastic inter-stock dynamics and hierarchical intra-stock dynamics influencing stock prices. In recent years, graph neural networks have achieved remarkable performance in this problem by formulating multiple stocks as graph-structured data. However, most of these approaches rely on artificially defined factors to construct static stock graphs, which fail to capture the intrinsic interdependencies between stocks that rapidly evolve. In addition, these methods often ignore the hierarchical features of the stocks and lose distinctive information within. In this work, we propose a novel graph learning approach implemented without expert knowledge to address these issues. First, our approach automatically constructs dynamic stock graphs by entropy-driven edge generation from a signal processing perspective. Then, we further learn task-optimal dependencies between stocks via a generalized graph diffusion process on constructed stock graphs. Last, a decoupled representation learning scheme is adopted to capture distinctive hierarchical intra-stock features. Experimental results demonstrate substantial improvements over state-of-the-art baselines on real-world datasets. Moreover, the ablation study and sensitivity study further illustrate the effectiveness of the proposed method in modeling the time-evolving inter-stock and intra-stock dynamics.
Existing contrastive language-image pre-training aims to learn a joint representation by matching abundant image-text pairs. However, the number of image-text pairs in medical datasets is usually orders of magnitude smaller than that in natural datasets. Besides, medical image-text pairs often involve numerous complex fine-grained correspondences. This paper aims to enhance the data efficiency by introducing multiple-to-multiple local relationship modeling to capture denser supervisions. More specifically, we propose a Medical Language-Image Pre-training (MLIP) framework, which exploits the limited image-text medical data more efficiently through patch-sentence matching. Furthermore, we introduce a masked contrastive learning strategy with semantic integrity estimation to reduce redundancy in images while preserving the underlying semantics. Our evaluation results show that MLIP outperforms previous work in zero/few-shot classification and few-shot segmentation tasks by a large margin.
The development of medical vision-language foundation models has attracted significant attention in the field of medicine and healthcare due to their promising prospect in various clinical applications. While previous studies have commonly focused on feature learning at a single learning scale, investigation on integrating multi-scale information is lacking, which may hinder the potential for mutual reinforcement among these features. This paper aims to bridge this gap by proposing a method that effectively exploits multi-scale information to enhance the performance of medical foundation models. The proposed method simultaneously exploits features at the local, instance, modality and global aspects, facilitating comprehensive representation learning within the models. We evaluate the effectiveness of the proposed method on six open-source datasets across different clinical tasks, demonstrating its ability to enhance the performance of medical foundation models.
Multi-label image classification presents a challenging task in many domains, including computer vision and medical imaging. Recent advancements have introduced graph-based and transformer-based methods to improve performance and capture label dependencies. However, these methods often include complex modules that entail heavy computation and lack interpretability. In this paper, we propose Probabilistic Multi-label Contrastive Learning (ProbMCL), a novel framework to address these challenges in multi-label image classification tasks. Our simple yet effective approach employs supervised contrastive learning, in which samples that share enough labels with an anchor image based on a decision threshold are introduced as a positive set. This structure captures label dependencies by pulling positive pair embeddings together and pushing away negative samples that fall below the threshold. We enhance representation learning by incorporating a mixture density network into contrastive learning and generating Gaussian mixture distributions to explore the epistemic uncertainty of the feature encoder. We validate the effectiveness of our framework through experimentation with datasets from the computer vision and medical imaging domains. Our method outperforms the existing state-of-the-art methods while achieving a low computational footprint on both datasets. Visualization analyses also demonstrate that ProbMCL-learned classifiers maintain a meaningful semantic topology.
What does learning to model relationships between strings teach large language models (LLMs) about the visual world? We systematically evaluate LLMs' abilities to generate and recognize an assortment of visual concepts of increasing complexity and then demonstrate how a preliminary visual representation learning system can be trained using models of text. As language models lack the ability to consume or output visual information as pixels, we use code to represent images in our study. Although LLM-generated images do not look like natural images, results on image generation and the ability of models to correct these generated images indicate that precise modeling of strings can teach language models about numerous aspects of the visual world. Furthermore, experiments on self-supervised visual representation learning, utilizing images generated with text models, highlight the potential to train vision models capable of making semantic assessments of natural images using just LLMs.
Thyroid cancer is the most common endocrine malignancy, and accurately distinguishing between benign and malignant thyroid tumors is crucial for developing effective treatment plans in clinical practice. Pathologically, thyroid tumors pose diagnostic challenges due to improper specimen sampling. In this study, we have designed a three-stage model using representation learning to integrate pixel-level and slice-level annotations for distinguishing thyroid tumors. This structure includes a pathology structure recognition method to predict structures related to thyroid tumors, an encoder-decoder network to extract pixel-level annotation information by learning the feature representations of image blocks, and an attention-based learning mechanism for the final classification task. This mechanism learns the importance of different image blocks in a pathological region, globally considering the information from each block. In the third stage, all information from the image blocks in a region is aggregated using attention mechanisms, followed by classification to determine the category of the region. Experimental results demonstrate that our proposed method can predict microscopic structures more accurately. After color-coding, the method achieves results on unstained pathology slides that approximate the quality of Hematoxylin and eosin staining, reducing the need for stained pathology slides. Furthermore, by leveraging the concept of indirect measurement and extracting polarized features from structures correlated with lesions, the proposed method can also classify samples where membrane structures cannot be obtained through sampling, providing a potential objective and highly accurate indirect diagnostic technique for thyroid tumors.
Proactively and naturally guiding the dialog from the non-recommendation context (e.g., Chit-chat) to the recommendation scenario (e.g., Music) is crucial for the Conversational Recommender System (CRS). Prior studies mainly focus on planning the next dialog goal~(e.g., chat on a movie star) conditioned on the previous dialog. However, we find the dialog goals can be simultaneously observed at different levels, which can be utilized to improve CRS. In this paper, we propose Dual-space Hierarchical Learning (DHL) to leverage multi-level goal sequences and their hierarchical relationships for conversational recommendation. Specifically, we exploit multi-level goal sequences from both the representation space and the optimization space. In the representation space, we propose the hierarchical representation learning where a cross attention module derives mutually enhanced multi-level goal representations. In the optimization space, we devise the hierarchical weight learning to reweight lower-level goal sequences, and introduce bi-level optimization for stable update. Additionally, we propose a soft labeling strategy to guide optimization gradually. Experiments on two real-world datasets verify the effectiveness of our approach. Code and data are available here.
The crux of graph classification lies in the effective representation learning for the entire graph. Typical graph neural networks focus on modeling the local dependencies when aggregating features of neighboring nodes, and obtain the representation for the entire graph by aggregating node features. Such methods have two potential limitations: 1) the global node saliency w.r.t. graph classification is not explicitly modeled, which is crucial since different nodes may have different semantic relevance to graph classification; 2) the graph representation directly aggregated from node features may have limited effectiveness to reflect graph-level information. In this work, we propose the Saliency-Aware Regularized Graph Neural Network (SAR-GNN) for graph classification, which consists of two core modules: 1) a traditional graph neural network serving as the backbone for learning node features and 2) the Graph Neural Memory designed to distill a compact graph representation from node features of the backbone. We first estimate the global node saliency by measuring the semantic similarity between the compact graph representation and node features. Then the learned saliency distribution is leveraged to regularize the neighborhood aggregation of the backbone, which facilitates the message passing of features for salient nodes and suppresses the less relevant nodes. Thus, our model can learn more effective graph representation. We demonstrate the merits of SAR-GNN by extensive experiments on seven datasets across various types of graph data. Code will be released.
Graphs are typical non-Euclidean data of complex structures. In recent years, Riemannian graph representation learning has emerged as an exciting alternative to Euclidean ones. However, Riemannian methods are still in an early stage: most of them present a single curvature (radius) regardless of structural complexity, suffer from numerical instability due to the exponential/logarithmic map, and lack the ability to capture motif regularity. In light of the issues above, we propose the problem of \emph{Motif-aware Riemannian Graph Representation Learning}, seeking a numerically stable encoder to capture motif regularity in a diverse-curvature manifold without labels. To this end, we present a novel Motif-aware Riemannian model with Generative-Contrastive learning (MotifRGC), which conducts a minmax game in Riemannian manifold in a self-supervised manner. First, we propose a new type of Riemannian GCN (D-GCN), in which we construct a diverse-curvature manifold by a product layer with the diversified factor, and replace the exponential/logarithmic map by a stable kernel layer. Second, we introduce a motif-aware Riemannian generative-contrastive learning to capture motif regularity in the constructed manifold and learn motif-aware node representation without external labels. Empirical results show the superiority of MofitRGC.
Self-supervised pre-training paradigms have been extensively explored in the field of skeleton-based action recognition. In particular, methods based on masked prediction have pushed the performance of pre-training to a new height. However, these methods take low-level features, such as raw joint coordinates or temporal motion, as prediction targets for the masked regions, which is suboptimal. In this paper, we show that using high-level contextualized features as prediction targets can achieve superior performance. Specifically, we propose Skeleton2vec, a simple and efficient self-supervised 3D action representation learning framework, which utilizes a transformer-based teacher encoder taking unmasked training samples as input to create latent contextualized representations as prediction targets. Benefiting from the self-attention mechanism, the latent representations generated by the teacher encoder can incorporate the global context of the entire training samples, leading to a richer training task. Additionally, considering the high temporal correlations in skeleton sequences, we propose a motion-aware tube masking strategy which divides the skeleton sequence into several tubes and performs persistent masking within each tube based on motion priors, thus forcing the model to build long-range spatio-temporal connections and focus on action-semantic richer regions. Extensive experiments on NTU-60, NTU-120, and PKU-MMD datasets demonstrate that our proposed Skeleton2vec outperforms previous methods and achieves state-of-the-art results.
In recent years, text-to-video retrieval methods based on CLIP have experienced rapid development. The primary direction of evolution is to exploit the much wider gamut of visual and textual cues to achieve alignment. Concretely, those methods with impressive performance often design a heavy fusion block for sentence (words)-video (frames) interaction, regardless of the prohibitive computation complexity. Nevertheless, these approaches are not optimal in terms of feature utilization and retrieval efficiency. To address this issue, we adopt multi-granularity visual feature learning, ensuring the model's comprehensiveness in capturing visual content features spanning from abstract to detailed levels during the training phase. To better leverage the multi-granularity features, we devise a two-stage retrieval architecture in the retrieval phase. This solution ingeniously balances the coarse and fine granularity of retrieval content. Moreover, it also strikes a harmonious equilibrium between retrieval effectiveness and efficiency. Specifically, in training phase, we design a parameter-free text-gated interaction block (TIB) for fine-grained video representation learning and embed an extra Pearson Constraint to optimize cross-modal representation learning. In retrieval phase, we use coarse-grained video representations for fast recall of top-k candidates, which are then reranked by fine-grained video representations. Extensive experiments on four benchmarks demonstrate the efficiency and effectiveness. Notably, our method achieves comparable performance with the current state-of-the-art methods while being nearly 50 times faster.
In specific scenarios, face sketch can be used to identify a person. However, drawing a face sketch often requires exceptional skill and is time-consuming, limiting its widespread applications in actual scenarios. The new framework of sketch less face image retrieval (SLFIR)[1] attempts to overcome the barriers by providing a means for humans and machines to interact during the drawing process. Considering SLFIR problem, there is a large gap between a partial sketch with few strokes and any whole face photo, resulting in poor performance at the early stages. In this study, we propose a multigranularity (MG) representation learning (MGRL) method to address the SLFIR problem, in which we learn the representation of different granularity regions for a partial sketch, and then, by combining all MG regions of the sketches and images, the final distance was determined. In the experiments, our method outperformed state-of-the-art baselines in terms of early retrieval on two accessible datasets. Codes are available at https://github.com/ddw2AIGROUP2CQUPT/MGRL.
Masked image modeling (MIM) has emerged as a promising approach for training Vision Transformers (ViTs). The essence of MIM lies in the token-wise prediction of masked tokens, which aims to predict targets tokenized from images or generated by pre-trained models like vision-language models. While using tokenizers or pre-trained models are plausible MIM targets, they often offer spatially inconsistent targets even for neighboring tokens, complicating models to learn unified and discriminative representations. Our pilot study identifies spatial inconsistencies and suggests that resolving them can accelerate representation learning. Building upon this insight, we introduce a novel self-supervision signal called Dynamic Token Morphing (DTM), which dynamically aggregates contextually related tokens to yield contextualized targets, thereby mitigating spatial inconsistency. DTM is compatible with various SSL frameworks; we showcase improved MIM results by employing DTM, barely introducing extra training costs. Our method facilitates training by using consistent targets, resulting in 1) faster training and 2) reduced losses. Experiments on ImageNet-1K and ADE20K demonstrate the superiority of our method compared with state-of-the-art, complex MIM methods. Furthermore, the comparative evaluation of the iNaturalists and fine-grained visual classification datasets further validates the transferability of our method on various downstream tasks. Code is available at https://github.com/naver-ai/dtm
As the deep learning revolution marches on, self-supervised learning has garnered increasing attention in recent years thanks to its remarkable representation learning ability and the low dependence on labeled data. Among these varied self-supervised techniques, masked modeling has emerged as a distinctive approach that involves predicting parts of the original data that are proportionally masked during training. This paradigm enables deep models to learn robust representations and has demonstrated exceptional performance in the context of computer vision, natural language processing, and other modalities. In this survey, we present a comprehensive review of the masked modeling framework and its methodology. We elaborate on the details of techniques within masked modeling, including diverse masking strategies, recovering targets, network architectures, and more. Then, we systematically investigate its wide-ranging applications across domains. Furthermore, we also explore the commonalities and differences between masked modeling methods in different fields. Toward the end of this paper, we conclude by discussing the limitations of current techniques and point out several potential avenues for advancing masked modeling research. A paper list project with this survey is available at \url{https://github.com/Lupin1998/Awesome-MIM}.
Understanding human actions from videos of first-person view poses significant challenges. Most prior approaches explore representation learning on egocentric videos only, while overlooking the potential benefit of exploiting existing large-scale third-person videos. In this paper, (1) we develop EgoInstructor, a retrieval-augmented multimodal captioning model that automatically retrieves semantically relevant third-person instructional videos to enhance the video captioning of egocentric videos. (2) For training the cross-view retrieval module, we devise an automatic pipeline to discover ego-exo video pairs from distinct large-scale egocentric and exocentric datasets. (3) We train the cross-view retrieval module with a novel EgoExoNCE loss that pulls egocentric and exocentric video features closer by aligning them to shared text features that describe similar actions. (4) Through extensive experiments, our cross-view retrieval module demonstrates superior performance across seven benchmarks. Regarding egocentric video captioning, EgoInstructor exhibits significant improvements by leveraging third-person videos as references. Project page is available at: https://jazzcharles.github.io/Egoinstructor/
Vector quantization (VQ) is a technique to deterministically learn features with discrete codebook representations. It is commonly performed with a variational autoencoding model, VQ-VAE, which can be further extended to hierarchical structures for making high-fidelity reconstructions. However, such hierarchical extensions of VQ-VAE often suffer from the codebook/layer collapse issue, where the codebook is not efficiently used to express the data, and hence degrades reconstruction accuracy. To mitigate this problem, we propose a novel unified framework to stochastically learn hierarchical discrete representation on the basis of the variational Bayes framework, called hierarchically quantized variational autoencoder (HQ-VAE). HQ-VAE naturally generalizes the hierarchical variants of VQ-VAE, such as VQ-VAE-2 and residual-quantized VAE (RQ-VAE), and provides them with a Bayesian training scheme. Our comprehensive experiments on image datasets show that HQ-VAE enhances codebook usage and improves reconstruction performance. We also validated HQ-VAE in terms of its applicability to a different modality with an audio dataset.
Code intelligence leverages machine learning techniques to extract knowledge from extensive code corpora, with the aim of developing intelligent tools to improve the quality and productivity of computer programming. Currently, there is already a thriving research community focusing on code intelligence, with efforts ranging from software engineering, machine learning, data mining, natural language processing, and programming languages. In this paper, we conduct a comprehensive literature review on deep learning for code intelligence, from the aspects of code representation learning, deep learning techniques, and application tasks. We also benchmark several state-of-the-art neural models for code intelligence, and provide an open-source toolkit tailored for the rapid prototyping of deep-learning-based code intelligence models. In particular, we inspect the existing code intelligence models under the basis of code representation learning, and provide a comprehensive overview to enhance comprehension of the present state of code intelligence. Furthermore, we publicly release the source code and data resources to provide the community with a ready-to-use benchmark, which can facilitate the evaluation and comparison of existing and future code intelligence models (https://xcodemind.github.io). At last, we also point out several challenging and promising directions for future research.
Patient representation learning based on electronic health records (EHR) is a critical task for disease prediction. This task aims to effectively extract useful information on dynamic features. Although various existing works have achieved remarkable progress, the model performance can be further improved by fully extracting the trends, variations, and the correlation between the trends and variations in dynamic features. In addition, sparse visit records limit the performance of deep learning models. To address these issues, we propose the Multi-perspective Patient Representation Extractor (MPRE) for disease prediction. Specifically, we propose Frequency Transformation Module (FTM) to extract the trend and variation information of dynamic features in the time-frequency domain, which can enhance the feature representation. In the 2D Multi-Extraction Network (2D MEN), we form the 2D temporal tensor based on trend and variation. Then, the correlations between trend and variation are captured by the proposed dilated operation. Moreover, we propose the First-Order Difference Attention Mechanism (FODAM) to calculate the contributions of differences in adjacent variations to the disease diagnosis adaptively. To evaluate the performance of MPRE and baseline methods, we conduct extensive experiments on two real-world public datasets. The experiment results show that MPRE outperforms state-of-the-art baseline methods in terms of AUROC and AUPRC.
We introduce SynCLR, a novel approach for learning visual representations exclusively from synthetic images and synthetic captions, without any real data. We synthesize a large dataset of image captions using LLMs, then use an off-the-shelf text-to-image model to generate multiple images corresponding to each synthetic caption. We perform visual representation learning on these synthetic images via contrastive learning, treating images sharing the same caption as positive pairs. The resulting representations transfer well to many downstream tasks, competing favorably with other general-purpose visual representation learners such as CLIP and DINO v2 in image classification tasks. Furthermore, in dense prediction tasks such as semantic segmentation, SynCLR outperforms previous self-supervised methods by a significant margin, e.g., improving over MAE and iBOT by 6.2 and 4.3 mIoU on ADE20k for ViT-B/16.
Domain generalization focuses on leveraging knowledge from multiple related domains with ample training data and labels to enhance inference on unseen in-distribution (IN) and out-of-distribution (OOD) domains. In our study, we introduce a two-phase representation learning technique using multi-task learning. This approach aims to cultivate a latent space from features spanning multiple domains, encompassing both native and cross-domains, to amplify generalization to IN and OOD territories. Additionally, we attempt to disentangle the latent space by minimizing the mutual information between the prior and latent space, effectively de-correlating spurious feature correlations. Collectively, the joint optimization will facilitate domain-invariant feature learning. We assess the model's efficacy across multiple cybersecurity datasets, using standard classification metrics on both unseen IN and OOD sets, and juxtapose the results with contemporary domain generalization methods.