We introduce Metis, a foundation model for unified speech generation. Unlike previous task-specific or multi-task models, Metis follows a pre-training and fine-tuning paradigm. It is pre-trained on large-scale unlabeled speech data using masked generative modeling and then fine-tuned to adapt to diverse speech generation tasks. Specifically, 1) Metis utilizes two discrete speech representations: SSL tokens derived from speech self-supervised learning (SSL) features, and acoustic tokens directly quantized from waveforms. 2) Metis performs masked generative pre-training on SSL tokens, utilizing 300K hours of diverse speech data, without any additional condition. 3) Through fine-tuning with task-specific conditions, Metis achieves efficient adaptation to various speech generation tasks while supporting multimodal input, even when using limited data and trainable parameters. Experiments demonstrate that Metis can serve as a foundation model for unified speech generation: Metis outperforms state-of-the-art task-specific or multi-task systems across five speech generation tasks, including zero-shot text-to-speech, voice conversion, target speaker extraction, speech enhancement, and lip-to-speech, even with fewer than 20M trainable parameters or 300 times less training data. Audio samples are are available at https://metis-demo.github.io/.
Self-supervised learning has been a powerful approach for learning meaningful representations from unlabeled data across various domains, reducing the reliance on large labeled datasets. Inspired by BERT's success in capturing deep bidirectional contexts in natural language processing, similar frameworks have been adapted to other modalities such as audio, with models like BEATs extending the bidirectional training paradigm to audio signals using vector quantization (VQ). However, these frameworks face challenges, notably their dependence on a single codebook for quantization, which may not capture the complex, multifaceted nature of signals. In addition, inefficiencies in codebook utilization lead to underutilized code vectors. To address these limitations, we introduce BRIDLE (Bidirectional Residual Quantization Interleaved Discrete Learning Encoder), a self-supervised encoder pretraining framework that incorporates residual quantization (RQ) into the bidirectional training process, and is generalized for pretraining with audio, image, and video. Using multiple hierarchical codebooks, RQ enables fine-grained discretization in the latent space, enhancing representation quality. BRIDLE involves an interleaved training procedure between the encoder and tokenizer. We evaluate BRIDLE on audio understanding tasks using classification benchmarks, achieving state-of-the-art results, and demonstrate competitive performance on image classification and video classification tasks, showing consistent improvements over traditional VQ methods in downstream performance.
Self-supervised learning (SSL) algorithms have emerged as powerful tools that can leverage large quantities of unlabeled audio data to pre-train robust representations that support strong performance on diverse downstream tasks. Up to now these have mostly been developed separately for speech and non-speech applications. Here, we explored the domain specificity of a convolutional model's pre-training data relative to different downstream speech and non-speech tasks using a self-supervised pre-training approach (BYOL-A). We found that these pre-trained models (regardless of whether they were pre-trained on speech data, non-speech data or both) enabled good performance on nearly all downstream tasks, beating or nearly matching the performance of popular domain-specific models. Only small domain-specificity advantages were observed between the different pre-training datasets. The popular domain-specific models used as baselines performed very well in their target domains, but generally faltered outside of them. Together, these results demonstrate that SSL methods can be a powerful way to learn flexible representations for domain specific data without labels. These models can be a powerful resource for later transfer learning, fine-tuning or data exploration applications when the downstream data are similar, but also perhaps when there may be a domain mismatch.
Graph representation learning has emerged as a cornerstone for tasks like node classification and link prediction, yet prevailing self-supervised learning (SSL) methods face challenges such as computational inefficiency, reliance on contrastive objectives, and representation collapse. Existing approaches often depend on feature reconstruction, negative sampling, or complex decoders, which introduce training overhead and hinder generalization. Further, current techniques which address such limitations fail to account for the contribution of node embeddings to a certain prediction in the absence of labeled nodes. To address these limitations, we propose a novel joint embedding predictive framework for graph SSL that eliminates contrastive objectives and negative sampling while preserving semantic and structural information. Additionally, we introduce a semantic-aware objective term that incorporates pseudo-labels derived from Gaussian Mixture Models (GMMs), enhancing node discriminability by evaluating latent feature contributions. Extensive experiments demonstrate that our framework outperforms state-of-the-art graph SSL methods across benchmarks, achieving superior performance without contrastive loss or complex decoders. Key innovations include (1) a non-contrastive, view-invariant joint embedding predictive architecture, (2) Leveraging single context and multiple targets relationship between subgraphs, and (3) GMM-based pseudo-label scoring to capture semantic contributions. This work advances graph SSL by offering a computationally efficient, collapse-resistant paradigm that bridges spatial and semantic graph features for downstream tasks. The code for our paper can be found at https://github.com/Deceptrax123/JPEB-GSSL
Recent developments in Self-Supervised Learning (SSL) have demonstrated significant potential for Speaker Verification (SV), but closing the performance gap with supervised systems remains an ongoing challenge. Standard SSL frameworks rely on anchor-positive pairs extracted from the same audio utterances. Hence, positives have channel characteristics similar to those of their corresponding anchors, even with extensive data-augmentation. Therefore, this positive sampling strategy is a fundamental limitation as it encodes too much information regarding the recording source in the learned representations. This article introduces Self-Supervised Positive Sampling (SSPS), a bootstrapped technique for sampling appropriate and diverse positives in SSL frameworks for SV. SSPS samples positives close to their anchor in the representation space, under the assumption that these pseudo-positives belong to the same speaker identity but correspond to different recording conditions. This method demonstrates consistent improvements in SV performance on VoxCeleb benchmarks when implemented in major SSL frameworks, such as SimCLR, SwAV, VICReg, and DINO. Using SSPS, SimCLR, and DINO achieve 2.57% and 2.53% EER on VoxCeleb1-O. SimCLR yields a 58% relative reduction in EER, getting comparable performance to DINO with a simpler training framework. Furthermore, SSPS lowers intra-class variance and reduces channel information in speaker representations while exhibiting greater robustness without data-augmentation.
Optical Coherence Tomography (OCT) is a non-invasive imaging modality essential for diagnosing various eye diseases. Despite its clinical significance, developing OCT-based diagnostic tools faces challenges, such as limited public datasets, sparse annotations, and privacy concerns. Although deep learning has made progress in automating OCT analysis, these challenges remain unresolved. To address these limitations, we introduce the Vision Transformer-based Dual-Stream Self-Supervised Pretraining Network (ViT-2SPN), a novel framework designed to enhance feature extraction and improve diagnostic accuracy. ViT-2SPN employs a three-stage workflow: Supervised Pretraining, Self-Supervised Pretraining (SSP), and Supervised Fine-Tuning. The pretraining phase leverages the OCTMNIST dataset (97,477 unlabeled images across four disease classes) with data augmentation to create dual-augmented views. A Vision Transformer (ViT-Base) backbone extracts features, while a negative cosine similarity loss aligns feature representations. Pretraining is conducted over 50 epochs with a learning rate of 0.0001 and momentum of 0.999. Fine-tuning is performed on a stratified 5.129% subset of OCTMNIST using 10-fold cross-validation. ViT-2SPN achieves a mean AUC of 0.93, accuracy of 0.77, precision of 0.81, recall of 0.75, and an F1 score of 0.76, outperforming existing SSP-based methods.
Time series anomaly detection presents various challenges due to the sequential and dynamic nature of time-dependent data. Traditional unsupervised methods frequently encounter difficulties in generalization, often overfitting to known normal patterns observed during training and struggling to adapt to unseen normality. In response to this limitation, self-supervised techniques for time series have garnered attention as a potential solution to undertake this obstacle and enhance the performance of anomaly detectors. This paper presents a comprehensive review of the recent methods that make use of self-supervised learning for time series anomaly detection. A taxonomy is proposed to categorize these methods based on their primary characteristics, facilitating a clear understanding of their diversity within this field. The information contained in this survey, along with additional details that will be periodically updated, is available on the following GitHub repository: https://github.com/Aitorzan3/Awesome-Self-Supervised-Time-Series-Anomaly-Detection.
Self-supervised learning (SSL) is an emerging paradigm that exploits supervisory signals generated from the data itself, and many recent studies have leveraged SSL to conduct graph anomaly detection. However, we empirically found that three important factors can substantially impact detection performance across datasets: 1) the specific SSL strategy employed; 2) the tuning of the strategy's hyperparameters; and 3) the allocation of combination weights when using multiple strategies. Most SSL-based graph anomaly detection methods circumvent these issues by arbitrarily or selectively (i.e., guided by label information) choosing SSL strategies, hyperparameter settings, and combination weights. While an arbitrary choice may lead to subpar performance, using label information in an unsupervised setting is label information leakage and leads to severe overestimation of a method's performance. Leakage has been criticized as "one of the top ten data mining mistakes", yet many recent studies on SSL-based graph anomaly detection have been using label information to select hyperparameters. To mitigate this issue, we propose to use an internal evaluation strategy (with theoretical analysis) to select hyperparameters in SSL for unsupervised anomaly detection. We perform extensive experiments using 10 recent SSL-based graph anomaly detection algorithms on various benchmark datasets, demonstrating both the prior issues with hyperparameter selection and the effectiveness of our proposed strategy.
The core of the general recommender systems lies in learning high-quality embedding representations of users and items to investigate their positional relations in the feature space. Unfortunately, data sparsity caused by difficult-to-access interaction data severely limits the effectiveness of recommender systems. Faced with such a dilemma, various types of self-supervised learning methods have been introduced into recommender systems in an attempt to alleviate the data sparsity through distribution modeling or data augmentation. However, most data augmentation relies on elaborate manual design, which is not only not universal, but the bloated and redundant augmentation process may significantly slow down model training progress. To tackle these limitations, we propose a novel Dual Mixing-based Recommendation Framework (MixRec) to empower data augmentation as we wish. Specifically, we propose individual mixing and collective mixing, respectively. The former aims to provide a new positive sample that is unique to the target (user or item) and to make the pair-wise recommendation loss benefit from it, while the latter aims to portray a new sample that contains group properties in a batch. The two mentioned mixing mechanisms allow for data augmentation with only one parameter that does not need to be set multiple times and can be done in linear time complexity. Besides, we propose the dual-mixing contrastive learning to maximize the utilization of these new-constructed samples to enhance the consistency between pairs of positive samples. Experimental results on four real-world datasets demonstrate the advantages of MixRec in terms of effectiveness, simplicity, efficiency, and scalability.
Computed tomography (CT) and clinical numeric data are essential modalities for cancer evaluation, but building large-scale multimodal training datasets for developing medical foundation models remains challenging due to the structural complexity of multi-slice CT data and high cost of expert annotation. In this study, we propose MEDFORM, a multimodal pre-training strategy that guides CT image representation learning using complementary information from clinical data for medical foundation model development. MEDFORM efficiently processes CT slice through multiple instance learning (MIL) and adopts a dual pre-training strategy: first pretraining the CT slice feature extractor using SimCLR-based self-supervised learning, then aligning CT and clinical modalities through cross-modal contrastive learning. Our model was pre-trained on three different cancer types: lung cancer (141,171 slices), breast cancer (8,100 slices), colorectal cancer (10,393 slices). The experimental results demonstrated that this dual pre-training strategy improves cancer classification performance and maintains robust performance in few-shot learning scenarios. Code available at https://github.com/DigitalHealthcareLab/25MultiModalFoundationModel.git
With the growing demand for oriented object detection (OOD), recent studies on point-supervised OOD have attracted significant interest. In this paper, we propose PointOBB-v3, a stronger single point-supervised OOD framework. Compared to existing methods, it generates pseudo rotated boxes without additional priors and incorporates support for the end-to-end paradigm. PointOBB-v3 functions by integrating three unique image views: the original view, a resized view, and a rotated/flipped (rot/flp) view. Based on the views, a scale augmentation module and an angle acquisition module are constructed. In the first module, a Scale-Sensitive Consistency (SSC) loss and a Scale-Sensitive Feature Fusion (SSFF) module are introduced to improve the model's ability to estimate object scale. To achieve precise angle predictions, the second module employs symmetry-based self-supervised learning. Additionally, we introduce an end-to-end version that eliminates the pseudo-label generation process by integrating a detector branch and introduces an Instance-Aware Weighting (IAW) strategy to focus on high-quality predictions. We conducted extensive experiments on the DIOR-R, DOTA-v1.0/v1.5/v2.0, FAIR1M, STAR, and RSAR datasets. Across all these datasets, our method achieves an average improvement in accuracy of 3.56% in comparison to previous state-of-the-art methods. The code will be available at https://github.com/ZpyWHU/PointOBB-v3.
The group recommendation (GR) aims to suggest items for a group of users in social networks. Existing work typically considers individual preferences as the sole factor in aggregating group preferences. Actually, social influence is also an important factor in modeling users' contributions to the final group decision. However, existing methods either neglect the social influence of individual members or bundle preferences and social influence together as a unified representation. As a result, these models emphasize the preferences of the majority within the group rather than the actual interaction items, which we refer to as the preference bias issue in GR. Moreover, the self-supervised learning (SSL) strategies they designed to address the issue of group data sparsity fail to account for users' contextual social weights when regulating group representations, leading to suboptimal results. To tackle these issues, we propose a novel model based on Disentangled Modeling of Preferences and Social Influence for Group Recommendation (DisRec). Concretely, we first design a user-level disentangling network to disentangle the preferences and social influence of group members with separate embedding propagation schemes based on (hyper)graph convolution networks. We then introduce a socialbased contrastive learning strategy, selectively excluding user nodes based on their social importance to enhance group representations and alleviate the group-level data sparsity issue. The experimental results demonstrate that our model significantly outperforms state-of-the-art methods on two realworld datasets.
Recent breakthroughs in autonomous driving have been propelled by advances in robust world modeling, fundamentally transforming how vehicles interpret dynamic scenes and execute safe decision-making. World models have emerged as a linchpin technology, offering high-fidelity representations of the driving environment that integrate multi-sensor data, semantic cues, and temporal dynamics. This paper systematically reviews recent advances in world models for autonomous driving, proposing a three-tiered taxonomy: (i) Generation of Future Physical World, covering Image-, BEV-, OG-, and PC-based generation methods that enhance scene evolution modeling through diffusion models and 4D occupancy forecasting; (ii) Behavior Planning for Intelligent Agents, combining rule-driven and learning-based paradigms with cost map optimization and reinforcement learning for trajectory generation in complex traffic conditions; (ii) Interaction between Prediction and Planning, achieving multi-agent collaborative decision-making through latent space diffusion and memory-augmented architectures. The study further analyzes training paradigms, including self-supervised learning, multimodal pretraining, and generative data augmentation, while evaluating world models' performance in scene understanding and motion prediction tasks. Future research must address key challenges in self-supervised representation learning, long-tail scenario generation, and multimodal fusion to advance the practical deployment of world models in complex urban environments. Overall, the comprehensive analysis provides a technical roadmap for harnessing the transformative potential of world models in advancing safe and reliable autonomous driving solutions.
Longitudinal MRI analysis is crucial for predicting disease outcomes, particularly in chronic conditions like hepatocellular carcinoma (HCC), where early detection can significantly influence treatment strategies and patient prognosis. Yet, due to challenges like limited data availability, subtle parenchymal changes, and the irregular timing of medical screenings, current approaches have so far focused on cross-sectional imaging data. To address this, we propose HCCNet, a novel model architecture that integrates a 3D adaptation of the ConvNeXt CNN architecture with a Transformer encoder, capturing both the intricate spatial features of 3D MRIs and the complex temporal dependencies across different time points. HCCNet utilizes a two-stage pre-training process tailored for longitudinal MRI data. The CNN backbone is pre-trained using a self-supervised learning framework adapted for 3D MRIs, while the Transformer encoder is pre-trained with a sequence-order-prediction task to enhance its understanding of disease progression over time. We demonstrate the effectiveness of HCCNet by applying it to a cohort of liver cirrhosis patients undergoing regular MRI screenings for HCC surveillance. Our results show that HCCNet significantly improves predictive accuracy and reliability over baseline models, providing a robust tool for personalized HCC surveillance. The methodological approach presented in this paper is versatile and can be adapted to various longitudinal MRI screening applications. Its ability to handle varying patient record lengths and irregular screening intervals establishes it as an invaluable framework for monitoring chronic diseases, where timely and accurate disease prognosis is critical for effective treatment planning.
Current self-supervised learning methods for 3D medical imaging rely on simple pretext formulations and organ- or modality-specific datasets, limiting their generalizability and scalability. We present 3DINO, a cutting-edge SSL method adapted to 3D datasets, and use it to pretrain 3DINO-ViT: a general-purpose medical imaging model, on an exceptionally large, multimodal, and multi-organ dataset of ~100,000 3D medical imaging scans from over 10 organs. We validate 3DINO-ViT using extensive experiments on numerous medical imaging segmentation and classification tasks. Our results demonstrate that 3DINO-ViT generalizes across modalities and organs, including out-of-distribution tasks and datasets, outperforming state-of-the-art methods on the majority of evaluation metrics and labeled dataset sizes. Our 3DINO framework and 3DINO-ViT will be made available to enable research on 3D foundation models or further finetuning for a wide range of medical imaging applications.
Medical images acquired from standardized protocols show consistent macroscopic or microscopic anatomical structures, and these structures consist of composable/decomposable organs and tissues, but existing self-supervised learning (SSL) methods do not appreciate such composable/decomposable structure attributes inherent to medical images. To overcome this limitation, this paper introduces a novel SSL approach called ACE to learn anatomically consistent embedding via composition and decomposition with two key branches: (1) global consistency, capturing discriminative macro-structures via extracting global features; (2) local consistency, learning fine-grained anatomical details from composable/decomposable patch features via corresponding matrix matching. Experimental results across 6 datasets 2 backbones, evaluated in few-shot learning, fine-tuning, and property analysis, show ACE's superior robustness, transferability, and clinical potential. The innovations of our ACE lie in grid-wise image cropping, leveraging the intrinsic properties of compositionality and decompositionality of medical images, bridging the semantic gap from high-level pathologies to low-level tissue anomalies, and providing a new SSL method for medical imaging.
Foundation models have revolutionized computer vision by achieving vastly superior performance across diverse tasks through large-scale pretraining on extensive datasets. However, their application in surgical computer vision has been limited. This study addresses this gap by introducing SurgeNetXL, a novel surgical foundation model that sets a new benchmark in surgical computer vision. Trained on the largest reported surgical dataset to date, comprising over 4.7 million video frames, SurgeNetXL achieves consistent top-tier performance across six datasets spanning four surgical procedures and three tasks, including semantic segmentation, phase recognition, and critical view of safety (CVS) classification. Compared with the best-performing surgical foundation models, SurgeNetXL shows mean improvements of 2.4, 9.0, and 12.6 percent for semantic segmentation, phase recognition, and CVS classification, respectively. Additionally, SurgeNetXL outperforms the best-performing ImageNet-based variants by 14.4, 4.0, and 1.6 percent in the respective tasks. In addition to advancing model performance, this study provides key insights into scaling pretraining datasets, extending training durations, and optimizing model architectures specifically for surgical computer vision. These findings pave the way for improved generalizability and robustness in data-scarce scenarios, offering a comprehensive framework for future research in this domain. All models and a subset of the SurgeNetXL dataset, including over 2 million video frames, are publicly available at: https://github.com/TimJaspers0801/SurgeNet.
Semantic segmentation is essential for comprehending images, but the process necessitates a substantial amount of detailed annotations at the pixel level. Acquiring such annotations can be costly in the real-world. Unsupervised domain adaptation (UDA) for semantic segmentation is a technique that uses virtual data with labels to train a model and adapts it to real data without labels. Some recent works use contrastive learning, which is a powerful method for self-supervised learning, to help with this technique. However, these works do not take into account the diversity of features within each class when using contrastive learning, which leads to errors in class prediction. We analyze the limitations of these works and propose a novel framework called Pseudo-label Guided Pixel Contrast (PGPC), which overcomes the disadvantages of previous methods. We also investigate how to use more information from target images without adding noise from pseudo-labels. We test our method on two standard UDA benchmarks and show that it outperforms existing methods. Specifically, we achieve relative improvements of 5.1% mIoU and 4.6% mIoU on the Grand Theft Auto V (GTA5) to Cityscapes and SYNTHIA to Cityscapes tasks based on DAFormer, respectively. Furthermore, our approach can enhance the performance of other UDA approaches without increasing model complexity. Code is available at https://github.com/embar111/pgpc
Recent advancements have highlighted the efficacy of self-supervised learning (SSL) features in various speech-related tasks, providing lightweight and versatile multi-view speech representations. However, our study reveals that while SSL features expedite model convergence, they conflict with traditional spectral features like FBanks in terms of update directions. In response, we propose a novel generalized feature fusion framework grounded in conditional computation, featuring a gradient-sensitive gating network and a multi-stage dropout strategy. This framework mitigates feature conflicts and bolsters model robustness to multi-view input features. By integrating SSL and spectral features, our approach accelerates convergence and maintains performance on par with spectral models across multiple speech translation tasks on the MUSTC dataset.
Neural View Synthesis (NVS), such as NeRF and 3D Gaussian Splatting, effectively creates photorealistic scenes from sparse viewpoints, typically evaluated by quality assessment methods like PSNR, SSIM, and LPIPS. However, these full-reference methods, which compare synthesized views to reference views, may not fully capture the perceptual quality of neurally synthesized scenes (NSS), particularly due to the limited availability of dense reference views. Furthermore, the challenges in acquiring human perceptual labels hinder the creation of extensive labeled datasets, risking model overfitting and reduced generalizability. To address these issues, we propose NVS-SQA, a NSS quality assessment method to learn no-reference quality representations through self-supervision without reliance on human labels. Traditional self-supervised learning predominantly relies on the "same instance, similar representation" assumption and extensive datasets. However, given that these conditions do not apply in NSS quality assessment, we employ heuristic cues and quality scores as learning objectives, along with a specialized contrastive pair preparation process to improve the effectiveness and efficiency of learning. The results show that NVS-SQA outperforms 17 no-reference methods by a large margin (i.e., on average 109.5% in SRCC, 98.6% in PLCC, and 91.5% in KRCC over the second best) and even exceeds 16 full-reference methods across all evaluation metrics (i.e., 22.9% in SRCC, 19.1% in PLCC, and 18.6% in KRCC over the second best).