LiDAR point cloud semantic segmentation plays a crucial role in autonomous driving. In recent years, semi-supervised methods have gained popularity due to their significant reduction in annotation labor and time costs. Current semi-supervised methods typically focus on point cloud spatial distribution or consider short-term temporal representations, e.g., only two adjacent frames, often overlooking the rich long-term temporal properties inherent in autonomous driving scenarios. In driving experience, we observe that nearby objects, such as roads and vehicles, remain stable while driving, whereas distant objects exhibit greater variability in category and shape. This natural phenomenon is also captured by LiDAR, which reflects lower temporal sensitivity for nearby objects and higher sensitivity for distant ones. To leverage these characteristics, we propose HiLoTs, which learns high-temporal sensitivity and low-temporal sensitivity representations from continuous LiDAR frames. These representations are further enhanced and fused using a cross-attention mechanism. Additionally, we employ a teacher-student framework to align the representations learned by the labeled and unlabeled branches, effectively utilizing the large amounts of unlabeled data. Experimental results on the SemanticKITTI and nuScenes datasets demonstrate that our proposed HiLoTs outperforms state-of-the-art semi-supervised methods, and achieves performance close to LiDAR+Camera multimodal approaches. Code is available on https://github.com/rdlin118/HiLoTs
Recommender systems (RecSys) have become essential in modern society, driving user engagement and satisfaction across diverse online platforms. Most RecSys focuses on designing a powerful encoder to embed users and items into high-dimensional vector representation space, with loss functions optimizing their representation distributions. Recent studies reveal that directly optimizing key properties of the representation distribution, such as alignment and uniformity, can outperform complex encoder designs. However, existing methods for optimizing critical attributes overlook the impact of dataset sparsity on the model: limited user-item interactions lead to sparse alignment, while excessive interactions result in uneven uniformity, both of which degrade performance. In this paper, we identify the sparse alignment and uneven uniformity issues, and further propose Regularized Alignment and Uniformity (RAU) to cope with these two issues accordingly. RAU consists of two novel regularization methods for alignment and uniformity to learn better user/item representation. 1) Center-strengthened alignment further aligns the average in-batch user/item representation to provide an enhanced alignment signal and further minimize the disparity between user and item representation. 2) Low-variance-guided uniformity minimizes the variance of pairwise distances along with uniformity, which provides extra guidance to a more stabilized uniformity increase during training. We conducted extensive experiments on three real-world datasets, and the proposed RAU resulted in significant performance improvements compared to current state-of-the-art CF methods, which confirms the advantages of the two proposed regularization methods.
While deep learning models play a crucial role in predicting antibody-antigen interactions (AAI), the scarcity of publicly available sequence-structure pairings constrains their generalization. Current AAI methods often focus on residue-level static details, overlooking fine-grained structural representations of antibodies and their inter-antibody similarities. To tackle this challenge, we introduce a multi-modality representation approach that integates 3D structural and 1D sequence data to unravel intricate intra-antibody hierarchical relationships. By harnessing these representations, we present MuLAAIP, an AAI prediction framework that utilizes graph attention networks to illuminate graph-level structural features and normalized adaptive graph convolution networks to capture inter-antibody sequence associations. Furthermore, we have curated an AAI benchmark dataset comprising both structural and sequence information along with interaction labels. Through extensive experiments on this benchmark, our results demonstrate that MuLAAIP outperforms current state-of-the-art methods in terms of predictive performance. The implementation code and dataset are publicly available at https://github.com/trashTian/MuLAAIP for reproducibility.
The rapid growth of video content demands efficient and precise retrieval systems. While vision-language models (VLMs) excel in representation learning, they often struggle with adaptive, time-sensitive video retrieval. This paper introduces a novel framework that combines vector similarity search with graph-based data structures. By leveraging VLM embeddings for initial retrieval and modeling contextual relationships among video segments, our approach enables adaptive query refinement and improves retrieval accuracy. Experiments demonstrate its precision, scalability, and robustness, offering an effective solution for interactive video retrieval in dynamic environments.
Structure-informed protein representation learning is essential for effective protein function annotation and \textit{de novo} design. However, the presence of inherent noise in both crystal and AlphaFold-predicted structures poses significant challenges for existing methods in learning robust protein representations. To address these issues, we propose a novel equivariant Transformer-State Space Model(SSM) hybrid framework, termed $E^3$former, designed for efficient protein representation. Our approach uses energy function-based receptive fields to construct proximity graphs and incorporates an equivariant high-tensor-elastic selective SSM within the transformer architecture. These components enable the model to adapt to complex atom interactions and extract geometric features with higher signal-to-noise ratios. Empirical results demonstrate that our model outperforms existing methods in structure-intensive tasks, such as inverse folding and binding site prediction, particularly when using predicted structures, owing to its enhanced tolerance to data deviation and noise. Our approach offers a novel perspective for conducting biological function research and drug discovery using noisy protein structure data.
Nearly all identifiability results in unsupervised representation learning inspired by, e.g., independent component analysis, factor analysis, and causal representation learning, rely on assumptions of additive independent noise or noiseless regimes. In contrast, we study the more general case where noise can take arbitrary forms, depend on latent variables, and be non-invertibly entangled within a nonlinear function. We propose a general framework for identifying latent variables in the nonparametric noisy settings. We first show that, under suitable conditions, the generative model is identifiable up to certain submanifold indeterminacies even in the presence of non-negligible noise. Furthermore, under the structural or distributional variability conditions, we prove that latent variables of the general nonlinear models are identifiable up to trivial indeterminacies. Based on the proposed theoretical framework, we have also developed corresponding estimation methods and validated them in various synthetic and real-world settings. Interestingly, our estimate of the true GDP growth from alternative measurements suggests more insightful information on the economies than official reports. We expect our framework to provide new insight into how both researchers and practitioners deal with latent variables in real-world scenarios.
Proteins are complex biomolecules that play a central role in various biological processes, making them critical targets for breakthroughs in molecular biology, medical research, and drug discovery. Deciphering their intricate, hierarchical structures, and diverse functions is essential for advancing our understanding of life at the molecular level. Protein Representation Learning (PRL) has emerged as a transformative approach, enabling the extraction of meaningful computational representations from protein data to address these challenges. In this paper, we provide a comprehensive review of PRL research, categorizing methodologies into five key areas: feature-based, sequence-based, structure-based, multimodal, and complex-based approaches. To support researchers in this rapidly evolving field, we introduce widely used databases for protein sequences, structures, and functions, which serve as essential resources for model development and evaluation. We also explore the diverse applications of these approaches in multiple domains, demonstrating their broad impact. Finally, we discuss pressing technical challenges and outline future directions to advance PRL, offering insights to inspire continued innovation in this foundational field.
Left ventricular ejection fraction (LVEF) is a critical metric for assessing cardiac function, widely used in diagnosing heart failure and guiding clinical decisions. Despite its importance, conventional LVEF estimation remains time-consuming and operator-dependent. Recent deep learning advancements have enhanced automation, yet many existing models are computationally demanding, hindering their feasibility for real-time clinical applications. Additionally, the interplay between spatial and temporal features is crucial for accurate estimation but is often overlooked. In this work, we propose Echo-E$^3$Net, an efficient Endo-Epi spatio-temporal network tailored for LVEF estimation. Our method introduces the Endo-Epi Cardial Border Detector (E$^2$CBD) module, which enhances feature extraction by leveraging spatial and temporal landmark cues. Complementing this, the Endo-Epi Feature Aggregator (E$^2$FA) distills statistical descriptors from backbone feature maps, refining the final EF prediction. These modules, along with a multi-component loss function tailored to align with the clinical definition of EF, collectively enhance spatial-temporal representation learning, ensuring robust and efficient EF estimation. We evaluate Echo-E$^3$Net on the EchoNet-Dynamic dataset, achieving a RMSE of 5.15 and an R$^2$ score of 0.82, setting a new benchmark in efficiency with 6.8 million parameters and only 8.49G Flops. Our model operates without pre-training, data augmentation, or ensemble methods, making it well-suited for real-time point-of-care ultrasound (PoCUS) applications. Our Code is publicly available on~\href{https://github.com/moeinheidari7829/Echo-E3Net}{\textcolor{magenta}{GitHub}}.
Breast cancer survival prediction in computational pathology presents a remarkable challenge due to tumor heterogeneity. For instance, different regions of the same tumor in the pathology image can show distinct morphological and molecular characteristics. This makes it difficult to extract representative features from whole slide images (WSIs) that truly reflect the tumor's aggressive potential and likely survival outcomes. In this paper, we present PathoHR, a novel pipeline for accurate breast cancer survival prediction that enhances any size of pathological images to enable more effective feature learning. Our approach entails (1) the incorporation of a plug-and-play high-resolution Vision Transformer (ViT) to enhance patch-wise WSI representation, enabling more detailed and comprehensive feature extraction, (2) the systematic evaluation of multiple advanced similarity metrics for comparing WSI-extracted features, optimizing the representation learning process to better capture tumor characteristics, (3) the demonstration that smaller image patches enhanced follow the proposed pipeline can achieve equivalent or superior prediction accuracy compared to raw larger patches, while significantly reducing computational overhead. Experimental findings valid that PathoHR provides the potential way of integrating enhanced image resolution with optimized feature learning to advance computational pathology, offering a promising direction for more accurate and efficient breast cancer survival prediction. Code will be available at https://github.com/AIGeeksGroup/PathoHR.
The weights of neural networks (NNs) have recently gained prominence as a new data modality in machine learning, with applications ranging from accuracy and hyperparameter prediction to representation learning or weight generation. One approach to leverage NN weights involves training autoencoders (AEs), using contrastive and reconstruction losses. This allows such models to be applied to a wide variety of downstream tasks, and they demonstrate strong predictive performance and low reconstruction error. However, despite the low reconstruction error, these AEs reconstruct NN models with deteriorated performance compared to the original ones, limiting their usability with regard to model weight generation. In this paper, we identify a limitation of weight-space AEs, specifically highlighting that a structural loss, that uses the Euclidean distance between original and reconstructed weights, fails to capture some features critical for reconstructing high-performing models. We analyze the addition of a behavioral loss for training AEs in weight space, where we compare the output of the reconstructed model with that of the original one, given some common input. We show a strong synergy between structural and behavioral signals, leading to increased performance in all downstream tasks evaluated, in particular NN weights reconstruction and generation.
We study the problem of unsupervised representation learning in slightly misspecified settings, and thus formalize the study of robustness of nonlinear representation learning. We focus on the case where the mixing is close to a local isometry in a suitable distance and show based on existing rigidity results that the mixing can be identified up to linear transformations and small errors. In a second step, we investigate Independent Component Analysis (ICA) with observations generated according to $x=f(s)=As+h(s)$ where $A$ is an invertible mixing matrix and $h$ a small perturbation. We show that we can approximately recover the matrix $A$ and the independent components. Together, these two results show approximate identifiability of nonlinear ICA with almost isometric mixing functions. Those results are a step towards identifiability results for unsupervised representation learning for real-world data that do not follow restrictive model classes.
Network-wide Traffic State Estimation (TSE), which aims to infer a complete image of network traffic states with sparsely deployed sensors, plays a vital role in intelligent transportation systems. With the development of data-driven methods, traffic dynamics modeling has advanced significantly. However, TSE poses fundamental challenges for data-driven approaches, since historical patterns cannot be learned locally at sensor-free segments. Although inductive graph learning shows promise in estimating states at locations without sensor, existing methods typically handle unobserved locations by filling them with zeros, introducing bias to the sensitive graph message propagation. The recently proposed Dirichlet Energy-based Feature Propagation (DEFP) method achieves State-Of-The-Art (SOTA) performance in unobserved node classification by eliminating the need for zero-filling. However, applying it to TSE faces three key challenges: inability to handle directed traffic networks, strong assumptions in traffic spatial correlation modeling, and overlooks distinct propagation rules of different patterns (e.g., congestion and free flow). We propose DGAE, a novel inductive graph representation model that addresses these challenges through theoretically derived DEFP for Directed graph (DEFP4D), enhanced spatial representation learning via DEFP4D-guided latent space encoding, and physics-guided propagation mechanisms that separately handles congested and free-flow patterns. Experiments on three traffic datasets demonstrate that DGAE outperforms existing SOTA methods and exhibits strong cross-city transferability. Furthermore, DEFP4D can serve as a standalone lightweight solution, showing superior performance under extremely sparse sensor conditions.
Egocentric video-language pretraining has significantly advanced video representation learning. Humans perceive and interact with a fully 3D world, developing spatial awareness that extends beyond text-based understanding. However, most previous works learn from 1D text or 2D visual cues, such as bounding boxes, which inherently lack 3D understanding. To bridge this gap, we introduce EgoDTM, an Egocentric Depth- and Text-aware Model, jointly trained through large-scale 3D-aware video pretraining and video-text contrastive learning. EgoDTM incorporates a lightweight 3D-aware decoder to efficiently learn 3D-awareness from pseudo depth maps generated by depth estimation models. To further facilitate 3D-aware video pretraining, we enrich the original brief captions with hand-object visual cues by organically combining several foundation models. Extensive experiments demonstrate EgoDTM's superior performance across diverse downstream tasks, highlighting its superior 3D-aware visual understanding. Our code will be released at https://github.com/xuboshen/EgoDTM.
LiDAR-based Vehicle-to-Everything (V2X) cooperative perception has demonstrated its impact on the safety and effectiveness of autonomous driving. Since current cooperative perception algorithms are trained and tested on the same dataset, the generalization ability of cooperative perception systems remains underexplored. This paper is the first work to study the Domain Generalization problem of LiDAR-based V2X cooperative perception (V2X-DG) for 3D detection based on four widely-used open source datasets: OPV2V, V2XSet, V2V4Real and DAIR-V2X. Our research seeks to sustain high performance not only within the source domain but also across other unseen domains, achieved solely through training on source domain. To this end, we propose Cooperative Mixup Augmentation based Generalization (CMAG) to improve the model generalization capability by simulating the unseen cooperation, which is designed compactly for the domain gaps in cooperative perception. Furthermore, we propose a constraint for the regularization of the robust generalized feature representation learning: Cooperation Feature Consistency (CFC), which aligns the intermediately fused features of the generalized cooperation by CMAG and the early fused features of the original cooperation in source domain. Extensive experiments demonstrate that our approach achieves significant performance gains when generalizing to other unseen datasets while it also maintains strong performance on the source dataset.
The past decade has witnessed notable achievements in self-supervised learning for video tasks. Recent efforts typically adopt the Masked Video Modeling (MVM) paradigm, leading to significant progress on multiple video tasks. However, two critical challenges remain: 1) Without human annotations, the random temporal sampling introduces uncertainty, increasing the difficulty of model training. 2) Previous MVM methods primarily recover the masked patches in the pixel space, leading to insufficient information compression for downstream tasks. To address these challenges jointly, we propose a self-supervised framework that leverages Temporal Correspondence for video Representation learning (T-CoRe). For challenge 1), we propose a sandwich sampling strategy that selects two auxiliary frames to reduce reconstruction uncertainty in a two-side-squeezing manner. Addressing challenge 2), we introduce an auxiliary branch into a self-distillation architecture to restore representations in the latent space, generating high-level semantic representations enriched with temporal information. Experiments of T-CoRe consistently present superior performance across several downstream tasks, demonstrating its effectiveness for video representation learning. The code is available at https://github.com/yafeng19/T-CORE.
Deep learning-based medical image segmentation has shown remarkable success; however, it typically requires extensive pixel-level annotations, which are both expensive and time-intensive. Semi-supervised medical image segmentation (SSMIS) offers a viable alternative, driven by advancements in CNNs and ViTs. However, these networks often rely on single fixed activation functions and linear modeling patterns, limiting their ability to effectively learn robust representations. Given the limited availability of labeled date, achieving robust representation learning becomes crucial. Inspired by Kolmogorov-Arnold Networks (KANs), we propose Semi-KAN, which leverages the untapped potential of KANs to enhance backbone architectures for representation learning in SSMIS. Our findings indicate that: (1) compared to networks with fixed activation functions, KANs exhibit superior representation learning capabilities with fewer parameters, and (2) KANs excel in high-semantic feature spaces. Building on these insights, we integrate KANs into tokenized intermediate representations, applying them selectively at the encoder's bottleneck and the decoder's top layers within a U-Net pipeline to extract high-level semantic features. Although learnable activation functions improve feature expansion, they introduce significant computational overhead with only marginal performance gains. To mitigate this, we reduce the feature dimensions and employ horizontal scaling to capture multiple pattern representations. Furthermore, we design a multi-branch U-Net architecture with uncertainty estimation to effectively learn diverse pattern representations. Extensive experiments on four public datasets demonstrate that Semi-KAN surpasses baseline networks, utilizing fewer KAN layers and lower computational cost, thereby underscoring the potential of KANs as a promising approach for SSMIS.
Multimodal Large Language Models (MLLMs) have achieved remarkable success in vision-language tasks but their remote sensing (RS) counterpart are relatively under explored. Unlike natural images, RS imagery presents unique challenges that current MLLMs struggle to handle, particularly in visual grounding and spatial reasoning. This study investigates the limitations of CLIP-based MLLMs in RS, highlighting their failure to differentiate visually distinct yet semantically similar RS images. To address this, we introduce a remote sensing multimodal visual patterns (RSMMVP) benchmark. It is designed to evaluate MLLMs in RS tasks by identifying the CLIP-blind pairs, where CLIP-based models incorrectly assign high similarity scores to visually distinct RS images. Through a visual question answering (VQA) evaluation, we analyze the performance of state-of-the-art MLLMs, revealing significant limitations in RS specific representation learning. The results provide valuable insights into the weaknesses of CLIP-based visual encoding and offer a foundation for future research to develop more effective MLLMs tailored for remote sensing applications.
The widespread use of deep learning classifiers necessitates Open-set recognition (OSR), which enables the identification of input data not only from classes known during training but also from unknown classes that might be present in test data. Many existing OSR methods are computationally expensive due to the reliance on complex generative models or suffer from high training costs. We investigate OSR from a representation-learning perspective, specifically through spherical embeddings. We introduce SphOR, a computationally efficient representation learning method that models the feature space as a mixture of von Mises-Fisher distributions. This approach enables the use of semantically ambiguous samples during training, to improve the detection of samples from unknown classes. We further explore the relationship between OSR performance and key representation learning properties which influence how well features are structured in high-dimensional space. Extensive experiments on multiple OSR benchmarks demonstrate the effectiveness of our method, producing state-of-the-art results, with improvements up-to 6% that validate its performance. Code at https://github.com/nadarasarbahavan/SpHOR
Graph-based collaborative filtering has been established as a prominent approach in recommendation systems, leveraging the inherent graph topology of user-item interactions to model high-order connectivity patterns and enhance recommendation performance. Recent advances in Graph Contrastive Learning (GCL) have demonstrated promising potential to alleviate data sparsity issues by improving representation learning through contrastive view generation and mutual information maximization. However, existing approaches lack effective data augmentation strategies. Structural augmentation risks distorting fundamental graph topology, while feature-level perturbation techniques predominantly employ uniform noise scales that fail to account for node-specific characteristics. To solve these challenges, we propose Diffusion-augmented Contrastive Learning (DGCL), an innovative framework that integrates diffusion models with contrastive learning for enhanced collaborative filtering. Our approach employs a diffusion process that learns node-specific Gaussian distributions of representations, thereby generating semantically consistent yet diversified contrastive views through reverse diffusion sampling. DGCL facilitates adaptive data augmentation based on reconstructed representations, considering both semantic coherence and node-specific features. In addition, it explores unrepresented regions of the latent sparse feature space, thereby enriching the diversity of contrastive views. Extensive experimental results demonstrate the effectiveness of DGCL on three public datasets.
Multimodal alignment constructs a joint latent vector space where modalities representing the same concept map to neighboring latent vectors. We formulate this as an inverse problem and show that, under certain conditions, paired data from each modality can map to equivalent latent vectors, which we refer to as perfect alignment. When perfect alignment cannot be achieved, it can be approximated using the Singular Value Decomposition (SVD) of a multimodal data matrix. Experiments on synthetic multimodal Gaussian data verify the effectiveness of our perfect alignment method compared to a learned contrastive alignment method. We further demonstrate the practical application of cross-modal transfer for human action recognition, showing that perfect alignment significantly enhances the model's accuracy. We conclude by discussing how these findings can be applied to various modalities and tasks and the limitations of our method. We hope these findings inspire further exploration of perfect alignment and its applications in representation learning.