Remote work and online courses have become important methods of knowledge dissemination, leading to a large number of document-based instructional videos. Unlike traditional video datasets, these videos mainly feature rich-text images and audio that are densely packed with information closely tied to the visual content, requiring advanced multimodal understanding capabilities. However, this domain remains underexplored due to dataset availability and its inherent complexity. In this paper, we introduce the DocVideoQA task and dataset for the first time, comprising 1454 videos across 23 categories with a total duration of about 828 hours. The dataset is annotated with 154k question-answer pairs generated manually and via GPT, assessing models' comprehension, temporal awareness, and modality integration capabilities. Initially, we establish a baseline using open-source MLLMs. Recognizing the challenges in modality comprehension for document-centric videos, we present DV-LLaMA, a robust video MLLM baseline. Our method enhances unimodal feature extraction with diverse instruction-tuning data and employs contrastive learning to strengthen modality integration. Through fine-tuning, the LLM is equipped with audio-visual capabilities, leading to significant improvements in document-centric video understanding. Extensive testing on the DocVideoQA dataset shows that DV-LLaMA significantly outperforms existing models. We'll release the code and dataset to facilitate future research.
Most existing graph-based semi-supervised hyperspectral image classification methods rely on superpixel partitioning techniques. However, they suffer from misclassification of certain pixels due to inaccuracies in superpixel boundaries, \ie, the initial inaccuracies in superpixel partitioning limit overall classification performance. In this paper, we propose a novel graph-weighted contrastive learning approach that avoids the use of superpixel partitioning and directly employs neural networks to learn hyperspectral image representation. Furthermore, while many approaches require all graph nodes to be available during training, our approach supports mini-batch training by processing only a subset of nodes at a time, reducing computational complexity and improving generalization to unseen nodes. Experimental results on three widely-used datasets demonstrate the effectiveness of the proposed approach compared to baselines relying on superpixel partitioning.
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.
While recent advancements in generative modeling have significantly improved text-image alignment, some residual misalignment between text and image representations still remains. Although many approaches have attempted to address this issue by fine-tuning models using various reward models, etc., we revisit the challenge from the perspective of representation alignment-an approach that has gained popularity with the success of REPresentation Alignment (REPA). We first argue that conventional text-to-image (T2I) diffusion models, typically trained on paired image and text data (i.e., positive pairs) by minimizing score matching or flow matching losses, is suboptimal from the standpoint of representation alignment. Instead, a better alignment can be achieved through contrastive learning that leverages both positive and negative pairs. To achieve this efficiently even with pretrained models, we introduce a lightweight contrastive fine tuning strategy called SoftREPA that uses soft text tokens. This approach improves alignment with minimal computational overhead by adding fewer than 1M trainable parameters to the pretrained model. Our theoretical analysis demonstrates that our method explicitly increases the mutual information between text and image representations, leading to enhanced semantic consistency. Experimental results across text-to-image generation and text-guided image editing tasks validate the effectiveness of our approach in improving the semantic consistency of T2I generative models.
Vision-language models for Earth observation (EO) typically rely on the visual spectrum of data as the only model input, thus failing to leverage the rich spectral information available in the multispectral channels recorded by satellites. Therefore, in this paper, we introduce Llama3-MS-CLIP, the first vision-language model pre-trained with contrastive learning on a large-scale multispectral dataset and report on the performance gains due to the extended spectral range. Furthermore, we present the largest-to-date image-caption dataset for multispectral data, consisting of one million Sentinel-2 samples and corresponding textual descriptions generated with Llama3-LLaVA-Next and Overture Maps data. We develop a scalable captioning pipeline, which is validated by domain experts. We evaluate Llama3-MS-CLIP on multispectral zero-shot image classification and retrieval using three datasets of varying complexity. Our results demonstrate that Llama3-MS-CLIP significantly outperforms other RGB-based approaches, improving classification accuracy by 6.77% on average and retrieval performance by 4.63% mAP compared to the second-best model. Our results emphasize the relevance of multispectral vision-language learning. We release the image-caption dataset, code, and model weights under an open-source license.
Precise audio-visual synchronization in speech videos is crucial for content quality and viewer comprehension. Existing methods have made significant strides in addressing this challenge through rule-based approaches and end-to-end learning techniques. However, these methods often rely on limited audio-visual representations and suboptimal learning strategies, potentially constraining their effectiveness in more complex scenarios. To address these limitations, we present UniSync, a novel approach for evaluating audio-visual synchronization using embedding similarities. UniSync offers broad compatibility with various audio representations (e.g., Mel spectrograms, HuBERT) and visual representations (e.g., RGB images, face parsing maps, facial landmarks, 3DMM), effectively handling their significant dimensional differences. We enhance the contrastive learning framework with a margin-based loss component and cross-speaker unsynchronized pairs, improving discriminative capabilities. UniSync outperforms existing methods on standard datasets and demonstrates versatility across diverse audio-visual representations. Its integration into talking face generation frameworks enhances synchronization quality in both natural and AI-generated content.
Machine unlearning methods have become increasingly important for selective concept removal in large pre-trained models. While recent work has explored unlearning in Euclidean contrastive vision-language models, the effectiveness of concept removal in hyperbolic spaces remains unexplored. This paper investigates machine unlearning in hyperbolic contrastive learning by adapting Alignment Calibration to MERU, a model that embeds images and text in hyperbolic space to better capture semantic hierarchies. Through systematic experiments and ablation studies, we demonstrate that hyperbolic geometry offers distinct advantages for concept removal, achieving near perfect forgetting with reasonable performance on retained concepts, particularly when scaling to multiple concept removal. Our approach introduces hyperbolic-specific components including entailment calibration and norm regularization that leverage the unique properties of hyperbolic space. Comparative analysis with Euclidean models reveals fundamental differences in unlearning dynamics, with hyperbolic unlearning reorganizing the semantic hierarchy while Euclidean approaches merely disconnect cross-modal associations. These findings not only advance machine unlearning techniques but also provide insights into the geometric properties that influence concept representation and removal in multimodal models. Source code available at https://github.com/alex-pv01/HAC
Cross-modal hashing is a promising approach for efficient data retrieval and storage optimization. However, contemporary methods exhibit significant limitations in semantic preservation, contextual integrity, and information redundancy, which constrains retrieval efficacy. We present PromptHash, an innovative framework leveraging affinity prompt-aware collaborative learning for adaptive cross-modal hashing. We propose an end-to-end framework for affinity-prompted collaborative hashing, with the following fundamental technical contributions: (i) a text affinity prompt learning mechanism that preserves contextual information while maintaining parameter efficiency, (ii) an adaptive gated selection fusion architecture that synthesizes State Space Model with Transformer network for precise cross-modal feature integration, and (iii) a prompt affinity alignment strategy that bridges modal heterogeneity through hierarchical contrastive learning. To the best of our knowledge, this study presents the first investigation into affinity prompt awareness within collaborative cross-modal adaptive hash learning, establishing a paradigm for enhanced semantic consistency across modalities. Through comprehensive evaluation on three benchmark multi-label datasets, PromptHash demonstrates substantial performance improvements over existing approaches. Notably, on the NUS-WIDE dataset, our method achieves significant gains of 18.22% and 18.65% in image-to-text and text-to-image retrieval tasks, respectively. The code is publicly available at https://github.com/ShiShuMo/PromptHash.
Skeleton-based Temporal Action Segmentation (STAS) aims to segment and recognize various actions from long, untrimmed sequences of human skeletal movements. Current STAS methods typically employ spatio-temporal modeling to establish dependencies among joints as well as frames, and utilize one-hot encoding with cross-entropy loss for frame-wise classification supervision. However, these methods overlook the intrinsic correlations among joints and actions within skeletal features, leading to a limited understanding of human movements. To address this, we propose a Text-Derived Relational Graph-Enhanced Network (TRG-Net) that leverages prior graphs generated by Large Language Models (LLM) to enhance both modeling and supervision. For modeling, the Dynamic Spatio-Temporal Fusion Modeling (DSFM) method incorporates Text-Derived Joint Graphs (TJG) with channel- and frame-level dynamic adaptation to effectively model spatial relations, while integrating spatio-temporal core features during temporal modeling. For supervision, the Absolute-Relative Inter-Class Supervision (ARIS) method employs contrastive learning between action features and text embeddings to regularize the absolute class distributions, and utilizes Text-Derived Action Graphs (TAG) to capture the relative inter-class relationships among action features. Additionally, we propose a Spatial-Aware Enhancement Processing (SAEP) method, which incorporates random joint occlusion and axial rotation to enhance spatial generalization. Performance evaluations on four public datasets demonstrate that TRG-Net achieves state-of-the-art results.
Heterogeneous graphs (HGs) are composed of multiple types of nodes and edges, making it more effective in capturing the complex relational structures inherent in the real world. However, in real-world scenarios, labeled data is often difficult to obtain, which limits the applicability of semi-supervised approaches. Self-supervised learning aims to enable models to automatically learn useful features from data, effectively addressing the challenge of limited labeling data. In this paper, we propose a novel contrastive learning framework for heterogeneous graphs (ASHGCL), which incorporates three distinct views, each focusing on node attributes, high-order and low-order structural information, respectively, to effectively capture attribute information, high-order structures, and low-order structures for node representation learning. Furthermore, we introduce an attribute-enhanced positive sample selection strategy that combines both structural information and attribute information, effectively addressing the issue of sampling bias. Extensive experiments on four real-world datasets show that ASHGCL outperforms state-of-the-art unsupervised baselines and even surpasses some supervised benchmarks.
The differing representation spaces required for visual understanding and generation pose a challenge in unifying them within the autoregressive paradigm of large language models. A vision tokenizer trained for reconstruction excels at capturing low-level perceptual details, making it well-suited for visual generation but lacking high-level semantic representations for understanding tasks. Conversely, a vision encoder trained via contrastive learning aligns well with language but struggles to decode back into the pixel space for generation tasks. To bridge this gap, we propose DualToken, a method that unifies representations for both understanding and generation within a single tokenizer. However, directly integrating reconstruction and semantic objectives in a single tokenizer creates conflicts, leading to degraded performance in both reconstruction quality and semantic performance. Instead of forcing a single codebook to handle both semantic and perceptual information, DualToken disentangles them by introducing separate codebooks for high and low-level features, effectively transforming their inherent conflict into a synergistic relationship. As a result, DualToken achieves state-of-the-art performance in both reconstruction and semantic tasks while demonstrating remarkable effectiveness in downstream MLLM understanding and generation tasks. Notably, we also show that DualToken, as a unified tokenizer, surpasses the naive combination of two distinct types vision encoders, providing superior performance within a unified MLLM.
We address the problem of semi-supervised domain generalization (SSDG), where the distributions of train and test data differ, and only a small amount of labeled data along with a larger amount of unlabeled data are available during training. Existing SSDG methods that leverage only the unlabeled samples for which the model's predictions are highly confident (confident-unlabeled samples), limit the full utilization of the available unlabeled data. To the best of our knowledge, we are the first to explore a method for incorporating the unconfident-unlabeled samples that were previously disregarded in SSDG setting. To this end, we propose UPCSC to utilize these unconfident-unlabeled samples in SSDG that consists of two modules: 1) Unlabeled Proxy-based Contrastive learning (UPC) module, treating unconfident-unlabeled samples as additional negative pairs and 2) Surrogate Class learning (SC) module, generating positive pairs for unconfident-unlabeled samples using their confusing class set. These modules are plug-and-play and do not require any domain labels, which can be easily integrated into existing approaches. Experiments on four widely used SSDG benchmarks demonstrate that our approach consistently improves performance when attached to baselines and outperforms competing plug-and-play methods. We also analyze the role of our method in SSDG, showing that it enhances class-level discriminability and mitigates domain gaps. The code is available at https://github.com/dongkwani/UPCSC.
Multimodal learning that integrates histopathology images and genomic data holds great promise for cancer survival prediction. However, existing methods face key limitations: 1) They rely on multimodal mapping and metrics in Euclidean space, which cannot fully capture the hierarchical structures in histopathology (among patches from different resolutions) and genomics data (from genes to pathways). 2) They discretize survival time into independent risk intervals, which ignores its continuous and ordinal nature and fails to achieve effective optimization. 3) They treat censorship as a binary indicator, excluding censored samples from model optimization and not making full use of them. To address these challenges, we propose HySurvPred, a novel framework for survival prediction that integrates three key modules: Multimodal Hyperbolic Mapping (MHM), Angle-aware Ranking-based Contrastive Loss (ARCL) and Censor-Conditioned Uncertainty Constraint (CUC). Instead of relying on Euclidean space, we design the MHM module to explore the inherent hierarchical structures within each modality in hyperbolic space. To better integrate multimodal features in hyperbolic space, we introduce the ARCL module, which uses ranking-based contrastive learning to preserve the ordinal nature of survival time, along with the CUC module to fully explore the censored data. Extensive experiments demonstrate that our method outperforms state-of-the-art methods on five benchmark datasets. The source code is to be released.
World models significantly enhance hierarchical understanding, improving data integration and learning efficiency. To explore the potential of the world model in the remote sensing (RS) field, this paper proposes a label-efficient remote sensing world model for multimodal data fusion (FusDreamer). The FusDreamer uses the world model as a unified representation container to abstract common and high-level knowledge, promoting interactions across different types of data, \emph{i.e.}, hyperspectral (HSI), light detection and ranging (LiDAR), and text data. Initially, a new latent diffusion fusion and multimodal generation paradigm (LaMG) is utilized for its exceptional information integration and detail retention capabilities. Subsequently, an open-world knowledge-guided consistency projection (OK-CP) module incorporates prompt representations for visually described objects and aligns language-visual features through contrastive learning. In this way, the domain gap can be bridged by fine-tuning the pre-trained world models with limited samples. Finally, an end-to-end multitask combinatorial optimization (MuCO) strategy can capture slight feature bias and constrain the diffusion process in a collaboratively learnable direction. Experiments conducted on four typical datasets indicate the effectiveness and advantages of the proposed FusDreamer. The corresponding code will be released at https://github.com/Cimy-wang/FusDreamer.
Accurately localizing audible objects based on audio-visual cues is the core objective of audio-visual segmentation. Most previous methods emphasize spatial or temporal multi-modal modeling, yet overlook challenges from ambiguous audio-visual correspondences such as nearby visually similar but acoustically different objects and frequent shifts in objects' sounding status. Consequently, they may struggle to reliably correlate audio and visual cues, leading to over- or under-segmentation. To address these limitations, we propose a novel framework with two primary components: an audio-guided modality alignment (AMA) module and an uncertainty estimation (UE) module. Instead of indiscriminately correlating audio-visual cues through a global attention mechanism, AMA performs audio-visual interactions within multiple groups and consolidates group features into compact representations based on their responsiveness to audio cues, effectively directing the model's attention to audio-relevant areas. Leveraging contrastive learning, AMA further distinguishes sounding regions from silent areas by treating features with strong audio responses as positive samples and weaker responses as negatives. Additionally, UE integrates spatial and temporal information to identify high-uncertainty regions caused by frequent changes in sound state, reducing prediction errors by lowering confidence in these areas. Experimental results demonstrate that our approach achieves superior accuracy compared to existing state-of-the-art methods, particularly in challenging scenarios where traditional approaches struggle to maintain reliable segmentation.
Recommendation systems have found extensive applications across diverse domains. However, the training data available typically comprises implicit feedback, manifested as user clicks and purchase behaviors, rather than explicit declarations of user preferences. This type of training data presents three main challenges for accurate ranking prediction: First, the unobservable nature of user preferences makes likelihood function modeling inherently difficult. Second, the resulting false positives (FP) and false negatives (FN) introduce noise into the learning process, disrupting parameter learning. Third, data bias arises as observed interactions tend to concentrate on a few popular items, exacerbating the feedback loop of popularity bias. To address these issues, we propose Variational BPR, a novel and easily implementable learning objective that integrates key components for enhancing collaborative filtering: likelihood optimization, noise reduction, and popularity debiasing. Our approach involves decomposing the pairwise loss under the ELBO-KL framework and deriving its variational lower bound to establish a manageable learning objective for approximate inference. Within this bound, we introduce an attention-based latent interest prototype contrastive mechanism, replacing instance-level contrastive learning, to effectively reduce noise from problematic samples. The process of deriving interest prototypes implicitly incorporates a flexible hard sample mining strategy, capable of simultaneously identifying hard positive and hard negative samples. Furthermore, we demonstrate that this hard sample mining strategy promotes feature distribution uniformity, thereby alleviating popularity bias. Empirically, we demonstrate the effectiveness of Variational BPR on popular backbone recommendation models. The code and data are available at: https://github.com/liubin06/VariationalBPR
We present V$^2$Dial - a novel expert-based model specifically geared towards simultaneously handling image and video input data for multimodal conversational tasks. Current multimodal models primarily focus on simpler tasks (e.g., VQA, VideoQA, video-text retrieval) and often neglect the more challenging conversational counterparts, such as video and visual/image dialog. Moreover, works on both conversational tasks evolved separately from each other despite their apparent similarities limiting their applicability potential. To this end, we propose to unify both tasks using a single model that for the first time jointly learns the spatial and temporal features of images and videos by routing them through dedicated experts and aligns them using matching and contrastive learning techniques. Furthermore, we systemically study the domain shift between the two tasks by investigating whether and to what extent these seemingly related tasks can mutually benefit from their respective training data. Extensive evaluations on the widely used video and visual dialog datasets of AVSD and VisDial show that our model achieves new state-of-the-art results across four benchmarks both in zero-shot and fine-tuning settings.
We present a framework for pre-training of 3D hand pose estimation from in-the-wild hand images sharing with similar hand characteristics, dubbed SimHand. Pre-training with large-scale images achieves promising results in various tasks, but prior methods for 3D hand pose pre-training have not fully utilized the potential of diverse hand images accessible from in-the-wild videos. To facilitate scalable pre-training, we first prepare an extensive pool of hand images from in-the-wild videos and design our pre-training method with contrastive learning. Specifically, we collect over 2.0M hand images from recent human-centric videos, such as 100DOH and Ego4D. To extract discriminative information from these images, we focus on the similarity of hands: pairs of non-identical samples with similar hand poses. We then propose a novel contrastive learning method that embeds similar hand pairs closer in the feature space. Our method not only learns from similar samples but also adaptively weights the contrastive learning loss based on inter-sample distance, leading to additional performance gains. Our experiments demonstrate that our method outperforms conventional contrastive learning approaches that produce positive pairs sorely from a single image with data augmentation. We achieve significant improvements over the state-of-the-art method (PeCLR) in various datasets, with gains of 15% on FreiHand, 10% on DexYCB, and 4% on AssemblyHands. Our code is available at https://github.com/ut-vision/SiMHand.
Self-supervised learning is a machine learning approach that generates implicit labels by learning underlined patterns and extracting discriminative features from unlabeled data without manual labelling. Contrastive learning introduces the concept of "positive" and "negative" samples, where positive pairs (e.g., variation of the same image/object) are brought together in the embedding space, and negative pairs (e.g., views from different images/objects) are pushed farther away. This methodology has shown significant improvements in image understanding and image text analysis without much reliance on labeled data. In this paper, we comprehensively discuss the terminologies, recent developments and applications of contrastive learning with respect to text-image models. Specifically, we provide an overview of the approaches of contrastive learning in text-image models in recent years. Secondly, we categorize the approaches based on different model structures. Thirdly, we further introduce and discuss the latest advances of the techniques used in the process such as pretext tasks for both images and text, architectural structures, and key trends. Lastly, we discuss the recent state-of-art applications of self-supervised contrastive learning Text-Image based models.
We present OuroMamba, the first data-free post-training quantization (DFQ) method for vision Mamba-based models (VMMs). We identify two key challenges in enabling DFQ for VMMs, (1) VMM's recurrent state transitions restricts capturing of long-range interactions and leads to semantically weak synthetic data, (2) VMM activations exhibit dynamic outlier variations across time-steps, rendering existing static PTQ techniques ineffective. To address these challenges, OuroMamba presents a two-stage framework: (1) OuroMamba-Gen to generate semantically rich and meaningful synthetic data. It applies contrastive learning on patch level VMM features generated through neighborhood interactions in the latent state space, (2) OuroMamba-Quant to employ mixed-precision quantization with lightweight dynamic outlier detection during inference. In specific, we present a thresholding based outlier channel selection strategy for activations that gets updated every time-step. Extensive experiments across vision and generative tasks show that our data-free OuroMamba surpasses existing data-driven PTQ techniques, achieving state-of-the-art performance across diverse quantization settings. Additionally, we implement efficient GPU kernels to achieve practical latency speedup of up to 2.36x. Code will be released soon.