Rebalanced Multimodal Learning with Data-aware Unimodal Sampling

To address the modality learning degeneration caused by modality imbalance, existing multimodal learning~(MML) approaches primarily attempt to balance the optimization process of each modality from the perspective of model learning. However, almost all existing methods ignore the modality imbalance caused by unimodal data sampling, i.e., equal unimodal data sampling often results in discrepancies in informational content, leading to modality imbalance. Therefore, in this paper, we propose a novel MML approach called \underline{D}ata-aware \underline{U}nimodal \underline{S}ampling~(\method), which aims to dynamically alleviate the modality imbalance caused by sampling. Specifically, we first propose a novel cumulative modality discrepancy to monitor the multimodal learning process. Based on the learning status, we propose a heuristic and a reinforcement learning~(RL)-based data-aware unimodal sampling approaches to adaptively determine the quantity of sampled data at each iteration, thus alleviating the modality imbalance from the perspective of sampling. Meanwhile, our method can be seamlessly incorporated into almost all existing multimodal learning approaches as a plugin. Experiments demonstrate that \method~can achieve the best performance by comparing with diverse state-of-the-art~(SOTA) baselines.

Long-Term TalkingFace Generation via Motion-Prior Conditional Diffusion Model

Recent advances in conditional diffusion models have shown promise for generating realistic TalkingFace videos, yet challenges persist in achieving consistent head movement, synchronized facial expressions, and accurate lip synchronization over extended generations. To address these, we introduce the \textbf{M}otion-priors \textbf{C}onditional \textbf{D}iffusion \textbf{M}odel (\textbf{MCDM}), which utilizes both archived and current clip motion priors to enhance motion prediction and ensure temporal consistency. The model consists of three key elements: (1) an archived-clip motion-prior that incorporates historical frames and a reference frame to preserve identity and context; (2) a present-clip motion-prior diffusion model that captures multimodal causality for accurate predictions of head movements, lip sync, and expressions; and (3) a memory-efficient temporal attention mechanism that mitigates error accumulation by dynamically storing and updating motion features. We also release the \textbf{TalkingFace-Wild} dataset, a multilingual collection of over 200 hours of footage across 10 languages. Experimental results demonstrate the effectiveness of MCDM in maintaining identity and motion continuity for long-term TalkingFace generation. Code, models, and datasets will be publicly available.

Vision-centric Token Compression in Large Language Model

Large Language Models (LLMs) have revolutionized natural language processing, excelling in handling longer sequences. However, the inefficiency and redundancy in processing extended in-context tokens remain a challenge. Many attempts to address this rely on compressing tokens with smaller text encoders, yet we question whether text encoders are truly indispensable. Our journey leads to an unexpected discovery-a much smaller vision encoder, applied directly to sequences of text tokens, can rival text encoders on text tasks. When pre-trained on large amounts of data and transferred to multiple mid-sized or small text understanding benchmarks, VIST leads to comparable results with 16% fewer FLOPs and 50% less memory usage. We further uncover significant token redundancy and devise a frequency-based masking strategy to guide the focus of the visual encoder toward the most critical tokens. Interestingly, we observe the trained visual encoder performs like a summarizer, selectively ignoring less important words such as prepositions and conjunctions. This approach delivers remarkable results, outperforming traditional text encoder-based methods by 5.7% on average over benchmarks like TriviaQA, NQ, PopQA, TREF, SST2, and SST5, setting a new standard for token efficiency in LLMs.

Merging Context Clustering with Visual State Space Models for Medical Image Segmentation

Medical image segmentation demands the aggregation of global and local feature representations, posing a challenge for current methodologies in handling both long-range and short-range feature interactions. Recently, vision mamba (ViM) models have emerged as promising solutions for addressing model complexities by excelling in long-range feature iterations with linear complexity. However, existing ViM approaches overlook the importance of preserving short-range local dependencies by directly flattening spatial tokens and are constrained by fixed scanning patterns that limit the capture of dynamic spatial context information. To address these challenges, we introduce a simple yet effective method named context clustering ViM (CCViM), which incorporates a context clustering module within the existing ViM models to segment image tokens into distinct windows for adaptable local clustering. Our method effectively combines long-range and short-range feature interactions, thereby enhancing spatial contextual representations for medical image segmentation tasks. Extensive experimental evaluations on diverse public datasets, i.e., Kumar, CPM17, ISIC17, ISIC18, and Synapse demonstrate the superior performance of our method compared to current state-of-the-art methods. Our code can be found at https://github.com/zymissy/CCViM.

Rebalanced Vision-Language Retrieval Considering Structure-Aware Distillation

Vision-language retrieval aims to search for similar instances in one modality based on queries from another modality. The primary objective is to learn cross-modal matching representations in a latent common space. Actually, the assumption underlying cross-modal matching is modal balance, where each modality contains sufficient information to represent the others. However, noise interference and modality insufficiency often lead to modal imbalance, making it a common phenomenon in practice. The impact of imbalance on retrieval performance remains an open question. In this paper, we first demonstrate that ultimate cross-modal matching is generally sub-optimal for cross-modal retrieval when imbalanced modalities exist. The structure of instances in the common space is inherently influenced when facing imbalanced modalities, posing a challenge to cross-modal similarity measurement. To address this issue, we emphasize the importance of meaningful structure-preserved matching. Accordingly, we propose a simple yet effective method to rebalance cross-modal matching by learning structure-preserved matching representations. Specifically, we design a novel multi-granularity cross-modal matching that incorporates structure-aware distillation alongside the cross-modal matching loss. While the cross-modal matching loss constraints instance-level matching, the structure-aware distillation further regularizes the geometric consistency between learned matching representations and intra-modal representations through the developed relational matching. Extensive experiments on different datasets affirm the superior cross-modal retrieval performance of our approach, simultaneously enhancing single-modal retrieval capabilities compared to the baseline models.

ExpRDiff: Short-exposure Guided Diffusion Model for Realistic Local Motion Deblurring

Removing blur caused by moving objects is challenging, as the moving objects are usually significantly blurry while the static background remains clear. Existing methods that rely on local blur detection often suffer from inaccuracies and cannot generate satisfactory results when focusing solely on blurred regions. To overcome these problems, we first design a context-based local blur detection module that incorporates additional contextual information to improve the identification of blurry regions. Considering that modern smartphones are equipped with cameras capable of providing short-exposure images, we develop a blur-aware guided image restoration method that utilizes sharp structural details from short-exposure images, facilitating accurate reconstruction of heavily blurred regions. Furthermore, to restore images realistically and visually-pleasant, we develop a short-exposure guided diffusion model that explores useful features from short-exposure images and blurred regions to better constrain the diffusion process. Finally, we formulate the above components into a simple yet effective network, named ExpRDiff. Experimental results show that ExpRDiff performs favorably against state-of-the-art methods.

Embedding and Enriching Explicit Semantics for Visible-Infrared Person Re-Identification

Visible-infrared person re-identification (VIReID) retrieves pedestrian images with the same identity across different modalities. Existing methods learn visual content solely from images, lacking the capability to sense high-level semantics. In this paper, we propose an Embedding and Enriching Explicit Semantics (EEES) framework to learn semantically rich cross-modality pedestrian representations. Our method offers several contributions. First, with the collaboration of multiple large language-vision models, we develop Explicit Semantics Embedding (ESE), which automatically supplements language descriptions for pedestrians and aligns image-text pairs into a common space, thereby learning visual content associated with explicit semantics. Second, recognizing the complementarity of multi-view information, we present Cross-View Semantics Compensation (CVSC), which constructs multi-view image-text pair representations, establishes their many-to-many matching, and propagates knowledge to single-view representations, thus compensating visual content with its missing cross-view semantics. Third, to eliminate noisy semantics such as conflicting color attributes in different modalities, we design Cross-Modality Semantics Purification (CMSP), which constrains the distance between inter-modality image-text pair representations to be close to that between intra-modality image-text pair representations, further enhancing the modality-invariance of visual content. Finally, experimental results demonstrate the effectiveness and superiority of the proposed EEES.

FoundIR: Unleashing Million-scale Training Data to Advance Foundation Models for Image Restoration

Despite the significant progress made by all-in-one models in universal image restoration, existing methods suffer from a generalization bottleneck in real-world scenarios, as they are mostly trained on small-scale synthetic datasets with limited degradations. Therefore, large-scale high-quality real-world training data is urgently needed to facilitate the emergence of foundational models for image restoration. To advance this field, we spare no effort in contributing a million-scale dataset with two notable advantages over existing training data: real-world samples with larger-scale, and degradation types with higher diversity. By adjusting internal camera settings and external imaging conditions, we can capture aligned image pairs using our well-designed data acquisition system over multiple rounds and our data alignment criterion. Moreover, we propose a robust model, FoundIR, to better address a broader range of restoration tasks in real-world scenarios, taking a further step toward foundation models. Specifically, we first utilize a diffusion-based generalist model to remove degradations by learning the degradation-agnostic common representations from diverse inputs, where incremental learning strategy is adopted to better guide model training. To refine the model's restoration capability in complex scenarios, we introduce degradation-aware specialist models for achieving final high-quality results. Extensive experiments show the value of our dataset and the effectiveness of our method.

Divide-and-Conquer: Confluent Triple-Flow Network for RGB-T Salient Object Detection

RGB-Thermal Salient Object Detection aims to pinpoint prominent objects within aligned pairs of visible and thermal infrared images. Traditional encoder-decoder architectures, while designed for cross-modality feature interactions, may not have adequately considered the robustness against noise originating from defective modalities. Inspired by hierarchical human visual systems, we propose the ConTriNet, a robust Confluent Triple-Flow Network employing a Divide-and-Conquer strategy. Specifically, ConTriNet comprises three flows: two modality-specific flows explore cues from RGB and Thermal modalities, and a third modality-complementary flow integrates cues from both modalities. ConTriNet presents several notable advantages. It incorporates a Modality-induced Feature Modulator in the modality-shared union encoder to minimize inter-modality discrepancies and mitigate the impact of defective samples. Additionally, a foundational Residual Atrous Spatial Pyramid Module in the separated flows enlarges the receptive field, allowing for the capture of multi-scale contextual information. Furthermore, a Modality-aware Dynamic Aggregation Module in the modality-complementary flow dynamically aggregates saliency-related cues from both modality-specific flows. Leveraging the proposed parallel triple-flow framework, we further refine saliency maps derived from different flows through a flow-cooperative fusion strategy, yielding a high-quality, full-resolution saliency map for the final prediction. To evaluate the robustness and stability of our approach, we collect a comprehensive RGB-T SOD benchmark, VT-IMAG, covering various real-world challenging scenarios. Extensive experiments on public benchmarks and our VT-IMAG dataset demonstrate that ConTriNet consistently outperforms state-of-the-art competitors in both common and challenging scenarios.

Precision Profile Pollution Attack on Sequential Recommenders via Influence Function

Sequential recommendation approaches have demonstrated remarkable proficiency in modeling user preferences. Nevertheless, they are susceptible to profile pollution attacks (PPA), wherein items are introduced into a user's interaction history deliberately to influence the recommendation list. Since retraining the model for each polluted item is time-consuming, recent PPAs estimate item influence based on gradient directions to identify the most effective attack candidates. However, the actual item representations diverge significantly from the gradients, resulting in disparate outcomes.To tackle this challenge, we introduce an INFluence Function-based Attack approach INFAttack that offers a more accurate estimation of the influence of polluting items. Specifically, we calculate the modifications to the original model using the influence function when generating polluted sequences by introducing specific items. Subsequently, we choose the sequence that has been most significantly influenced to substitute the original sequence, thus promoting the target item. Comprehensive experiments conducted on five real-world datasets illustrate that INFAttack surpasses all baseline methods and consistently delivers stable attack performance for both popular and unpopular items.