Recent pre-trained vision-language models (PT-VLMs) often face a Multi-Domain Class-Incremental Learning (MCIL) scenario in practice, where several classes and domains of multi-modal tasks are incrementally arrived. Without access to previously learned tasks and unseen tasks, memory-constrained MCIL suffers from forward and backward forgetting. To alleviate the above challenges, parameter-efficient fine-tuning techniques (PEFT), such as prompt tuning, are employed to adapt the PT-VLM to the diverse incrementally learned tasks. To achieve effective new task adaptation, existing methods only consider the effect of PEFT strategy selection, but neglect the influence of PEFT parameter setting (e.g., prompting). In this paper, we tackle the challenge of optimizing prompt designs for diverse tasks in MCIL and propose an Instance-Aware Prompting (IAP) framework. Specifically, our Instance-Aware Gated Prompting (IA-GP) module enhances adaptation to new tasks while mitigating forgetting by dynamically assigning prompts across transformer layers at the instance level. Our Instance-Aware Class-Distribution-Driven Prompting (IA-CDDP) improves the task adaptation process by determining an accurate task-label-related confidence score for each instance. Experimental evaluations across 11 datasets, using three performance metrics, demonstrate the effectiveness of our proposed method. Code can be found at https://github.com/FerdinandZJU/IAP.
Multi-modal Large Langue Models (MLLMs) often process thousands of visual tokens, which consume a significant portion of the context window and impose a substantial computational burden. Prior work has empirically explored visual token pruning methods based on MLLMs' intermediate states (e.g., attention scores). However, they have limitations in precisely defining visual redundancy due to their inability to capture the influence of visual tokens on MLLMs' visual understanding (i.e., the predicted probabilities for textual token candidates). To address this issue, we manipulate the visual input and investigate variations in the textual output from both token-centric and context-centric perspectives, achieving intuitive and comprehensive analysis. Experimental results reveal that visual tokens with low ViT-[cls] association and low text-to-image attention scores can contain recognizable information and significantly contribute to images' overall information. To develop a more reliable method for identifying and pruning redundant visual tokens, we integrate these two perspectives and introduce a context-independent condition to identify redundant prototypes from training images, which probes the redundancy of each visual token during inference. Extensive experiments on single-image, multi-image and video comprehension tasks demonstrate the effectiveness of our method, notably achieving 90% to 110% of the performance while pruning 80% to 90% of visual tokens.
Due to the high diversity of image styles, the scalability to various styles plays a critical role in real-world applications. To accommodate a large amount of styles, previous multi-style transfer approaches rely on enlarging the model size while arbitrary-style transfer methods utilize heavy backbones. However, the additional computational cost introduced by more model parameters hinders these methods to be deployed on resource-limited devices. To address this challenge, in this paper, we develop a style transfer framework by decoupling the style modeling and transferring. Specifically, for style modeling, we propose a style representation learning scheme to encode the style information into a compact representation. Then, for style transferring, we develop a style-aware multi-style transfer network (SaMST) to adapt to diverse styles using pluggable style representations. In this way, our framework is able to accommodate diverse image styles in the learned style representations without introducing additional overhead during inference, thereby maintaining efficiency. Experiments show that our style representation can extract accurate style information. Moreover, qualitative and quantitative results demonstrate that our method achieves state-of-the-art performance in terms of both accuracy and efficiency. The codes are available in https://github.com/The-Learning-And-Vision-Atelier-LAVA/SaMST.
Transformer-based methods have achieved remarkable results in image super-resolution tasks because they can capture non-local dependencies in low-quality input images. However, this feature-intensive modeling approach is computationally expensive because it calculates the similarities between numerous features that are irrelevant to the query features when obtaining attention weights. These unnecessary similarity calculations not only degrade the reconstruction performance but also introduce significant computational overhead. How to accurately identify the features that are important to the current query features and avoid similarity calculations between irrelevant features remains an urgent problem. To address this issue, we propose a novel and effective Progressive Focused Transformer (PFT) that links all isolated attention maps in the network through Progressive Focused Attention (PFA) to focus attention on the most important tokens. PFA not only enables the network to capture more critical similar features, but also significantly reduces the computational cost of the overall network by filtering out irrelevant features before calculating similarities. Extensive experiments demonstrate the effectiveness of the proposed method, achieving state-of-the-art performance on various single image super-resolution benchmarks.
We introduce EditCLIP, a novel representation-learning approach for image editing. Our method learns a unified representation of edits by jointly encoding an input image and its edited counterpart, effectively capturing their transformation. To evaluate its effectiveness, we employ EditCLIP to solve two tasks: exemplar-based image editing and automated edit evaluation. In exemplar-based image editing, we replace text-based instructions in InstructPix2Pix with EditCLIP embeddings computed from a reference exemplar image pair. Experiments demonstrate that our approach outperforms state-of-the-art methods while being more efficient and versatile. For automated evaluation, EditCLIP assesses image edits by measuring the similarity between the EditCLIP embedding of a given image pair and either a textual editing instruction or the EditCLIP embedding of another reference image pair. Experiments show that EditCLIP aligns more closely with human judgments than existing CLIP-based metrics, providing a reliable measure of edit quality and structural preservation.
This report presents Wan, a comprehensive and open suite of video foundation models designed to push the boundaries of video generation. Built upon the mainstream diffusion transformer paradigm, Wan achieves significant advancements in generative capabilities through a series of innovations, including our novel VAE, scalable pre-training strategies, large-scale data curation, and automated evaluation metrics. These contributions collectively enhance the model's performance and versatility. Specifically, Wan is characterized by four key features: Leading Performance: The 14B model of Wan, trained on a vast dataset comprising billions of images and videos, demonstrates the scaling laws of video generation with respect to both data and model size. It consistently outperforms the existing open-source models as well as state-of-the-art commercial solutions across multiple internal and external benchmarks, demonstrating a clear and significant performance superiority. Comprehensiveness: Wan offers two capable models, i.e., 1.3B and 14B parameters, for efficiency and effectiveness respectively. It also covers multiple downstream applications, including image-to-video, instruction-guided video editing, and personal video generation, encompassing up to eight tasks. Consumer-Grade Efficiency: The 1.3B model demonstrates exceptional resource efficiency, requiring only 8.19 GB VRAM, making it compatible with a wide range of consumer-grade GPUs. Openness: We open-source the entire series of Wan, including source code and all models, with the goal of fostering the growth of the video generation community. This openness seeks to significantly expand the creative possibilities of video production in the industry and provide academia with high-quality video foundation models. All the code and models are available at https://github.com/Wan-Video/Wan2.1.
Language Bottleneck Models (LBMs) are proposed to achieve interpretable image recognition by classifying images based on textual concept bottlenecks. However, current LBMs simply list all concepts together as the bottleneck layer, leading to the spurious cue inference problem and cannot generalized to unseen classes. To address these limitations, we propose the Attribute-formed Language Bottleneck Model (ALBM). ALBM organizes concepts in the attribute-formed class-specific space, where concepts are descriptions of specific attributes for specific classes. In this way, ALBM can avoid the spurious cue inference problem by classifying solely based on the essential concepts of each class. In addition, the cross-class unified attribute set also ensures that the concept spaces of different classes have strong correlations, as a result, the learned concept classifier can be easily generalized to unseen classes. Moreover, to further improve interpretability, we propose Visual Attribute Prompt Learning (VAPL) to extract visual features on fine-grained attributes. Furthermore, to avoid labor-intensive concept annotation, we propose the Description, Summary, and Supplement (DSS) strategy to automatically generate high-quality concept sets with a complete and precise attribute. Extensive experiments on 9 widely used few-shot benchmarks demonstrate the interpretability, transferability, and performance of our approach. The code and collected concept sets are available at https://github.com/tiggers23/ALBM.
The rapid advancement of large vision-language models (LVLMs) has driven significant progress in multimodal tasks, enabling models to interpret, reason, and generate outputs across both visual and textual domains. While excelling in generative tasks, existing LVLMs often face limitations in tasks requiring high-fidelity representation learning, such as generating image or text embeddings for retrieval. Recent work has proposed finetuning LVLMs for representational learning, but the fine-tuned model often loses its generative capabilities due to the representational learning training paradigm. To address this trade-off, we introduce CAFe, a contrastive-autoregressive fine-tuning framework that enhances LVLMs for both representation and generative tasks. By integrating a contrastive objective with autoregressive language modeling, our approach unifies these traditionally separate tasks, achieving state-of-the-art results in both multimodal retrieval and multimodal generative benchmarks, including object hallucination (OH) mitigation. CAFe establishes a novel framework that synergizes embedding and generative functionalities in a single model, setting a foundation for future multimodal models that excel in both retrieval precision and coherent output generation.
Advanced analytics have transformed how sports teams operate, particularly in episodic sports like baseball. Their impact on continuous invasion sports, such as soccer and ice hockey, has been limited due to increased game complexity and restricted access to high-resolution game tracking data. In this demo, we present a method to collect and utilize simulated soccer tracking data from the Google Research Football environment to support the development of models designed for continuous tracking data. The data is stored in a schema that is representative of real tracking data and we provide processes that extract high-level features and events. We include examples of established tracking data models to showcase the efficacy of the simulated data. We address the scarcity of publicly available tracking data, providing support for research at the intersection of artificial intelligence and sports analytics.
Pre-trained video large language models (Video LLMs) exhibit remarkable reasoning capabilities, yet adapting these models to new tasks involving additional modalities or data types (e.g., audio or 3D information) remains challenging. In this paper, we present PAVE, a flexible framework for adapting pre-trained Video LLMs to downstream tasks with side-channel signals, such as audio, 3D cues, or multi-view videos. PAVE introduces lightweight adapters, referred to as "patches," which add a small number of parameters and operations to a base model without modifying its architecture or pre-trained weights. In doing so, PAVE can effectively adapt the pre-trained base model to support diverse downstream tasks, including audio-visual question answering, 3D reasoning, multi-view video recognition, and high frame rate video understanding. Across these tasks, PAVE significantly enhances the performance of the base model, surpassing state-of-the-art task-specific models while incurring a minor cost of ~0.1% additional FLOPs and parameters. Further, PAVE supports multi-task learning and generalizes well across different Video LLMs. Our code is available at https://github.com/dragonlzm/PAVE.
Image generation technology has brought significant advancements across various fields but has also raised concerns about data misuse and potential rights infringements, particularly with respect to creating visual artworks. Current methods aimed at safeguarding artworks often employ adversarial attacks. However, these methods face challenges such as poor transferability, high computational costs, and the introduction of noticeable noise, which compromises the aesthetic quality of the original artwork. To address these limitations, we propose a Structurally Imperceptible and Transferable Adversarial (SITA) attacks. SITA leverages a CLIP-based destylization loss, which decouples and disrupts the robust style representation of the image. This disruption hinders style extraction during stylized image generation, thereby impairing the overall stylization process. Importantly, SITA eliminates the need for a surrogate diffusion model, leading to significantly reduced computational overhead. The method's robust style feature disruption ensures high transferability across diverse models. Moreover, SITA introduces perturbations by embedding noise within the imperceptible structural details of the image. This approach effectively protects against style extraction without compromising the visual quality of the artwork. Extensive experiments demonstrate that SITA offers superior protection for artworks against unauthorized use in stylized generation. It significantly outperforms existing methods in terms of transferability, computational efficiency, and noise imperceptibility. Code is available at https://github.com/A-raniy-day/SITA.
Image-event joint depth estimation methods leverage complementary modalities for robust perception, yet face challenges in generalizability stemming from two factors: 1) limited annotated image-event-depth datasets causing insufficient cross-modal supervision, and 2) inherent frequency mismatches between static images and dynamic event streams with distinct spatiotemporal patterns, leading to ineffective feature fusion. To address this dual challenge, we propose Frequency-decoupled Unified Self-supervised Encoder (FUSE) with two synergistic components: The Parameter-efficient Self-supervised Transfer (PST) establishes cross-modal knowledge transfer through latent space alignment with image foundation models, effectively mitigating data scarcity by enabling joint encoding without depth ground truth. Complementing this, we propose the Frequency-Decoupled Fusion module (FreDFuse) to explicitly decouple high-frequency edge features from low-frequency structural components, resolving modality-specific frequency mismatches through physics-aware fusion. This combined approach enables FUSE to construct a universal image-event encoder that only requires lightweight decoder adaptation for target datasets. Extensive experiments demonstrate state-of-the-art performance with 14% and 24.9% improvements in Abs.Rel on MVSEC and DENSE datasets. The framework exhibits remarkable zero-shot adaptability to challenging scenarios including extreme lighting and motion blur, significantly advancing real-world deployment capabilities. The source code for our method is publicly available at: https://github.com/sunpihai-up/FUSE
Video camouflaged object segmentation (VCOS), aiming at segmenting camouflaged objects that seamlessly blend into their environment, is a fundamental vision task with various real-world applications. With the release of SAM2, video segmentation has witnessed significant progress. However, SAM2's capability of segmenting camouflaged videos is suboptimal, especially when given simple prompts such as point and box. To address the problem, we propose Camouflaged SAM2 (CamSAM2), which enhances SAM2's ability to handle camouflaged scenes without modifying SAM2's parameters. Specifically, we introduce a decamouflaged token to provide the flexibility of feature adjustment for VCOS. To make full use of fine-grained and high-resolution features from the current frame and previous frames, we propose implicit object-aware fusion (IOF) and explicit object-aware fusion (EOF) modules, respectively. Object prototype generation (OPG) is introduced to abstract and memorize object prototypes with informative details using high-quality features from previous frames. Extensive experiments are conducted to validate the effectiveness of our approach. While CamSAM2 only adds negligible learnable parameters to SAM2, it substantially outperforms SAM2 on three VCOS datasets, especially achieving 12.2 mDice gains with click prompt on MoCA-Mask and 19.6 mDice gains with mask prompt on SUN-SEG-Hard, with Hiera-T as the backbone. The code will be available at https://github.com/zhoustan/CamSAM2.
The hallucination of large multimodal models (LMMs), providing responses that appear correct but are actually incorrect, limits their reliability and applicability. This paper aims to study the hallucination problem of LMMs in video modality, which is dynamic and more challenging compared to static modalities like images and text. From this motivation, we first present a comprehensive benchmark termed HAVEN for evaluating hallucinations of LMMs in video understanding tasks. It is built upon three dimensions, i.e., hallucination causes, hallucination aspects, and question formats, resulting in 6K questions. Then, we quantitatively study 7 influential factors on hallucinations, e.g., duration time of videos, model sizes, and model reasoning, via experiments of 16 LMMs on the presented benchmark. In addition, inspired by recent thinking models like OpenAI o1, we propose a video-thinking model to mitigate the hallucinations of LMMs via supervised reasoning fine-tuning (SRFT) and direct preference optimization (TDPO)-- where SRFT enhances reasoning capabilities while TDPO reduces hallucinations in the thinking process. Extensive experiments and analyses demonstrate the effectiveness. Remarkably, it improves the baseline by 7.65% in accuracy on hallucination evaluation and reduces the bias score by 4.5%. The code and data are public at https://github.com/Hongcheng-Gao/HAVEN.
Deep learning-based image registration methods have shown state-of-the-art performance and rapid inference speeds. Despite these advances, many existing approaches fall short in capturing spatially varying information in non-local regions of feature maps due to the reliance on spatially-shared convolution kernels. This limitation leads to suboptimal estimation of deformation fields. In this paper, we propose a 3D Spatial-Awareness Convolution Block (SACB) to enhance the spatial information within feature representations. Our SACB estimates the spatial clusters within feature maps by leveraging feature similarity and subsequently parameterizes the adaptive convolution kernels across diverse regions. This adaptive mechanism generates the convolution kernels (weights and biases) tailored to spatial variations, thereby enabling the network to effectively capture spatially varying information. Building on SACB, we introduce a pyramid flow estimator (named SACB-Net) that integrates SACBs to facilitate multi-scale flow composition, particularly addressing large deformations. Experimental results on the brain IXI and LPBA datasets as well as Abdomen CT datasets demonstrate the effectiveness of SACB and the superiority of SACB-Net over the state-of-the-art learning-based registration methods. The code is available at https://github.com/x-xc/SACB_Net .
Accurate object segmentation is crucial for high-quality scene understanding in the 3D vision domain. However, 3D segmentation based on 3D Gaussian Splatting (3DGS) struggles with accurately delineating object boundaries, as Gaussian primitives often span across object edges due to their inherent volume and the lack of semantic guidance during training. In order to tackle these challenges, we introduce Clear Object Boundaries for 3DGS Segmentation (COB-GS), which aims to improve segmentation accuracy by clearly delineating blurry boundaries of interwoven Gaussian primitives within the scene. Unlike existing approaches that remove ambiguous Gaussians and sacrifice visual quality, COB-GS, as a 3DGS refinement method, jointly optimizes semantic and visual information, allowing the two different levels to cooperate with each other effectively. Specifically, for the semantic guidance, we introduce a boundary-adaptive Gaussian splitting technique that leverages semantic gradient statistics to identify and split ambiguous Gaussians, aligning them closely with object boundaries. For the visual optimization, we rectify the degraded suboptimal texture of the 3DGS scene, particularly along the refined boundary structures. Experimental results show that COB-GS substantially improves segmentation accuracy and robustness against inaccurate masks from pre-trained model, yielding clear boundaries while preserving high visual quality. Code is available at https://github.com/ZestfulJX/COB-GS.
The progress on generative models has led to significant advances on text-to-video (T2V) generation, yet the motion controllability of generated videos remains limited. Existing motion transfer methods explored the motion representations of reference videos to guide generation. Nevertheless, these methods typically rely on sample-specific optimization strategy, resulting in high computational burdens. In this paper, we propose EfficientMT, a novel and efficient end-to-end framework for video motion transfer. By leveraging a small set of synthetic paired motion transfer samples, EfficientMT effectively adapts a pretrained T2V model into a general motion transfer framework that can accurately capture and reproduce diverse motion patterns. Specifically, we repurpose the backbone of the T2V model to extract temporal information from reference videos, and further propose a scaler module to distill motion-related information. Subsequently, we introduce a temporal integration mechanism that seamlessly incorporates reference motion features into the video generation process. After training on our self-collected synthetic paired samples, EfficientMT enables general video motion transfer without requiring test-time optimization. Extensive experiments demonstrate that our EfficientMT outperforms existing methods in efficiency while maintaining flexible motion controllability. Our code will be available https://github.com/PrototypeNx/EfficientMT.
Long-context video modeling is essential for enabling generative models to function as world simulators, as they must maintain temporal coherence over extended time spans. However, most existing models are trained on short clips, limiting their ability to capture long-range dependencies, even with test-time extrapolation. While training directly on long videos is a natural solution, the rapid growth of vision tokens makes it computationally prohibitive. To support exploring efficient long-context video modeling, we first establish a strong autoregressive baseline called Frame AutoRegressive (FAR). FAR models temporal dependencies between continuous frames, converges faster than video diffusion transformers, and outperforms token-level autoregressive models. Based on this baseline, we observe context redundancy in video autoregression. Nearby frames are critical for maintaining temporal consistency, whereas distant frames primarily serve as context memory. To eliminate this redundancy, we propose the long short-term context modeling using asymmetric patchify kernels, which apply large kernels to distant frames to reduce redundant tokens, and standard kernels to local frames to preserve fine-grained detail. This significantly reduces the training cost of long videos. Our method achieves state-of-the-art results on both short and long video generation, providing an effective baseline for long-context autoregressive video modeling.
Recent advances in graph learning have paved the way for innovative retrieval-augmented generation (RAG) systems that leverage the inherent relational structures in graph data. However, many existing approaches suffer from rigid, fixed settings and significant engineering overhead, limiting their adaptability and scalability. Additionally, the RAG community has largely overlooked the decades of research in the graph database community regarding the efficient retrieval of interesting substructures on large-scale graphs. In this work, we introduce the RAG-on-Graphs Library (RGL), a modular framework that seamlessly integrates the complete RAG pipeline-from efficient graph indexing and dynamic node retrieval to subgraph construction, tokenization, and final generation-into a unified system. RGL addresses key challenges by supporting a variety of graph formats and integrating optimized implementations for essential components, achieving speedups of up to 143x compared to conventional methods. Moreover, its flexible utilities, such as dynamic node filtering, allow for rapid extraction of pertinent subgraphs while reducing token consumption. Our extensive evaluations demonstrate that RGL not only accelerates the prototyping process but also enhances the performance and applicability of graph-based RAG systems across a range of tasks.
This paper proposes a general solution to enable point cloud recognition models to handle distribution shifts at test time. Unlike prior methods, which rely heavily on training data (often inaccessible during online inference) and are limited to recognizing a fixed set of point cloud classes predefined during training, we explore a more practical and challenging scenario: adapting the model solely based on online test data to recognize both previously seen classes and novel, unseen classes at test time. To this end, we develop \textbf{Point-Cache}, a hierarchical cache model that captures essential clues of online test samples, particularly focusing on the global structure of point clouds and their local-part details. Point-Cache, which serves as a rich 3D knowledge base, is dynamically managed to prioritize the inclusion of high-quality samples. Designed as a plug-and-play module, our method can be flexibly integrated into large multimodal 3D models to support open-vocabulary point cloud recognition. Notably, our solution operates with efficiency comparable to zero-shot inference, as it is entirely training-free. Point-Cache demonstrates substantial gains across 8 challenging benchmarks and 4 representative large 3D models, highlighting its effectiveness. Code is available at https://github.com/auniquesun/Point-Cache.