Zongze Wu

Xudong Tian, Zhizhong Zhang, Cong Wang et al.

Information Bottleneck (IB) based multi-view learning provides an information theoretic principle for seeking shared information contained in heterogeneous data descriptions. However, its great success is generally attributed to estimate the multivariate mutual information which is intractable when the network becomes complicated. Moreover, the representation learning tradeoff, {\it i.e.}, prediction-compression and sufficiency-consistency tradeoff, makes the IB hard to satisfy both requirements simultaneously. In this paper, we design several variational information bottlenecks to exploit two key characteristics ({\it i.e.}, sufficiency and consistency) for multi-view representation learning. Specifically, we propose a Multi-View Variational Distillation (MV$^2$D) strategy to provide a scalable, flexible and analytical solution to fitting MI by giving arbitrary input of viewpoints but without explicitly estimating it. Under rigorously theoretical guarantee, our approach enables IB to grasp the intrinsic correlation between observations and semantic labels, producing predictive and compact representations naturally. Also, our information-theoretic constraint can effectively neutralize the sensitivity to heterogeneous data by eliminating both task-irrelevant and view-specific information, preventing both tradeoffs in multiple view cases. To verify our theoretically grounded strategies, we apply our approaches to various benchmarks under three different applications. Extensive experiments to quantitatively and qualitatively demonstrate the effectiveness of our approach against state-of-the-art methods.

Jaskirat Singh, Boyang Zheng, Zongze Wu et al.

Representation Autoencoders (RAE) replace traditional VAE with pretrained vision encoders. In this paper, we systematically investigate several design choices and find three insights which simplify and improve RAE. First, we study a generalized formulation where the representation is defined as sum of the last k encoder layers rather than solely the final layer. This simple change greatly improves reconstruction without encoder finetuning or specialized data (e.g., text, faces). Second, we study the prevalent assumption that RAE (using pretrained representation as encoder) replaces representation alignment (REPA), which distills the same representation to intermediate layers instead. Through large-scale empirical analysis, we uncover a surprising finding: RAE and REPA exhibit complementary working mechanisms, allowing the same representation to be used as both encoder and target for intermediate diffusion layers. Finally, the original RAE struggles with classifier-free guidance (CFG) and requires training a second, weaker diffusion model for AutoGuidance (AG). We show that REPA itself can be viewed as x-prediction in RAE latent space. By simply re-parameterizing the output of the DiT model, it can provide guidance for "free". Overall, RAEv2 leads to more than 10x faster convergence over the original RAE, achieving a state-of-the-art gFID of 1.06 in just 80 epochs on ImageNet-256. On FDr^k, RAEv2 achieves a state-of-the-art 2.17 at just 80 epochs compared to the previous best 3.26 (800 epochs) without any post-training. This motivates EP_FID@k (epochs to reach unguided gFID <= k) as a measure of training efficiency. RAEv2 attains an EP_FID@2 of 35 epochs, versus 177 for the original RAE. We also validate our approach across diverse settings for text-to-image generation and navigation world models, showing consistent improvements. Code is available at https://raev2.github.io.

Zongze Wu, Nicholas Kolkin, Jonathan Brandt et al.

We address the challenges of precise image inversion and disentangled image editing in the context of few-step diffusion models. We introduce an encoder based iterative inversion technique. The inversion network is conditioned on the input image and the reconstructed image from the previous step, allowing for correction of the next reconstruction towards the input image. We demonstrate that disentangled controls can be easily achieved in the few-step diffusion model by conditioning on an (automatically generated) detailed text prompt. To manipulate the inverted image, we freeze the noise maps and modify one attribute in the text prompt (either manually or via instruction based editing driven by an LLM), resulting in the generation of a new image similar to the input image with only one attribute changed. It can further control the editing strength and accept instructive text prompt. Our approach facilitates realistic text-guided image edits in real-time, requiring only 8 number of functional evaluations (NFEs) in inversion (one-time cost) and 4 NFEs per edit. Our method is not only fast, but also significantly outperforms state-of-the-art multi-step diffusion editing techniques.

Xiaopin Zhong, Guankun Wang, Weixiang Liu et al.

As a fundamental computer vision task, crowd counting plays an important role in public safety. Currently, deep learning based head detection is a promising method for crowd counting. However, the highly concerned object detection networks cannot be well applied to this problem for three reasons: (1) Existing loss functions fail to address sample imbalance in highly dense and complex scenes; (2) Canonical object detectors lack spatial coherence in loss calculation, disregarding the relationship between object location and background region; (3) Most of the head detection datasets are only annotated with the center points, i.e. without bounding boxes. To overcome these issues, we propose a novel Mask Focal Loss (MFL) based on heatmap via the Gaussian kernel. MFL provides a unifying framework for the loss functions based on both heatmap and binary feature map ground truths. Additionally, we introduce GTA_Head, a synthetic dataset with comprehensive annotations, for evaluation and comparison. Extensive experimental results demonstrate the superior performance of our MFL across various detectors and datasets, and it can reduce MAE and RMSE by up to 47.03% and 61.99%, respectively. Therefore, our work presents a strong foundation for advancing crowd counting methods based on density estimation.

Mengqi Lei, Siqi Li, Yihong Wu et al.

The YOLO series models reign supreme in real-time object detection due to their superior accuracy and computational efficiency. However, both the convolutional architectures of YOLO11 and earlier versions and the area-based self-attention mechanism introduced in YOLOv12 are limited to local information aggregation and pairwise correlation modeling, lacking the capability to capture global multi-to-multi high-order correlations, which limits detection performance in complex scenarios. In this paper, we propose YOLOv13, an accurate and lightweight object detector. To address the above-mentioned challenges, we propose a Hypergraph-based Adaptive Correlation Enhancement (HyperACE) mechanism that adaptively exploits latent high-order correlations and overcomes the limitation of previous methods that are restricted to pairwise correlation modeling based on hypergraph computation, achieving efficient global cross-location and cross-scale feature fusion and enhancement. Subsequently, we propose a Full-Pipeline Aggregation-and-Distribution (FullPAD) paradigm based on HyperACE, which effectively achieves fine-grained information flow and representation synergy within the entire network by distributing correlation-enhanced features to the full pipeline. Finally, we propose to leverage depthwise separable convolutions to replace vanilla large-kernel convolutions, and design a series of blocks that significantly reduce parameters and computational complexity without sacrificing performance. We conduct extensive experiments on the widely used MS COCO benchmark, and the experimental results demonstrate that our method achieves state-of-the-art performance with fewer parameters and FLOPs. Specifically, our YOLOv13-N improves mAP by 3.0\% over YOLO11-N and by 1.5\% over YOLOv12-N. The code and models of our YOLOv13 model are available at: https://github.com/iMoonLab/yolov13.

Jiezhou Chen, Guankun Wang, Weixiang Liu et al.

Head detection provides distribution information of pedestrian, which is crucial for scene statistical analysis, traffic management, and risk assessment and early warning. However, scene complexity and large-scale variation in the real world make accurate detection more difficult. Therefore, we present a modified Yolov8 which improves head detection performance through reinforcing target perception. An Auxiliary Learning Feature Fusion (ALFF) module comprised of LSTM and convolutional blocks is used as the auxiliary task to help the model perceive targets. In addition, we introduce Noise Calibration into Distribution Focal Loss to facilitate model fitting and improve the accuracy of detection. Considering the requirements of high accuracy and speed for the head detection task, our method is adapted with two kinds of backbone, namely Yolov8n and Yolov8m. The results demonstrate the superior performance of our approach in improving detection accuracy and robustness.

Jianye Yi, Xiaopin Zhong, Weixiang Liu et al.

Semantic segmentation is a classic and fundamental computer vision problem dedicated to assigning each pixel with its corresponding class. Some recent methods introduce edge-based information for improving the segmentation performance. However these methods are specific and limited to certain network architectures, and they can not be transferred to other models or tasks. Therefore, we propose an abstract and universal edge supervision method called Edge-aware Plug-and-play Scheme (EPS), which can be easily and quickly applied to any semantic segmentation models. The core is edge-width/thickness preserving guided for semantic segmentation. The EPS first extracts the Edge Ground Truth (Edge GT) with a predefined edge thickness from the training data; and then for any network architecture, it directly copies the decoder head for the auxiliary task with the Edge GT supervision. To ensure the edge thickness preserving consistantly, we design a new boundarybased loss, called Polar Hausdorff (PH) Loss, for the auxiliary supervision. We verify the effectiveness of our EPS on the Cityscapes dataset using 22 models. The experimental results indicate that the proposed method can be seamlessly integrated into any state-of-the-art (SOTA) models with zero modification, resulting in promising enhancement of the segmentation performance.

Jaskirat Singh, Xingjian Leng, Zongze Wu et al.

Representation alignment (REPA) guides generative training by distilling representations from a strong, pretrained vision encoder to intermediate diffusion features. We investigate a fundamental question: what aspect of the target representation matters for generation, its \textit{global} \revision{semantic} information (e.g., measured by ImageNet-1K accuracy) or its spatial structure (i.e. pairwise cosine similarity between patch tokens)? Prevalent wisdom holds that stronger global semantic performance leads to better generation as a target representation. To study this, we first perform a large-scale empirical analysis across 27 different vision encoders and different model scales. The results are surprising; spatial structure, rather than global performance, drives the generation performance of a target representation. To further study this, we introduce two straightforward modifications, which specifically accentuate the transfer of \emph{spatial} information. We replace the standard MLP projection layer in REPA with a simple convolution layer and introduce a spatial normalization layer for the external representation. Surprisingly, our simple method (implemented in $<$4 lines of code), termed iREPA, consistently improves convergence speed of REPA, across a diverse set of vision encoders, model sizes, and training variants (such as REPA, REPA-E, Meanflow, JiT etc). %, etc. Our work motivates revisiting the fundamental working mechanism of representational alignment and how it can be leveraged for improved training of generative models. The code and project page are available at https://end2end-diffusion.github.io/irepa

Zhengyang Geng, Yiyang Lu, Zongze Wu et al.

MeanFlow (MF) has recently been established as a framework for one-step generative modeling. However, its ``fastforward'' nature introduces key challenges in both the training objective and the guidance mechanism. First, the original MF's training target depends not only on the underlying ground-truth fields but also on the network itself. To address this issue, we recast the objective as a loss on the instantaneous velocity $v$, re-parameterized by a network that predicts the average velocity $u$. Our reformulation yields a more standard regression problem and improves the training stability. Second, the original MF fixes the classifier-free guidance scale during training, which sacrifices flexibility. We tackle this issue by formulating guidance as explicit conditioning variables, thereby retaining flexibility at test time. The diverse conditions are processed through in-context conditioning, which reduces model size and benefits performance. Overall, our $\textbf{improved MeanFlow}$ ($\textbf{iMF}$) method, trained entirely from scratch, achieves $\textbf{1.72}$ FID with a single function evaluation (1-NFE) on ImageNet 256$\times$256. iMF substantially outperforms prior methods of this kind and closes the gap with multi-step methods while using no distillation. We hope our work will further advance fastforward generative modeling as a stand-alone paradigm.

Yifan Feng, Chengwu Yang, Xingliang Hou et al.

Existing benchmarks like NLGraph and GraphQA evaluate LLMs on graphs by focusing mainly on pairwise relationships, overlooking the high-order correlations found in real-world data. Hypergraphs, which can model complex beyond-pairwise relationships, offer a more robust framework but are still underexplored in the context of LLMs. To address this gap, we introduce LLM4Hypergraph, the first comprehensive benchmark comprising 21,500 problems across eight low-order, five high-order, and two isomorphism tasks, utilizing both synthetic and real-world hypergraphs from citation networks and protein structures. We evaluate six prominent LLMs, including GPT-4o, demonstrating our benchmark's effectiveness in identifying model strengths and weaknesses. Our specialized prompting framework incorporates seven hypergraph languages and introduces two novel techniques, Hyper-BAG and Hyper-COT, which enhance high-order reasoning and achieve an average 4% (up to 9%) performance improvement on structure classification tasks. This work establishes a foundational testbed for integrating hypergraph computational capabilities into LLMs, advancing their comprehension. The source codes are at https://github.com/iMoonLab/LLM4Hypergraph.

Rohit Gandikota, Zongze Wu, Richard Zhang et al.

We present SliderSpace, a framework for automatically decomposing the visual capabilities of diffusion models into controllable and human-understandable directions. Unlike existing control methods that require a user to specify attributes for each edit direction individually, SliderSpace discovers multiple interpretable and diverse directions simultaneously from a single text prompt. Each direction is trained as a low-rank adaptor, enabling compositional control and the discovery of surprising possibilities in the model's latent space. Through extensive experiments on state-of-the-art diffusion models, we demonstrate SliderSpace's effectiveness across three applications: concept decomposition, artistic style exploration, and diversity enhancement. Our quantitative evaluation shows that SliderSpace-discovered directions decompose the visual structure of model's knowledge effectively, offering insights into the latent capabilities encoded within diffusion models. User studies further validate that our method produces more diverse and useful variations compared to baselines. Our code, data and trained weights are available at https://sliderspace.baulab.info

Xingjian Bai, Guande He, Zhengqi Li et al.

Causality -- referring to temporal, uni-directional cause-effect relationships between components -- underlies many complex generative processes, including videos, language, and robot trajectories. Current causal diffusion models entangle temporal reasoning with iterative denoising, applying causal attention across all layers, at every denoising step, and over the entire context. In this paper, we show that the causal reasoning in these models is separable from the multi-step denoising process. Through systematic probing of autoregressive video diffusers, we uncover two key regularities: (1) early layers produce highly similar features across denoising steps, indicating redundant computation along the diffusion trajectory; and (2) deeper layers exhibit sparse cross-frame attention and primarily perform intra-frame rendering. Motivated by these findings, we introduce Separable Causal Diffusion (SCD), a new architecture that explicitly decouples once-per-frame temporal reasoning, via a causal transformer encoder, from multi-step frame-wise rendering, via a lightweight diffusion decoder. Extensive experiments on both pretraining and post-training tasks across synthetic and real benchmarks show that SCD significantly improves throughput and per-frame latency while matching or surpassing the generation quality of strong causal diffusion baselines.

Jianye Yi, Xiaopin Zhong, Weixiang Liu et al.

The continuous development of the photovoltaic (PV) industry has raised high requirements for the quality of monocrystalline of PV module cells. When learning to segment defect regions in PV module cell images, Tiny Hidden Cracks (THC) lead to extremely-imbalanced samples. The ratio of defect pixels to normal pixels can be as low as 1:2000. This extreme imbalance makes it difficult to segment the THC of PV module cells, which is also a challenge for semantic segmentation. To address the problem of segmenting defects on extremely-imbalanced THC data, the paper makes contributions from three aspects: (1) it proposes an explicit measure for output imbalance; (2) it generalizes a distribution-based loss that can handle different types of output imbalances; and (3) it introduces a compound loss with our adaptive hyperparameter selection algorithm that can keep the consistency of training and inference for harmonizing the output imbalance on extremelyimbalanced input data. The proposed method is evaluated on four widely-used deep learning architectures and four datasets with varying degrees of input imbalance. The experimental results show that the proposed method outperforms existing methods.

Junhao Cai, Deyu Zeng, Junhao Pang et al.

Current text-to-3D generation methods excel in natural scenes but struggle with industrial applications due to two critical limitations: domain adaptation challenges where conventional LoRA fusion causes knowledge interference across categories, and geometric reasoning deficiencies where pairwise consistency constraints fail to capture higher-order structural dependencies essential for precision manufacturing. We propose a novel framework named ForgeDreamer addressing both challenges through two key innovations. First, we introduce a Multi-Expert LoRA Ensemble mechanism that consolidates multiple category-specific LoRA models into a unified representation, achieving superior cross-category generalization while eliminating knowledge interference. Second, building on enhanced semantic understanding, we develop a Cross-View Hypergraph Geometric Enhancement approach that captures structural dependencies spanning multiple viewpoints simultaneously. These components work synergistically improved semantic understanding, enables more effective geometric reasoning, while hypergraph modeling ensures manufacturing-level consistency. Extensive experiments on a custom industrial dataset demonstrate superior semantic generalization and enhanced geometric fidelity compared to state-of-the-art approaches. Our code and data are provided in the supplementary material attached in the appendix for review purposes.

Yuxuan Lin, Yang Chang, Xuan Tong et al.

In the advancement of industrial informatization, unsupervised anomaly detection technology effectively overcomes the scarcity of abnormal samples and significantly enhances the automation and reliability of smart manufacturing. As an important branch, industrial image anomaly detection focuses on automatically identifying visual anomalies in industrial scenarios (such as product surface defects, assembly errors, and equipment appearance anomalies) through computer vision techniques. With the rapid development of Unsupervised industrial Image Anomaly Detection (UIAD), excellent detection performance has been achieved not only in RGB setting but also in 3D and multimodal (RGB and 3D) settings. However, existing surveys primarily focus on UIAD tasks in RGB setting, with little discussion in 3D and multimodal settings. To address this gap, this artical provides a comprehensive review of UIAD tasks in the three modal settings. Specifically, we first introduce the task concept and process of UIAD. We then overview the research on UIAD in three modal settings (RGB, 3D, and multimodal), including datasets and methods, and review multimodal feature fusion strategies in multimodal setting. Finally, we summarize the main challenges faced by UIAD tasks in the three modal settings, and offer insights into future development directions, aiming to provide researchers with a comprehensive reference and offer new perspectives for the advancement of industrial informatization. Corresponding resources are available at https://github.com/Sunny5250/Awesome-Multi-Setting-UIAD.

Zuo Zuo, Jiahao Dong, Yue Gao et al.

In the manufacturing industry, defect detection is an essential but challenging task aiming to detect defects generated in the process of production. Though traditional YOLO models presents a good performance in defect detection, they still have limitations in capturing high-order feature interrelationships, which hurdles defect detection in the complex scenarios and across the scales. To this end, we introduce hypergraph computation into YOLO framework, dubbed HyperDefect-YOLO (HD-YOLO), to improve representative ability and semantic exploitation. HD-YOLO consists of Defect Aware Module (DAM) and Mixed Graph Network (MGNet) in the backbone, which specialize for perception and extraction of defect features. To effectively aggregate multi-scale features, we propose HyperGraph Aggregation Network (HGANet) which combines hypergraph and attention mechanism to aggregate multi-scale features. Cross-Scale Fusion (CSF) is proposed to adaptively fuse and handle features instead of simple concatenation and convolution. Finally, we propose Semantic Aware Module (SAM) in the neck to enhance semantic exploitation for accurately localizing defects with different sizes in the disturbed background. HD-YOLO undergoes rigorous evaluation on public HRIPCB and NEU-DET datasets with significant improvements compared to state-of-the-art methods. We also evaluate HD-YOLO on self-built MINILED dataset collected in real industrial scenarios to demonstrate the effectiveness of the proposed method. The source codes are at https://github.com/Jay-zzcoder/HD-YOLO.

Zuo Zuo, Jiahao Dong, Yao Wu et al.

Industrial anomaly classification (AC) is an indispensable task in industrial manufacturing, which guarantees quality and safety of various product. To address the scarcity of data in industrial scenarios, lots of few-shot anomaly detection methods emerge recently. In this paper, we propose an effective few-shot anomaly classification (FSAC) framework with one-stage training, dubbed CLIP-FSAC++. Specifically, we introduce a cross-modality interaction module named Anomaly Descriptor following image and text encoders, which enhances the correlation of visual and text embeddings and adapts the representations of CLIP from pre-trained data to target data. In anomaly descriptor, image-to-text cross-attention module is used to obtain image-specific text embeddings and text-to-image cross-attention module is used to obtain text-specific visual embeddings. Then these modality-specific embeddings are used to enhance original representations of CLIP for better matching ability. Comprehensive experiment results are provided for evaluating our method in few-normal shot anomaly classification on VisA and MVTEC-AD for 1, 2, 4 and 8-shot settings. The source codes are at https://github.com/Jay-zzcoder/clip-fsac-pp

Shivam Duggal, Xingjian Bai, Zongze Wu et al.

Latent diffusion models (LDMs) enable high-fidelity synthesis by operating in learned latent spaces. However, training state-of-the-art LDMs requires complex staging: a tokenizer must be trained first, before the diffusion model can be trained in the frozen latent space. We propose UNITE - an autoencoder architecture for unified tokenization and latent diffusion. UNITE consists of a Generative Encoder that serves as both image tokenizer and latent generator via weight sharing. Our key insight is that tokenization and generation can be viewed as the same latent inference problem under different conditioning regimes: tokenization infers latents from fully observed images, whereas generation infers them from noise together with text or class conditioning. Motivated by this, we introduce a single-stage training procedure that jointly optimizes both tasks via two forward passes through the same Generative Encoder. The shared parameters enable gradients to jointly shape the latent space, encouraging a "common latent language". Across image and molecule modalities, UNITE achieves near state of the art performance without adversarial losses or pretrained encoders (e.g., DINO), reaching FID 2.12 and 1.73 for Base and Large models on ImageNet 256 x 256. We further analyze the Generative Encoder through the lenses of representation alignment and compression. These results show that single stage joint training of tokenization & generation from scratch is feasible.

You Zhou, Jiangshan Zhao, Deyu Zeng et al.

Unsupervised Continuous Anomaly Detection (UCAD) faces significant challenges in multi-task representation learning, with existing methods suffering from incomplete representation and catastrophic forgetting. Unlike supervised models, unsupervised scenarios lack prior information, making it difficult to effectively distinguish redundant and complementary multimodal features. To address this, we propose the Multimodal Task Representation Memory Bank (MTRMB) method through two key technical innovations: A Key-Prompt-Multimodal Knowledge (KPMK) mechanism that uses concise key prompts to guide cross-modal feature interaction between BERT and ViT. Refined Structure-based Contrastive Learning (RSCL) leveraging Grounding DINO and SAM to generate precise segmentation masks, pulling features of the same structural region closer while pushing different structural regions apart. Experiments on MVtec AD and VisA datasets demonstrate MTRMB's superiority, achieving an average detection accuracy of 0.921 at the lowest forgetting rate, significantly outperforming state-of-the-art methods. We plan to open source on GitHub.

Yotam Nitzan, Zongze Wu, Richard Zhang et al.

We introduce a novel diffusion transformer, LazyDiffusion, that generates partial image updates efficiently. Our approach targets interactive image editing applications in which, starting from a blank canvas or an image, a user specifies a sequence of localized image modifications using binary masks and text prompts. Our generator operates in two phases. First, a context encoder processes the current canvas and user mask to produce a compact global context tailored to the region to generate. Second, conditioned on this context, a diffusion-based transformer decoder synthesizes the masked pixels in a "lazy" fashion, i.e., it only generates the masked region. This contrasts with previous works that either regenerate the full canvas, wasting time and computation, or confine processing to a tight rectangular crop around the mask, ignoring the global image context altogether. Our decoder's runtime scales with the mask size, which is typically small, while our encoder introduces negligible overhead. We demonstrate that our approach is competitive with state-of-the-art inpainting methods in terms of quality and fidelity while providing a 10x speedup for typical user interactions, where the editing mask represents 10% of the image.

Churong Liang, Jinling Gan, Kairan Hong et al.

A critical limitation in large-scale multi-agent systems is the cascading of errors. And without intermediate verification, downstream agents exacerbate upstream inaccuracies, resulting in significant quality degradation. To bridge this gap, we introduce \textbf{COCO} (\textbf{C}ognitive \textbf{O}perating System with \textbf{C}ontinuous \textbf{O}versight), a theoretically grounded framework for asynchronous self-monitoring and adaptive error correction in multi-agent systems. COCO reconciles the fundamental tension between quality assurance and computational efficiency via a novel decoupled architecture. This design isolates error detection from the critical execution path and incorporates an automated configuration engine to minimize deployment complexity. The framework relies on three algorithmic innovations to mitigate both systematic and stochastic errors: (1) a Contextual Rollback Mechanism that leverages execution history for informed state recovery rather than naive retries; (2) a Bidirectional Reflection Protocol to ensure convergence and prevent oscillatory control loops; and (3) a Heterogeneous Cross-Validation Mechanism that utilizes ensemble disagreement to identify bias and hallucinations. Extensive experiments on diverse benchmarks demonstrate that COCO delivers a 6.5\% average performance improvement. Notably, the framework achieves 95.1\% of large-model performance with a 30$\times$ parameter reduction, confirming the potential for efficient, high-reliability deployment, and establishing COCO as a practical, annotation-based solution for critical autonomous domains.

Wenjie Pei, Weina Xu, Zongze Wu et al.

The crux of graph classification lies in the effective representation learning for the entire graph. Typical graph neural networks focus on modeling the local dependencies when aggregating features of neighboring nodes, and obtain the representation for the entire graph by aggregating node features. Such methods have two potential limitations: 1) the global node saliency w.r.t. graph classification is not explicitly modeled, which is crucial since different nodes may have different semantic relevance to graph classification; 2) the graph representation directly aggregated from node features may have limited effectiveness to reflect graph-level information. In this work, we propose the Saliency-Aware Regularized Graph Neural Network (SAR-GNN) for graph classification, which consists of two core modules: 1) a traditional graph neural network serving as the backbone for learning node features and 2) the Graph Neural Memory designed to distill a compact graph representation from node features of the backbone. We first estimate the global node saliency by measuring the semantic similarity between the compact graph representation and node features. Then the learned saliency distribution is leveraged to regularize the neighborhood aggregation of the backbone, which facilitates the message passing of features for salient nodes and suppresses the less relevant nodes. Thus, our model can learn more effective graph representation. We demonstrate the merits of SAR-GNN by extensive experiments on seven datasets across various types of graph data. Code will be released.

Yifan Feng, Shiquan Liu, Xiangmin Han et al.

Hypergraph neural networks (HGNNs) effectively model complex high-order relationships in domains like protein interactions and social networks by connecting multiple vertices through hyperedges, enhancing modeling capabilities, and reducing information loss. Developing foundation models for hypergraphs is challenging due to their distinct data, which includes both vertex features and intricate structural information. We present Hyper-FM, a Hypergraph Foundation Model for multi-domain knowledge extraction, featuring Hierarchical High-Order Neighbor Guided Vertex Knowledge Embedding for vertex feature representation and Hierarchical Multi-Hypergraph Guided Structural Knowledge Extraction for structural information. Additionally, we curate 10 text-attributed hypergraph datasets to advance research between HGNNs and LLMs. Experiments on these datasets show that Hyper-FM outperforms baseline methods by approximately 13.3\%, validating our approach. Furthermore, we propose the first scaling law for hypergraph foundation models, demonstrating that increasing domain diversity significantly enhances performance, unlike merely augmenting vertex and hyperedge counts. This underscores the critical role of domain diversity in scaling hypergraph models.

Liangbin Xie, Daniil Pakhomov, Zhonghao Wang et al.

This paper introduces TurboFill, a fast image inpainting model that enhances a few-step text-to-image diffusion model with an inpainting adapter for high-quality and efficient inpainting. While standard diffusion models generate high-quality results, they incur high computational costs. We overcome this by training an inpainting adapter on a few-step distilled text-to-image model, DMD2, using a novel 3-step adversarial training scheme to ensure realistic, structurally consistent, and visually harmonious inpainted regions. To evaluate TurboFill, we propose two benchmarks: DilationBench, which tests performance across mask sizes, and HumanBench, based on human feedback for complex prompts. Experiments show that TurboFill outperforms both multi-step BrushNet and few-step inpainting methods, setting a new benchmark for high-performance inpainting tasks. Our project page: https://liangbinxie.github.io/projects/TurboFill/

Rulin Zhou, Yingjie Feng, Guankun Wang et al.

Adenoid hypertrophy stands as a common cause of obstructive sleep apnea-hypopnea syndrome in children. It is characterized by snoring, nasal congestion, and growth disorders. Computed Tomography (CT) emerges as a pivotal medical imaging modality, utilizing X-rays and advanced computational techniques to generate detailed cross-sectional images. Within the realm of pediatric airway assessments, CT imaging provides an insightful perspective on the shape and volume of enlarged adenoids. Despite the advances of deep learning methods for medical imaging analysis, there remains an emptiness in the segmentation of adenoid hypertrophy in CT scans. To address this research gap, we introduce TSUBF-Nett (Trans-Spatial UNet-like Network based on Bi-direction Fusion), a 3D medical image segmentation framework. TSUBF-Net is engineered to effectively discern intricate 3D spatial interlayer features in CT scans and enhance the extraction of boundary-blurring features. Notably, we propose two innovative modules within the U-shaped network architecture:the Trans-Spatial Perception module (TSP) and the Bi-directional Sampling Collaborated Fusion module (BSCF).These two modules are in charge of operating during the sampling process and strategically fusing down-sampled and up-sampled features, respectively. Furthermore, we introduce the Sobel loss term, which optimizes the smoothness of the segmentation results and enhances model accuracy. Extensive 3D segmentation experiments are conducted on several datasets. TSUBF-Net is superior to the state-of-the-art methods with the lowest HD95: 7.03, IoU:85.63, and DSC: 92.26 on our own AHSD dataset. The results in the other two public datasets also demonstrate that our methods can robustly and effectively address the challenges of 3D segmentation in CT scans.

Nikita Starodubcev, Daniil Pakhomov, Zongze Wu et al.

Diffusion transformers typically incorporate textual information via attention layers and a modulation mechanism using a pooled text embedding. Nevertheless, recent approaches discard modulation-based text conditioning and rely exclusively on attention. In this paper, we address whether modulation-based text conditioning is necessary and whether it can provide any performance advantage. Our analysis shows that, in its conventional usage, the pooled embedding contributes little to overall performance, suggesting that attention alone is generally sufficient for faithfully propagating prompt information. However, we reveal that the pooled embedding can provide significant gains when used from a different perspective-serving as guidance and enabling controllable shifts toward more desirable properties. This approach is training-free, simple to implement, incurs negligible runtime overhead, and can be applied to various diffusion models, bringing improvements across diverse tasks, including text-to-image/video generation and image editing.

Shelly Golan, Yotam Nitzan, Zongze Wu et al.

Creative generation is the synthesis of new, surprising, and valuable samples that reflect user intent yet cannot be envisioned in advance. This task aims to extend human imagination, enabling the discovery of visual concepts that exist in the unexplored spaces between familiar domains. While text-to-image diffusion models excel at rendering photorealistic scenes that faithfully match user prompts, they still struggle to generate genuinely novel content. Existing approaches to enhance generative creativity either rely on interpolation of image features, which restricts exploration to predefined categories, or require time-intensive procedures such as embedding optimization or model fine-tuning. We propose VLM-Guided Adaptive Negative-Prompting, a training-free, inference-time method that promotes creative image generation while preserving the validity of the generated object. Our approach utilizes a vision-language model (VLM) that analyzes intermediate outputs of the generation process and adaptively steers it away from conventional visual concepts, encouraging the emergence of novel and surprising outputs. We evaluate creativity through both novelty and validity, using statistical metrics in the CLIP embedding space. Through extensive experiments, we show consistent gains in creative novelty with negligible computational overhead. Moreover, unlike existing methods that primarily generate single objects, our approach extends to complex scenarios, such as generating coherent sets of creative objects and preserving creativity within elaborate compositional prompts. Our method integrates seamlessly into existing diffusion pipelines, offering a practical route to producing creative outputs that venture beyond the constraints of textual descriptions.

Zongze Wu, Yani Guo, Churong Liang et al.

Despite remarkable advances in Large Language Model capabilities, tool retrieval for agent-based systems remains fundamentally limited by reliance on semantic similarity, which fails to capture functional viability. Current methods often retrieve textually relevant but functionally inoperative tools due to parameter mismatches, authentication failures, and execution constraints--a phenomenon we term the semantic-functional gap. We introduce GRETEL, to address this gap through systematic empirical validation. GRETEL implements an agentic workflow that processes semantically retrieved candidates through sandboxed plan-execute-evaluate cycles, generating execution-grounded evidence to distinguish truly functional tools from merely descriptive matches. Our comprehensive evaluation on the ToolBench benchmark demonstrates substantial improvements across all metrics: Pass Rate (at 10) increases from 0.690 to 0.826, Recall (at 10) improves from 0.841 to 0.867, and NDCG (at 10) rises from 0.807 to 0.857.. These results establish that execution-based validation provides a more reliable foundation for tool selection than semantic similarity alone, enabling more robust agent performance in real-world applications.

Zuo Zuo, Jiahao Dong, Yanyun Qu et al.

Industrial anomaly detection (AD) plays a significant role in manufacturing where a long-standing challenge is data scarcity. A growing body of works have emerged to address insufficient anomaly data via anomaly generation. However, these anomaly generation methods suffer from lack of fidelity or need to be trained with extra data. To this end, we propose a training-free anomaly generation framework dubbed AAG, which is based on Stable Diffusion (SD)'s strong generation ability for effective anomaly image generation. Given a normal image, mask and a simple text prompt, AAG can generate realistic and natural anomalies in the specific regions and simultaneously keep contents in other regions unchanged. In particular, we propose Cross-Attention Enhancement (CAE) to re-engineer the cross-attention mechanism within Stable Diffusion based on the given mask. CAE increases the similarity between visual tokens in specific regions and text embeddings, which guides these generated visual tokens in accordance with the text description. Besides, generated anomalies need to be more natural and plausible with object in given image. We propose Self-Attention Enhancement (SAE) which improves similarity between each normal visual token and anomaly visual tokens. SAE ensures that generated anomalies are coherent with original pattern. Extensive experiments on MVTec AD and VisA datasets demonstrate effectiveness of AAG in anomaly generation and its utility. Furthermore, anomaly images generated by AAG can bolster performance of various downstream anomaly inspection tasks.

Chao Qu, Shuo Zhu, Yuhang Wang et al.

The fields of imaging in the nighttime dynamic and other extremely dark conditions have seen impressive and transformative advancements in recent years, partly driven by the rise of novel sensing approaches, e.g., near-infrared (NIR) cameras with high sensitivity and event cameras with minimal blur. However, inappropriate exposure ratios of near-infrared cameras make them susceptible to distortion and blur. Event cameras are also highly sensitive to weak signals at night yet prone to interference, often generating substantial noise and significantly degrading observations and analysis. Herein, we develop a new framework for low-light imaging combined with NIR imaging and event-based techniques, named synergistic neuromorphic imaging, which can jointly achieve NIR image deblurring and event denoising. Harnessing cross-modal features of NIR images and visible events via spectral consistency and higher-order interaction, the NIR images and events are simultaneously fused, enhanced, and bootstrapped. Experiments on real and realistically simulated sequences demonstrate the effectiveness of our method and indicate better accuracy and robustness than other methods in practical scenarios. This study gives impetus to enhance both NIR images and events, which paves the way for high-fidelity low-light imaging and neuromorphic reasoning.

Zuo Zuo, Jiahao Dong, Yao Wu et al.

Few-shot anomaly detection methods can effectively address data collecting difficulty in industrial scenarios. Compared to 2D few-shot anomaly detection (2D-FSAD), 3D few-shot anomaly detection (3D-FSAD) is still an unexplored but essential task. In this paper, we propose CLIP3D-AD, an efficient 3D-FSAD method extended on CLIP. We successfully transfer strong generalization ability of CLIP into 3D-FSAD. Specifically, we synthesize anomalous images on given normal images as sample pairs to adapt CLIP for 3D anomaly classification and segmentation. For classification, we introduce an image adapter and a text adapter to fine-tune global visual features and text features. Meanwhile, we propose a coarse-to-fine decoder to fuse and facilitate intermediate multi-layer visual representations of CLIP. To benefit from geometry information of point cloud and eliminate modality and data discrepancy when processed by CLIP, we project and render point cloud to multi-view normal and anomalous images. Then we design multi-view fusion module to fuse features of multi-view images extracted by CLIP which are used to facilitate visual representations for further enhancing vision-language correlation. Extensive experiments demonstrate that our method has a competitive performance of 3D few-shot anomaly classification and segmentation on MVTec-3D AD dataset.

Jie Liu, Yao Wu, Xiaotong Luo et al.

In industrial scenarios, it is crucial not only to identify anomalous items but also to classify the type of anomaly. However, research on anomaly multi-classification remains largely unexplored. This paper proposes a novel and valuable research task called anomaly multi-classification. Given the challenges in applying few-shot learning to this task, due to limited training data and unique characteristics of anomaly images, we introduce a baseline model that combines RelationNet and PatchCore. We propose a data generation method that creates pseudo classes and a corresponding proxy task, aiming to bridge the gap in transferring few-shot learning to industrial scenarios. Furthermore, we utilize contrastive learning to improve the vanilla baseline, achieving much better performance than directly fine-tune a ResNet. Experiments conducted on MvTec AD and MvTec3D AD demonstrate that our approach shows superior performance in this novel task.

Yuval Alaluf, Or Patashnik, Zongze Wu et al.

StyleGAN is arguably one of the most intriguing and well-studied generative models, demonstrating impressive performance in image generation, inversion, and manipulation. In this work, we explore the recent StyleGAN3 architecture, compare it to its predecessor, and investigate its unique advantages, as well as drawbacks. In particular, we demonstrate that while StyleGAN3 can be trained on unaligned data, one can still use aligned data for training, without hindering the ability to generate unaligned imagery. Next, our analysis of the disentanglement of the different latent spaces of StyleGAN3 indicates that the commonly used W/W+ spaces are more entangled than their StyleGAN2 counterparts, underscoring the benefits of using the StyleSpace for fine-grained editing. Considering image inversion, we observe that existing encoder-based techniques struggle when trained on unaligned data. We therefore propose an encoding scheme trained solely on aligned data, yet can still invert unaligned images. Finally, we introduce a novel video inversion and editing workflow that leverages the capabilities of a fine-tuned StyleGAN3 generator to reduce texture sticking and expand the field of view of the edited video.

Zongze Wu, Yotam Nitzan, Eli Shechtman et al.

In this paper, we perform an in-depth study of the properties and applications of aligned generative models. We refer to two models as aligned if they share the same architecture, and one of them (the child) is obtained from the other (the parent) via fine-tuning to another domain, a common practice in transfer learning. Several works already utilize some basic properties of aligned StyleGAN models to perform image-to-image translation. Here, we perform the first detailed exploration of model alignment, also focusing on StyleGAN. First, we empirically analyze aligned models and provide answers to important questions regarding their nature. In particular, we find that the child model's latent spaces are semantically aligned with those of the parent, inheriting incredibly rich semantics, even for distant data domains such as human faces and churches. Second, equipped with this better understanding, we leverage aligned models to solve a diverse set of tasks. In addition to image translation, we demonstrate fully automatic cross-domain image morphing. We further show that zero-shot vision tasks may be performed in the child domain, while relying exclusively on supervision in the parent domain. We demonstrate qualitatively and quantitatively that our approach yields state-of-the-art results, while requiring only simple fine-tuning and inversion.

Or Patashnik, Zongze Wu, Eli Shechtman et al.

Inspired by the ability of StyleGAN to generate highly realistic images in a variety of domains, much recent work has focused on understanding how to use the latent spaces of StyleGAN to manipulate generated and real images. However, discovering semantically meaningful latent manipulations typically involves painstaking human examination of the many degrees of freedom, or an annotated collection of images for each desired manipulation. In this work, we explore leveraging the power of recently introduced Contrastive Language-Image Pre-training (CLIP) models in order to develop a text-based interface for StyleGAN image manipulation that does not require such manual effort. We first introduce an optimization scheme that utilizes a CLIP-based loss to modify an input latent vector in response to a user-provided text prompt. Next, we describe a latent mapper that infers a text-guided latent manipulation step for a given input image, allowing faster and more stable text-based manipulation. Finally, we present a method for mapping a text prompts to input-agnostic directions in StyleGAN's style space, enabling interactive text-driven image manipulation. Extensive results and comparisons demonstrate the effectiveness of our approaches.

Zongze Wu, Dani Lischinski, Eli Shechtman

We explore and analyze the latent style space of StyleGAN2, a state-of-the-art architecture for image generation, using models pretrained on several different datasets. We first show that StyleSpace, the space of channel-wise style parameters, is significantly more disentangled than the other intermediate latent spaces explored by previous works. Next, we describe a method for discovering a large collection of style channels, each of which is shown to control a distinct visual attribute in a highly localized and disentangled manner. Third, we propose a simple method for identifying style channels that control a specific attribute, using a pretrained classifier or a small number of example images. Manipulation of visual attributes via these StyleSpace controls is shown to be better disentangled than via those proposed in previous works. To show this, we make use of a newly proposed Attribute Dependency metric. Finally, we demonstrate the applicability of StyleSpace controls to the manipulation of real images. Our findings pave the way to semantically meaningful and well-disentangled image manipulations via simple and intuitive interfaces.

Xi Liu, Badong Chen, Bin Xu et al.

The unscented transformation (UT) is an efficient method to solve the state estimation problem for a non-linear dynamic system, utilizing a derivative-free higher-order approximation by approximating a Gaussian distribution rather than approximating a non-linear function. Applying the UT to a Kalman filter type estimator leads to the well-known unscented Kalman filter (UKF). Although the UKF works very well in Gaussian noises, its performance may deteriorate significantly when the noises are non-Gaussian, especially when the system is disturbed by some heavy-tailed impulsive noises. To improve the robustness of the UKF against impulsive noises, a new filter for nonlinear systems is proposed in this work, namely the maximum correntropy unscented filter (MCUF). In MCUF, the UT is applied to obtain the prior estimates of the state and covariance matrix, and a robust statistical linearization regression based on the maximum correntropy criterion (MCC) is then used to obtain the posterior estimates of the state and covariance. The satisfying performance of the new algorithm is confirmed by two illustrative examples.