Weisheng Li

Da Guo, Shijia Wang, Qiang Xiao et al.

Generative retrieval has emerged as a promising paradigm in recommender systems, offering superior sequence modeling capabilities over traditional dual-tower architectures. However, in large-scale industrial scenarios, such models often suffer from inherent myopia: due to single-step inference and strict latency constraints, they tend to collapse diverse user intents into locally optimal predictions, failing to capture long-horizon and multi-item consumption patterns. Moreover, real-world retrieval systems must follow explicit retrieval instructions, such as category-level control and policy constraints. Incorporating such instruction-following behavior into generative retrieval remains challenging, as existing conditioning or post-hoc filtering approaches often compromise relevance or efficiency. In this work, we present Climber-Pilot, a unified generative retrieval framework to address both limitations. First, we introduce Time-Aware Multi-Item Prediction (TAMIP), a novel training paradigm designed to mitigate inherent myopia in generative retrieval. By distilling long-horizon, multi-item foresight into model parameters through time-aware masking, TAMIP alleviates locally optimal predictions while preserving efficient single-step inference. Second, to support flexible instruction-following retrieval, we propose Condition-Guided Sparse Attention (CGSA), which incorporates business constraints directly into the generative process via sparse attention, without introducing additional inference steps. Extensive offline experiments and online A/B testing at NetEase Cloud Music, one of the largest music streaming platforms, demonstrate that Climber-Pilot significantly outperforms state-of-the-art baselines, achieving a 4.24\% lift of the core business metric.

Xiuli Bi, Bo Liu, Fan Yang et al.

As deep learning technology continues to evolve, the images yielded by generative models are becoming more and more realistic, triggering people to question the authenticity of images. Existing generated image detection methods detect visual artifacts in generated images or learn discriminative features from both real and generated images by massive training. This learning paradigm will result in efficiency and generalization issues, making detection methods always lag behind generation methods. This paper approaches the generated image detection problem from a new perspective: Start from real images. By finding the commonality of real images and mapping them to a dense subspace in feature space, the goal is that generated images, regardless of their generative model, are then projected outside the subspace. As a result, images from different generative models can be detected, solving some long-existing problems in the field. Experimental results show that although our method was trained only by real images and uses 99.9\% less training data than other deep learning-based methods, it can compete with state-of-the-art methods and shows excellent performance in detecting emerging generative models with high inference efficiency. Moreover, the proposed method shows robustness against various post-processing. These advantages allow the method to be used in real-world scenarios.

Weisheng Li, Lin Huang

We analyzed the network structure of real-time object detection models and found that the features in the feature concatenation stage are very rich. Applying an attention module here can effectively improve the detection accuracy of the model. However, the commonly used attention module or self-attention module shows poor performance in detection accuracy and inference efficiency. Therefore, we propose a novel self-attention module, called 2D local feature superimposed self-attention, for the feature concatenation stage of the neck network. This self-attention module reflects global features through local features and local receptive fields. We also propose and optimize an efficient decoupled head and AB-OTA, and achieve SOTA results. Average precisions of 49.0% (71FPS, 14ms), 46.1% (85FPS, 11.7ms), and 39.1% (107FPS, 9.3ms) were obtained for large, medium, and small-scale models built using our proposed improvements. Our models exceeded YOLOv5 by 0.8% -- 3.1% in average precision.

Lin Huang, Weisheng Li, Yujuan Tan et al.

In this study, the structural problems of the YOLOv5 model were analyzed emphatically. Based on the characteristics of fine defects in artificial leather, four innovative structures, namely DFP, IFF, AMP, and EOS, were designed. These advancements led to the proposal of a high-performance artificial leather fine defect detection model named YOLOD. YOLOD demonstrated outstanding performance on the artificial leather defect dataset, achieving an impressive increase of 11.7% - 13.5% in AP_50 compared to YOLOv5, along with a significant reduction of 5.2% - 7.2% in the error detection rate. Moreover, YOLOD also exhibited remarkable performance on the general MS-COCO dataset, with an increase of 0.4% - 2.6% in AP compared to YOLOv5, and a rise of 2.5% - 4.1% in AP_S compared to YOLOv5. These results demonstrate the superiority of YOLOD in both artificial leather defect detection and general object detection tasks, making it a highly efficient and effective model for real-world applications.

Liming Xu, Hanqi Li, Bochuan Zheng et al.

Hashing methods have made significant progress in cross-modal retrieval tasks with fast query speed and low storage cost. Among them, deep learning-based hashing achieves better performance on large-scale data due to its excellent extraction and representation ability for nonlinear heterogeneous features. However, there are still two main challenges in catastrophic forgetting when data with new categories arrive continuously, and time-consuming for non-continuous hashing retrieval to retrain for updating. To this end, we, in this paper, propose a novel deep lifelong cross-modal hashing to achieve lifelong hashing retrieval instead of re-training hash function repeatedly when new data arrive. Specifically, we design lifelong learning strategy to update hash functions by directly training the incremental data instead of retraining new hash functions using all the accumulated data, which significantly reduce training time. Then, we propose lifelong hashing loss to enable original hash codes participate in lifelong learning but remain invariant, and further preserve the similarity and dis-similarity among original and incremental hash codes to maintain performance. Additionally, considering distribution heterogeneity when new data arriving continuously, we introduce multi-label semantic similarity to supervise hash learning, and it has been proven that the similarity improves performance with detailed analysis. Experimental results on benchmark datasets show that the proposed methods achieves comparative performance comparing with recent state-of-the-art cross-modal hashing methods, and it yields substantial average increments over 20\% in retrieval accuracy and almost reduces over 80\% training time when new data arrives continuously.

Dan He, Weisheng Li, Guofen Wang et al.

Multimodal medical image fusion (MMIF) extracts the most meaningful information from multiple source images, enabling a more comprehensive and accurate diagnosis. Achieving high-quality fusion results requires a careful balance of brightness, color, contrast, and detail; this ensures that the fused images effectively display relevant anatomical structures and reflect the functional status of the tissues. However, existing MMIF methods have limited capacity to capture detailed features during conventional training and suffer from insufficient cross-modal feature interaction, leading to suboptimal fused image quality. To address these issues, this study proposes a two-stage diffusion model-based fusion network (DM-FNet) to achieve unified MMIF. In Stage I, a diffusion process trains UNet for image reconstruction. UNet captures detailed information through progressive denoising and represents multilevel data, providing a rich set of feature representations for the subsequent fusion network. In Stage II, noisy images at various steps are input into the fusion network to enhance the model's feature recognition capability. Three key fusion modules are also integrated to process medical images from different modalities adaptively. Ultimately, the robust network structure and a hybrid loss function are integrated to harmonize the fused image's brightness, color, contrast, and detail, enhancing its quality and information density. The experimental results across various medical image types demonstrate that the proposed method performs exceptionally well regarding objective evaluation metrics. The fused image preserves appropriate brightness, a comprehensive distribution of radioactive tracers, rich textures, and clear edges. The code is available at https://github.com/HeDan-11/DM-FNet.

Dan He, Guofen Wang, Weisheng Li et al.

Multimodal image fusion (MMIF) aims to integrate information from different modalities to obtain a comprehensive image, aiding downstream tasks. However, existing methods tend to prioritize natural image fusion and focus on information complementary and network training strategies. They ignore the essential distinction between natural and medical image fusion and the influence of underlying components. This paper dissects the significant differences between the two tasks regarding fusion goals, statistical properties, and data distribution. Based on this, we rethink the suitability of the normalization strategy and convolutional kernels for end-to-end MMIF.Specifically, this paper proposes a mixture of instance normalization and group normalization to preserve sample independence and reinforce intrinsic feature correlation.This strategy promotes the potential of enriching feature maps, thus boosting fusion performance. To this end, we further introduce the large kernel convolution, effectively expanding receptive fields and enhancing the preservation of image detail. Moreover, the proposed multipath adaptive fusion module recalibrates the decoder input with features of various scales and receptive fields, ensuring the transmission of crucial information. Extensive experiments demonstrate that our method exhibits state-of-the-art performance in multiple fusion tasks and significantly improves downstream applications. The code is available at https://github.com/HeDan-11/LKC-FUNet.

Xiuli Bi, Jian Lu, Bo Liu et al.

Benefiting from large-scale pre-training of text-video pairs, current text-to-video (T2V) diffusion models can generate high-quality videos from the text description. Besides, given some reference images or videos, the parameter-efficient fine-tuning method, i.e. LoRA, can generate high-quality customized concepts, e.g., the specific subject or the motions from a reference video. However, combining the trained multiple concepts from different references into a single network shows obvious artifacts. To this end, we propose CustomTTT, where we can joint custom the appearance and the motion of the given video easily. In detail, we first analyze the prompt influence in the current video diffusion model and find the LoRAs are only needed for the specific layers for appearance and motion customization. Besides, since each LoRA is trained individually, we propose a novel test-time training technique to update parameters after combination utilizing the trained customized models. We conduct detailed experiments to verify the effectiveness of the proposed methods. Our method outperforms several state-of-the-art works in both qualitative and quantitative evaluations.

Lin Huang, Yujuan Tan, Weisheng Li et al.

One-stage object detection, particularly the YOLO series, strikes a favorable balance between accuracy and efficiency. However, existing YOLO detectors lack explicit modeling of heterogeneous object responses within shared feature channels, which limits further performance gains. To address this, we propose YOLO-DS, a framework built around a novel Dual-Statistic Synergy Operator (DSO). The DSO decouples object features by jointly modeling the channel-wise mean and the peak-to-mean difference. Building upon the DSO, we design two lightweight gating modules: the Dual-Statistic Synergy Gating (DSG) module for adaptive channel-wise feature selection, and the Multi-Path Segmented Gating (MSG) module for depth-wise feature weighting. On the MS-COCO benchmark, YOLO-DS consistently outperforms YOLOv8 across five model scales (N, S, M, L, X), achieving AP gains of 1.1% to 1.7% with only a minimal increase in inference latency. Extensive visualization, ablation, and comparative studies validate the effectiveness of our approach, demonstrating its superior capability in discriminating heterogeneous objects with high efficiency.

Xiaoyu Qiao, Weisheng Li, Bin Xiao et al.

Score matching with Langevin dynamics (SMLD) method has been successfully applied to accelerated MRI. However, the hyperparameters in the sampling process require subtle tuning, otherwise the results can be severely corrupted by hallucination artifacts, especially with out-of-distribution test data. To address the limitations, we proposed a novel workflow where naive SMLD samples serve as additional priors to guide model-driven network training. First, we adopted a pretrained score network to generate samples as preliminary guidance images (PGI), obviating the need for network retraining, parameter tuning and in-distribution test data. Although PGIs are corrupted by hallucination artifacts, we believe they can provide extra information through effective denoising steps to facilitate reconstruction. Therefore, we designed a denoising module (DM) in the second step to coarsely eliminate artifacts and noises from PGIs. The features are extracted from a score-based information extractor (SIE) and a cross-domain information extractor (CIE), which directly map to the noise patterns. Third, we designed a model-driven network guided by denoised PGIs (DGIs) to further recover fine details. DGIs are densely connected with intermediate reconstructions in each cascade to enrich the information and are periodically updated to provide more accurate guidance. Our experiments on different datasets reveal that despite the low average quality of PGIs, the proposed workflow can effectively extract valuable information to guide the network training, even with severely reduced training data and sampling steps. Our method outperforms other cutting-edge techniques by effectively mitigating hallucination artifacts, yielding robust and high-quality reconstruction results.

Xiaoyu Qiao, Weisheng Li, Guofen Wang et al.

Deep unfolding networks (DUNs) have demonstrated significant potential in accelerating magnetic resonance imaging (MRI). However, they often encounter high computational costs and slow convergence rates. Besides, they struggle to fully exploit the complementarity when incorporating multiple priors. In this study, we propose a multi-prior collaborative DUN, termed MCU-Net, to address these limitations. Our method features a parallel structure consisting of different optimization-inspired subnetworks based on low-rank and sparsity, respectively. We design a gates-controlled spatial attention module (GSAM), evaluating the relative confidence (RC) and overall confidence (OC) maps for intermediate reconstructions produced by different subnetworks. RC allocates greater weights to the image regions where each subnetwork excels, enabling precise element-wise collaboration. We design correction modules to enhance the effectiveness in regions where both subnetworks exhibit limited performance, as indicated by low OC values, thereby obviating the need for additional branches. The gate units within GSAMs are designed to preserve necessary information across multiple iterations, improving the accuracy of the learned confidence maps and enhancing robustness against accumulated errors. Experimental results on multiple datasets show significant improvements on PSNR and SSIM results with relatively low FLOPs compared to cutting-edge methods. Additionally, the proposed strategy can be conveniently applied to various DUN structures to enhance their performance.

Mengjingcheng Mo, Xinyang Tong, Jiaxu Leng et al.

While unmanned aerial vehicles (UAVs) offer wide-area, high-altitude coverage for anomaly detection, they face challenges such as dynamic viewpoints, scale variations, and complex scenes. Existing datasets and methods, mainly designed for fixed ground-level views, struggle to adapt to these conditions, leading to significant performance drops in drone-view scenarios. To bridge this gap, we introduce A2Seek (Aerial Anomaly Seek), a large-scale, reasoning-centric benchmark dataset for aerial anomaly understanding. This dataset covers various scenarios and environmental conditions, providing high-resolution real-world aerial videos with detailed annotations, including anomaly categories, frame-level timestamps, region-level bounding boxes, and natural language explanations for causal reasoning. Building on this dataset, we propose A2Seek-R1, a novel reasoning framework that generalizes R1-style strategies to aerial anomaly understanding, enabling a deeper understanding of "Where" anomalies occur and "Why" they happen in aerial frames. To this end, A2Seek-R1 first employs a graph-of-thought (GoT)-guided supervised fine-tuning approach to activate the model's latent reasoning capabilities on A2Seek. Then, we introduce Aerial Group Relative Policy Optimization (A-GRPO) to design rule-based reward functions tailored to aerial scenarios. Furthermore, we propose a novel "seeking" mechanism that simulates UAV flight behavior by directing the model's attention to informative regions. Extensive experiments demonstrate that A2Seek-R1 achieves up to a 22.04% improvement in AP for prediction accuracy and a 13.9% gain in mIoU for anomaly localization, exhibiting strong generalization across complex environments and out-of-distribution scenarios. Our dataset and code will be released at https://hayneyday.github.io/A2Seek/.

Lin Huang, Yujuan Tan, Weisheng Li et al.

This paper addresses the inherent limitations of conventional bottleneck structures (diminished instance discriminability due to overemphasis on batch statistics) and decoupled heads (computational redundancy) in object detection frameworks by proposing two novel modules: the Instance-Specific Bottleneck with full-channel global self-attention (ISB) and the Instance-Specific Asymmetric Decoupled Head (ISADH). The ISB module innovatively reconstructs feature maps to establish an efficient full-channel global attention mechanism through synergistic fusion of batch-statistical and instance-specific features. Complementing this, the ISADH module pioneers an asymmetric decoupled architecture enabling hierarchical multi-dimensional feature integration via dual-stream batch-instance representation fusion. Extensive experiments on the MS-COCO benchmark demonstrate that the coordinated deployment of ISB and ISADH in the YOLO-PRO framework achieves state-of-the-art performance across all computational scales. Specifically, YOLO-PRO surpasses YOLOv8 by 1.0-1.6% AP (N/S/M/L/X scales) and outperforms YOLO11 by 0.1-0.5% AP in critical N/M/L/X groups, while maintaining competitive computational efficiency. This work provides practical insights for developing high-precision detectors deployable on edge devices.

Lin Huang, Weisheng Li, Yujuan Tan et al.

In this study, we examine the associations between channel features and convolutional kernels during the processes of feature purification and gradient backpropagation, with a focus on the forward and backward propagation within the network. Consequently, we propose a method called Dense Channel Compression for Feature Spatial Solidification. Drawing upon the central concept of this method, we introduce two innovative modules for backbone and head networks: the Dense Channel Compression for Feature Spatial Solidification Structure (DCFS) and the Asymmetric Multi-Level Compression Decoupled Head (ADH). When integrated into the YOLOv5 model, these two modules demonstrate exceptional performance, resulting in a modified model referred to as YOLOCS. Evaluated on the MSCOCO dataset, the large, medium, and small YOLOCS models yield AP of 50.1%, 47.6%, and 42.5%, respectively. Maintaining inference speeds remarkably similar to those of the YOLOv5 model, the large, medium, and small YOLOCS models surpass the YOLOv5 model's AP by 1.1%, 2.3%, and 5.2%, respectively.

Lei Li, Fuping Wu, Sihan Wang et al.

Assessment of myocardial viability is essential in diagnosis and treatment management of patients suffering from myocardial infarction, and classification of pathology on myocardium is the key to this assessment. This work defines a new task of medical image analysis, i.e., to perform myocardial pathology segmentation (MyoPS) combining three-sequence cardiac magnetic resonance (CMR) images, which was first proposed in the MyoPS challenge, in conjunction with MICCAI 2020. The challenge provided 45 paired and pre-aligned CMR images, allowing algorithms to combine the complementary information from the three CMR sequences for pathology segmentation. In this article, we provide details of the challenge, survey the works from fifteen participants and interpret their methods according to five aspects, i.e., preprocessing, data augmentation, learning strategy, model architecture and post-processing. In addition, we analyze the results with respect to different factors, in order to examine the key obstacles and explore potential of solutions, as well as to provide a benchmark for future research. We conclude that while promising results have been reported, the research is still in the early stage, and more in-depth exploration is needed before a successful application to the clinics. Note that MyoPS data and evaluation tool continue to be publicly available upon registration via its homepage (www.sdspeople.fudan.edu.cn/zhuangxiahai/0/myops20/).

Hang Zhang, Yeyun Gong, Yelong Shen et al.

In this paper, we introduce a two-level attention schema, Poolingformer, for long document modeling. Its first level uses a smaller sliding window pattern to aggregate information from neighbors. Its second level employs a larger window to increase receptive fields with pooling attention to reduce both computational cost and memory consumption. We first evaluate Poolingformer on two long sequence QA tasks: the monolingual NQ and the multilingual TyDi QA. Experimental results show that Poolingformer sits atop three official leaderboards measured by F1, outperforming previous state-of-the-art models by 1.9 points (79.8 vs. 77.9) on NQ long answer, 1.9 points (79.5 vs. 77.6) on TyDi QA passage answer, and 1.6 points (67.6 vs. 66.0) on TyDi QA minimal answer. We further evaluate Poolingformer on a long sequence summarization task. Experimental results on the arXiv benchmark continue to demonstrate its superior performance.

Bin Xiao, Tao Geng, Xiuli Bi et al.

Local binary pattern (LBP) as a kind of local feature has shown its simplicity, easy implementation and strong discriminating power in image recognition. Although some LBP variants are specifically investigated for color image recognition, the color information of images is not adequately considered and the curse of dimensionality in classification is easily caused in these methods. In this paper, a color-related local binary pattern (cLBP) which learns the dominant patterns from the decoded LBP is proposed for color images recognition. This paper first proposes a relative similarity space (RSS) that represents the color similarity between image channels for describing a color image. Then, the decoded LBP which can mine the correlation information between the LBP feature maps correspond to each color channel of RSS traditional RGB spaces, is employed for feature extraction. Finally, a feature learning strategy is employed to learn the dominant color-related patterns for reducing the dimension of feature vector and further improving the discriminatively of features. The theoretic analysis show that the proposed RSS can provide more discriminative information, and has higher noise robustness as well as higher illumination variation robustness than traditional RGB space. Experimental results on four groups, totally twelve public color image datasets show that the proposed method outperforms most of the LBP variants for color image recognition in terms of dimension of features, recognition accuracy under noise-free, noisy and illumination variation conditions.

Bo Liu, Ranglei Wu, Xiuli Bi et al.

Recently, many detection methods based on convolutional neural networks (CNNs) have been proposed for image splicing forgery detection. Most of these detection methods focus on the local patches or local objects. In fact, image splicing forgery detection is a global binary classification task that distinguishes the tampered and non-tampered regions by image fingerprints. However, some specific image contents are hardly retained by CNN-based detection networks, but if included, would improve the detection accuracy of the networks. To resolve these issues, we propose a novel network called dual-encoder U-Net (D-Unet) for image splicing forgery detection, which employs an unfixed encoder and a fixed encoder. The unfixed encoder autonomously learns the image fingerprints that differentiate between the tampered and non-tampered regions, whereas the fixed encoder intentionally provides the direction information that assists the learning and detection of the network. This dual-encoder is followed by a spatial pyramid global-feature extraction module that expands the global insight of D-Unet for classifying the tampered and non-tampered regions more accurately. In an experimental comparison study of D-Unet and state-of-the-art methods, D-Unet outperformed the other methods in image-level and pixel-level detection, without requiring pre-training or training on a large number of forgery images. Moreover, it was stably robust to different attacks.