Large language models (LLMs) have shown great potential in enhancing search and recommender systems by providing rich semantic representations from unstructured texts. However, directly integrating LLM embeddings into industrial recommendation pipelines often results in subpar performance due to the semantic and distributional mismatch between pre-trained LLM features and domain-specific, feedback-driven representations. Existing approaches struggle to effectively align LLM embeddings with recommendation objectives, often facing challenges such as label misalignment or the potential loss of semantic diversity during fine-tuning. In this work, we present TreeBridge, a novel framework that introduces a structure-aware generative encoding tree to bridge the semantic gap between LLM embeddings and recommendation tasks. It preserves the external semantic richness of LLM embeddings, while learning label-informed structures that capture user preferences and interaction patterns. This enables the generation of task-adaptive representations without compromising embedding diversity. We further adopt an online-offline hybrid service paradigm to ensure low-latency real-world deployment. TreeBridge has been deployed on the Shopee e-commerce platform, one of the largest online shopping platforms in Southeast Asia serving hundreds of millions of users. Since its deployment in May 2025, it has helped the company achieve a commercially significant 1.55% relative improvement in gross merchandise volume (GMV). The deployment experience demonstrates the effectiveness, scalability, and significant commercial value of TreeBridge.

Denoising Diffusion Probabilistic Model (DDPM) has demonstrated exceptional performance in low-light enhancement task. However, the dependency on paired training datas has left the generality of DDPM in low-light enhancement largely untapped. Therefore, this paper proposes a mutually reinforcing learning framework of decoupled degradation and diffusion enhancement, named MRLIE, which leverages style guidance from unpaired low-light images to generate pseudo-image pairs that are consistent with the target domain, thereby optimizing the latter diffusion enhancement network in a supervised manner. During the degradation process, the diffusion loss of fixed enhancement network serves as a evaluation metric for structure consistency and is combined with adversarial style loss to form the optimization objective for degradation network. Such loss design ensures that scene structure information is retained during the degradation process. During the enhancement process, the degradation network with frozen parameters continuously generates pseudo-paired low-/normal-light image pairs as training datas, thus the diffusion enhancement network could be progressively optimized. On the whole, the two processes are interdependent and could achieve cooperative improvement in terms of degradation realism and enhancement quality through iterative optimization. Additionally, we propose the Retinex-based decoupled degradation strategy for simulating the complex degradation in real low-light imaging, which ensures the color correction and noise suppression capabilities of latter diffusion enhancement network. Extensive experiments show that MRLIE can achieve promising results and better generality across various datasets.

Infrared image nonuniformity correction aims to remove the column-wise stripe noise. Most existing methods just consider stripe noise whereas failing to handle real captured nonuniformity, as directional characteristic of stripe is severely disrupted by random Gaussian noise. Moreover, deep learning-based methods proposed in recent years are blocked by limited receptive field thus cannot accurately distinguish vertical structure and vertical stripes. To address these issues, we propose a universal infrared image nonuniformity correction method based on stripe-aware attention network. We seek to improve the performance of our algorithm by first restoring the damaged stripe directional characteristics, then maximizing the utilization of the prior characteristics. On the one hand, we construct the two-stage framework, in which denoising network is firstly applied to eliminate Gaussian noise and preserve stripes as scene information. As a result, the prior directional characteristics are restored, thereby enhancing the ability of subsequent sub-network to perceive stripe noise. On the other hand, due to the distinct long-range pixel correlations of vertical structures and vertical textures, we introduce a column-wise stripe attention mechanism (CSA) that can capture long-range dependencies of target pixels in the vertical direction. This significantly improves the discriminative ability of algorithm towards vertical structures and stripes, with minimal computational cost. Extensive experiments show that the proposed method can achieve promising results and has better universality for different infrared scenarios.

Low-light image enhancement aims to recover normal-light images from the images captured under dim environments. Most existing methods could just improve the light appearance globally whereas failing to handle other degradation such as dense noise, color offset and extremely low-light. Moreover, unsupervised methods proposed in recent years lack reliable physical model as the basis, thus universality is greatly limited. To address these problems, we propose a novel low-light image enhancement method via Retinex-inline cycle-consistent generative adversarial network named Cycle-Retinex, whose training is totally dependent on unpaired datasets. Specifically, we organically combine Retinex theory with CycleGAN, by which we decouple low-light image enhancement task into two sub-tasks, i.e. illumination map enhancement and reflectance map restoration. Retinex theory helps CycleGAN simplify low-light image enhancement problem and CycleGAN provides synthetic paired images to guide the training of Retinex decomposition network. We further introduce a self-augmented method to address the color distortion and noise problem, thus making the network learn to enhance low-light images adaptively. Extensive experiments show that the proposed method can achieve promising results.

Temporal graphs offer more accurate modeling of many real-world scenarios than static graphs. However, neighbor aggregation, a critical building block of graph networks, for temporal graphs, is currently straightforwardly extended from that of static graphs. It can be computationally expensive when involving all historical neighbors during such aggregation. In practice, typically only a subset of the most recent neighbors are involved. However, such subsampling leads to incomplete and biased neighbor information. To address this limitation, we propose a novel framework for temporal neighbor aggregation that uses the recurrent neural network with node-wise hidden states to integrate information from all historical neighbors for each node to acquire the complete neighbor information. We demonstrate the superior theoretical expressiveness of the proposed framework as well as its state-of-the-art performance in real-world applications. Notably, it achieves a significant +9.4% improvement on averaged precision in a real-world Ecommerce dataset over existing methods on 2-layer models.

In this paper, we propose a novel deep decomposition approach based on Retinex theory for multi-exposure image fusion, termed as DMEF. According to the assumption of Retinex theory, we firstly decompose the source images into illumination and reflection maps by the data-driven decomposition network, among which we introduce the pathwise interaction block that reactivates the deep features lost in one path and embeds them into another path. Therefore, loss of illumination and reflection features during decomposition can be effectively suppressed. And then the high dynamic range illumination map could be obtained by fusing the separated illumination maps in the fusion network. Thus, the reconstructed details in under-exposed and over-exposed regions will be clearer with the help of the fused reflection map which contains complete high-frequency scene information. Finally, the fused illumination and reflection maps are multiplied pixel-by-pixel to obtain the final fused image. Moreover, to retain the discontinuity in the illumination map where gradient of reflection map changes steeply, we introduce the structure-preservation smoothness loss function to retain the structure information and eliminate visual artifacts in these regions. The superiority of our proposed network is demonstrated by applying extensive experiments compared with other state-of-the-art fusion methods subjectively and objectively.

For a natural scene with nonuniform environment light, the captured visible images are always under- or over-exposed because of the limited dynamic range of digital imaging devices. Multi-exposure image fusion (MEF) is a mainstream and effective solution. For a local region that has friendly visual effect in one exposure setting but extremely bad-exposed in another, most existing MEF methods have the ability to transfer the scene detail information to the fused images. However, they will be affected by the over-high or -low light inevitably thus resulting in local visibility reduction. To address this issue, we propose an adaptive clarity evaluation-guided network with illumination correction for MEF in a coarse-to-fine manner, which is termed as ACE-MEF. To be specific, our ACE-MEF is mainly composed of two modules: clarity preservation network (CPN) and illumination adjustment network (IAN). Based on the adaptive clarity evaluation, CPN could be trained to coarsely preserve the environment light and texture details of the clearer regions in source images. Therefore, the need for labeled reference images that are time-consuming to obtain could be mitigated. By measuring the parameter maps of gamma function, IAN is able to refine and correct the local bad-exposed regions so that more details could be further revealed. Extensive experiments demonstrate that our method outperforms multiple state-of-the-art algorithms qualitatively and quantitatively.

In this paper, we propose a novel and practical convolutional neural network method for building footprint generation in remote sensing images, in order to deal with the problem that the detailed information and geometric structure of ground objects in high-resolution images become more abundant, which leads to a large increase in the calculation amount. So we introduce a deepened space module, which can ignore the channels with weak target features and emphasize the effective features. It is embedded in each splicing layer in the upsampling process of U-net to achieve the effect of feature selection. By means of clipping and data enhancement, we carry out iterative training and model optimization learning on Inria aerial image label dataset, and realize the automatic generation of building footprint. Compared with FCN8s, Unet, SegNet, PSPNet, Deeplabv3 + and GLNet, experimental results show that the method we use to generate building footprint is more accurate, and in IoU, mPA, PA three indicators are better than the comparison algorithms.

In view of the difficulty and low accuracy of small object detection in remote sensing images, this paper proposes a bidirectional cross-scale connection feature fusion network with an information direct connection layer and a shallow information fusion layer. Aiming at the problem that the detection targets in remote sensing images are mainly small and medium-sized targets, we fuse the shallow feature maps with rich spatial information in the bidirectional cross-scale connection feature fusion network instead of directly using the shallow feature maps for regression and classification. While ensuring the model inference speed, the detection accuracy of small objects is improved. At the same time, we use the information direct connection layer to perform feature fusion with the initial information in each iteration of the bidirectional cross-scale connection feature fusion pyramid to prevent the loss of small object information. Experimental results show that the algorithm proposed in this paper can obtain good accuracy and real-time performance on the NWPU VHR-10 dataset.

In this paper, in order to enhance the infrared target in infrared image and retain the edge and detail information in visible image, we propose a multi-scale decomposition fusion method based on phase congruency and saliency. In this method, the Laplacian pyramid is first used to decompose the source image into detail layers and base layers. Secondly, we use a method based on phase congruency for the fusion of detail layers. Thirdly, for the base layer, we decompose it into saliency map and residual map. The “max absolute” rule and “averag” rule are adopted for the fusion of saliency map and residual map, then the fused saliency map and residual map are added to attain the fused base image. Finally, we use the inverse transform of Laplacian pyramid to reconstruct the fused image. The experimental results show that the proposed method have better fusion effect than other methods. What's outstanding is that the infrared targets in the fused image are enhanced and abundant edges are preserved.