论文检索

Hongyang Liu, Zhu Sun, Tianjun Wei, Yan Wang, Jiajie Zhu, Xinghua Qu

Recent advances in large language models (LLMs) have enabled realistic user simulators for developing and evaluating recommender systems (RSs). However, existing LLM-based simulators for RSs face two major limitations: (1) static and single-step prompt-based inference that leads to inaccurate and incomplete user profile construction; (2) unrealistic and single-round recommendation-feedback interaction pattern that fails to capture real-world scenarios. To address these limitations, we propose DGDPO (Diagnostic-Guided Dynamic Profile Optimization), a novel framework that constructs user profile through a dynamic and iterative optimization process to enhance the simulation fidelity. Specifically, DGDPO incorporates two core modules within each optimization loop: firstly, a specialized LLM-based diagnostic module, calibrated through our novel training strategy, accurately identifies specific defects in the user profile. Subsequently, a generalized LLM-based treatment module analyzes the diagnosed defect and generates targeted suggestions to refine the profile. Furthermore, unlike existing LLM-based user simulators that are limited to single-round interactions, we are the first to integrate DGDPO with sequential recommenders, enabling a bidirectional evolution where user profiles and recommendation strategies adapt to each other over multi-round interactions. Extensive experiments conducted on three real-world datasets demonstrate the effectiveness of our proposed framework.

Emily Liu, Kuan Han, Minfeng Zhan, Bocheng Zhao, Guanyu Mu, Yang Song

Watch time is widely used as a proxy for user satisfaction in video recommendation platforms. However, raw watch times are influenced by confounding factors such as video duration, popularity, and individual user behaviors, potentially distorting preference signals and resulting in biased recommendation models. We propose a novel relative advantage debiasing framework that corrects watch time by comparing it to empirically derived reference distributions conditioned on user and item groups. This approach yields a quantile-based preference signal and introduces a two-stage architecture that explicitly separates distribution estimation from preference learning. Additionally, we present distributional embeddings to efficiently parameterize watch-time quantiles without requiring online sampling or storage of historical data. Both offline and online experiments demonstrate significant improvements in recommendation accuracy and robustness compared to existing baseline methods.

Aoyu Liu, Yaying Zhang

As urban data expands, existing spatio-temporal models encounter challenges such as high context dependency, poor cross-scenario generalization, and inefficient computational performance. To address these issues, we propose UrbanPG, an efficient and scalable framework for spatio-temporal learning. UrbanPG separates task-specific personalized patterns from general patterns, enabling unified spatio-temporal modeling and efficient knowledge generalization across scenarios. The key innovations of UrbanPG include: the development of a lightweight, context-independent general backbone utilizing linear spatio-temporal attention for scalable cross-scenario deployment; a personalized context prompt mechanism designed to model heterogeneity through spatio-temporal embeddings and random perturbation regularization, interacting with the backbone to enhance pattern differentiation; the proposal of multi spatio-temporal learning paradigms for rapid knowledge transfer and generalization to downstream tasks through fine-tuning personalized context prompts while freezing the backbone. Experimental results demonstrate that UrbanPG achieves state-of-the-art performance in large-scale forecasting, few-shot transfer, and continual learning tasks across eight real-world datasets, showcasing exceptional performance, strong generalization, and significant reductions in computational overhead.

Run Ling, Wenji Wang, Yuting Liu, Guibing Guo, Haowei Liu, Jian Lu, Quanwei Zhang, Yexing Xu, Shuo Lu, Yun Wang 等

Personalized image generation is crucial for improving the user experience, as it renders reference images into preferred ones according to user visual preferences. Although effective, existing methods face two main issues. First, existing methods treat all items in the user's historical sequence equally when extracting user preferences, overlooking the varying semantic similarities between historical items and the reference item. Disproportionately high weights for low-similarity items distort user visual preferences for the reference item. Second, existing methods heavily rely on consistency between generated and reference images to optimize generation, which leads to underfitting user preferences and hinders personalization. To address these issues, we propose Retrieval Augmented Personalized Image GenerAtion guided by Recommendation (RAGAR). Our approach uses a retrieval mechanism to assign different weights to historical items according to their similarities to the reference item, thereby extracting more refined users' visual preferences for the reference item. Then we introduce a novel rank task based on the multi-modal ranking model to optimize the personalization of the generated images instead of forcing depend on consistency. Extensive experiments and human evaluations on three real-world datasets demonstrate that RAGAR achieves significant improvements in both personalization and semantic metrics compared to five baselines.

Yuxi Lin, Yongkang Li, Jie Xing, Zipei Fan

Among the diverse services provided by Location-Based Social Networks (LBSNs), Next Point-of-Interest (POI) recommendation plays a crucial role in inferring user preferences from historical check-in trajectories. However, existing sequential and graph-based methods frequently neglect significant mobility variations across distinct contextual scenarios (e.g., tourists versus locals). This oversight results in suboptimal performance due to two fundamental limitations: the inability to capture scenario-specific features and the failure to resolve inherent inter-scenario conflicts. To overcome these limitations, we propose the Multifaceted Scenario-Aware Hypergraph Learning method (MSAHG), a framework that adopts a scenario-splitting paradigm for next POI recommendation. Our main contributions are: (1) Construction of scenario-specific, multi-view disentangled sub-hypergraphs to capture distinct mobility patterns; (2) A parameter-splitting mechanism to adaptively resolve conflicting optimization directions across scenarios while preserving generalization capability. Extensive experiments on three real-world datasets demonstrate that MSAHG consistently outperforms five state-of-the-art methods across diverse scenarios, confirming its effectiveness in multi-scenario POI recommendation.

Yingde Lin, Yuanbo Xu, Lu Jiang, Pengyang Wang

Urban region embedding, which learns dense vector representations for urban zones, plays a foundational role in data-driven urban intelligence. These representations are critical for downstream applications like public safety management and infrastructure development, requiring nuanced understanding of urban functionality. A core challenge remains effective fusion of multi-view data (e.g., human mobility flows and static regional attributes) into unified zone representations. To this end, we propose MVJC, a Multi-view Joint Learning and Contrastive Learning framework, which employs: (1) Multi-view Joint Learning (MVJL) layer to model intra-view dependencies to extract view-specific features and (2) Multi-view Contrastive Learning (MVCL) layer to perform cross-region aggregation to derive consensus representations while capturing the regional complementarity. We further introduce a structure-aware contrastive loss that mitigates false negatives by aligning representations through region topology instead of instance identity. Extensive experiments on New York City datasets demonstrate MVJC’s superiority: it reduces crime prediction MAE by 9.1% (vs. 66.9 baseline) and improves land use clustering F-measure by 55.6% (vs. 0.45 baseline) over state-of-the-art method, which is attributed to MVJC’s synergy of joint and contrastive learning, yielding representations that are simultaneously predictive and semantically discriminative.

Mingkai Lin, Yinke Wang, Xiaobin Hong, Wenzhong Li

Multivariate motif discovery aims to identify frequently occurring subsequences within multi-dimensional time series, which is a critical machine learning task with wide applications. However, previous motif discovery algorithms often miss complex multivariate motifs and struggle with high computational costs as data scale and dimensionality grow. We propose a novel learnable multivariate matrix profile method (L-MAP) that captures inter-dimensional dependencies for comprehensive analysis of multivariate time series. The time series is partitioned into subsequences using the Fourier transform in the frequency domain, with locality-sensitive hashing (LSH) assigning them to buckets based on distinct patterns. Each subsequence is modeled as a graph for multivariate fusion, where triplet learning is used to capture cross-dimensional relationships and form graph embeddings. Unlike prior methods relying on Euclidean distance modeling, our graph-based approach computes all-pairs similarity in a latent space, which constructs the multivariate matrix profile from distributions formed by embedding clusters. Extensive experiments on multivariate datasets from diverse domains demonstrate that L-MAP outperforms state-of-the-art methods in motif discovery, offering superior quality, diversity, and scalability efficiency.

Chenxiao Lin, Ye Luo, KunHong Liu, Qingqiang Wu

Retrieval-augmented generation (RAG) enhances the reasoning capabilities of large language models (LLMs) by incorporating external knowledge. Among available sources, knowledge graphs (KGs) offer a structured and reliable foundation for factual information, making them increasingly popular in efforts to improve reasoning faithfulness in RAG. Most existing KG-based RAG methods rely on LLMs to extract knowledge from KGs. However, these approaches often require costly fine-tuning and struggle to navigate deep graph structures, limiting their effectiveness in multi-hop reasoning tasks. To address these challenges, we propose Stepwise Contrastive Reasoning (SCR), a lightweight framework that integrates graph structure and textual context for efficient and interpretable RAG over KGs. SCR combines relational message passing layers to encode KG entities with a Transformer encoder for processing question text. It decomposes reasoning into a series of alignment steps. At each step, SCR compares the current topic entity and its neighbors with the question representation, selecting the most relevant entity as the next topic entity. The question is then updated with this entity's textual description. This process continues until the selected entity no longer changes, indicating that the answer entity has been reached. Through stepwise alignment, SCR enables compact models to perform faithful and interpretable reasoning over large-scale KGs. Extensive experiments on several widely used KGQA benchmarks demonstrate that SCR not only achieves state-of-the-art performance but also effectively boosts the capabilities of smaller language models to match those of LLMs.

Jingyi Liao, Yongyi Su, Rong-Cheng Tu, Zhao Jin, Wenhao Sun, Yiting Li, Xun Xu, Dacheng Tao, Xulei Yang

While Multimodal Large Language Models (MLLMs) demonstrate remarkable capabilities across diverse domains, their application to specialized anomaly detection (AD) remains constrained by domain adaptation challenges. Existing Group Relative Policy Optimization (GRPO) based approaches suffer from two critical limitations: inadequate training data utilization when models produce uniform responses, and insufficient supervision over reasoning processes that encourage immediate binary decisions without deliberative analysis. We propose a comprehensive framework addressing these limitations through two synergistic innovations. First, we introduce a multi-stage deliberative reasoning process that guides models from region identification to focused examination, generating diverse response patterns essential for GRPO optimization while enabling structured supervision over analytical workflows. Second, we develop a fine-grained reward mechanism incorporating classification accuracy and localization supervision, transforming binary feedback into continuous signals that distinguish genuine analytical insight from spurious correctness. Comprehensive evaluation across multiple industrial datasets shows that our method achieves superior accuracy by enabling general-purpose MLLMs to acquire fine-grained visual discrimination for detecting subtle manufacturing defects.

Yahong Lian, Haotian Tian, Chunyao Song, Tingjian Ge

A multimodal recommendation system (MRS), which leverages rich multimodal information to model user preferences, has recently attracted significant research interest. Most existing MRSs focus primarily on developing sophisticated encoders for feature extraction, typically relying on simple aggregation of interaction-based features for final predictions. However, this conventional paradigm fails to account for the critical semantic difference between high- and low-rating interactions: while high ratings indicate user preference, low ratings explicitly convey dissatisfaction. Such oversight of negative feedback semantics may significantly limit the system’s recommendation performance. Recently, sign graphs—which model positive and negative feedback signals separately—have gained considerable attention. Inspired by this approach, we propose Sign-Aware Multimodal Graph Recommendation (SiMGR), a novel framework incorporating signed graphs into multimodal recommendation systems. SiMGR fuses multimodal features with signed interactions in a unified graph framework by integrating modality-specific representations and applying user-level thresholds to separate positive and negative subgraphs. A balanced pseudo-edge augmentation strategy is introduced to alleviate sparsity and enhance generalization. Experiments on three public multimodal recommendation datasets show that SiMGR outperforms state-of-the-art baselines, achieving an average 4.28% improvement in NDCG@20.

Jie Lian, Zhihao Wu, Jielong Lu, Jiajun Yu, Qianqian Shen, Haishuai Wang

Graph anomaly detection is emerging as a critical technology for addressing increasingly complex and dynamic risk environments. Although unsupervised graph anomaly detection has advanced under the graph representation learning, directly applying these paradigms remains fundamentally misaligned with anomaly detection objectives. In this work, we highlight two key insights: graph neural networks are often suboptimal as feature extractors due to neighborhood aggregation diluting anomaly signals, and reliance on local inconsistency mining is inadequate for comprehensive anomaly detection, as it often fails to identify anomalies hidden within camouflaged communities. Based on these insights, we propose multiscale inconsistency learning for graph anomaly detection (MI-GAD), a novel framework that integrates both local and global anomaly signals. Specifically, individual node representations are projected onto a common hypersphere to ensure uniformity. At the local scale, the graph structure is leveraged for affinity-aware modeling via group discrimination. At the global scale, we introduce node deviation, a metric that distinguishes anomalies by optimizing representation centers. This unified approach enables robust and comprehensive detection of diverse graph anomalies. Experiments on seven real datasets demonstrate that our method consistently outperforms state-of-the-art baselines in both effectiveness and scalability.

Zuqing Li, Junhao Gan, Jianzhong Qi

Diffusion-based tabular data synthesis models have yielded promising results. However, we observe that when the data dimensionality increases, existing models tend to degenerate and may perform even worse than simpler, non-diffusion-based models. This is because limited training samples in high-dimensional space often hinder generative models from capturing the distribution accurately. To mitigate the insufficient learning signals and to stabilize training under such conditions, we propose CtrTab, a condition-controlled diffusion model that injects perturbed ground-truth samples as auxiliary inputs during training. This design introduces an implicit $L_2$ regularization on the model’s sensitivity to the control signal, improving robustness and stability in high-dimensional, low-data scenarios. Experimental results across multiple datasets show that CtrTab outperforms state-of-the-art models, with a performance gap in accuracy over 90% on average.

Zilong Li, Jia Zhu, Chenglei Huang, Zhangze Chen, Hanghui Guo, Guoqing Ma, Jianxia Ling

Multimodal sequential recommender systems leverage diverse modal inputs to enhance the accuracy and relevance of personalized recommendations. However, existing fusion strategies often struggle to capture intricate cross-modal interactions, especially under the evolving dynamics of user intent. Moreover, they frequently neglect modality imbalance issues, leading to suboptimal utilization of multimodal information. To address these challenges, we propose DuAF-MAT, a novel framework for robust multimodal sequential recommendation. Our approach consists of three key components: (1) a Dual-Aware Adaptive Fusion (DuAF) module dynamically calibrates modality contributions by jointly modeling user preferences and temporal information, enabling the extraction of multimodal features aligned with evolving user interests; (2) by integrating Modality Adversarial Training with the Mixture-of-Experts paradigm, MAT-MoE employs an ensemble of expert generators to dynamically reconstruct missing modality representations, effectively mitigating modality imbalance challenges; (3) to address the inherent sparsity of sequential behavior data, we propose a Multi-Supervised Contrastive Learning strategy that integrates cross-modal alignment and virtual sequence augmentation. This approach enhances user interest modeling by leveraging diverse learning signals, resulting in improved model robustness and generalization capability. Extensive experiments on four public datasets demonstrate that DuAF-MAT significantly outperforms state-of-the-art baselines.

Yupeng Li, Mingyue Cheng, Yucong Luo, Yitong Zhou, Qingyang Mao, Shijin Wang

Multi-behavior sequential recommendation aims to capture users' dynamic interests by modeling diverse types of user interactions over time. Although several studies have explored this setting, the recommendation performance remains suboptimal, mainly due to two fundamental challenges: the heterogeneity of user behaviors and data sparsity. To address these challenges, we propose BLADE, a framework that enhances multi-behavior modeling while mitigating data sparsity. Specifically, to handle behavior heterogeneity, we introduce a dual item-behavior fusion architecture that incorporates behavior information at both the input and intermediate levels, enabling preference modeling from multiple perspectives. To mitigate data sparsity, we design three behavior-level data augmentation methods that operate directly on behavior sequences rather than core item sequences. These methods generate diverse augmented views while preserving the semantic consistency of item sequences. These augmented views further enhance representation learning and generalization via contrastive learning. Experiments on three real-world datasets demonstrate the effectiveness of our approach.

Yujie Li, Zezhi Shao, Chengqing Yu, Yisong Fu, Tao Sun, Yongjun Xu, Fei Wang

Time series forecasting under distribution shift remains challenging, as existing deep learning models often rely on local statistical normalization (e.g., mean and variance) that fails to capture global distribution shift. Methods like RevIN and its variants attempt to decouple distribution and pattern but still struggle with missing values, noisy observations, and invalid channel-wise affine transformation. To address these limitations, we propose Affine Prototype-Timestamp(APT), a lightweight and flexible plug-in module that injects global distribution features into the normalization–forecasting pipeline. By leveraging timestamp-conditioned prototype learning, APT dynamically generates affine parameters that modulate both input and output series, enabling the backbone to learn from self-supervised, distribution-aware clustered instances. APT is compatible with arbitrary forecasting backbones and normalization strategies while introducing minimal computational overhead. Extensive experiments across six benchmark datasets and multiple backbone-normalization combinations demonstrate that APT significantly improves forecasting performance under distribution shift.

Yuan Li, Jun Hu, Bryan Hooi, Bingsheng He, Cheng Chen

Real-world fraud detection applications benefit from graph learning techniques that jointly exploit node features—often rich in textual data—and graph structural information. Recently, Graph-Enhanced LLMs have emerged as a promising graph learning approach that converts graph information into prompts, exploiting LLMs' ability to reason over both textual and structural information. Among them, text-only prompting, which converts graph information into prompts consisting solely of text tokens, offers a solution that relies only on LLM tuning without requiring additional graph-specific encoders. However, text-only prompting struggles on heterogeneous fraud-detection graphs: multi-hop relations expand exponentially with each additional hop, leading to rapidly growing neighborhoods associated with dense textual information. These neighborhoods may overwhelm the model with long, irrelevant content in the prompt and suppress key signals from the target node, thereby degrading performance. To address this challenge, we propose Dual Granularity Prompting (DGP), which mitigates information overload by preserving fine-grained textual details for the target node while summarizing neighbor information into coarse-grained text prompts. DGP introduces tailored summarization strategies for different data modalities—bi-level semantic abstraction for textual fields and statistical aggregation for numerical features—enabling effective compression of verbose neighbor content into concise, informative prompts. Experiments across public and industry datasets demonstrate that DGP operates within a manageable token budget while improving fraud detection performance by up to 6.8% (AUPRC) over state-of-the-art methods, showing the potential of Graph-Enhanced LLMs for fraud detection.

Yichen Li, Yichen Tan, Yijing Shan, Haozhao Wang, Rui Zhang, Imran Razzak, Ruixuan Li

Data-Centric Sequential Recommendation (DaCSR) has emerged as a promising technique that enhances dataset quality to better capture user preferences without increasing training complexity. However, mining item relations to improve data quality remains challenging due to the intricate nature of interaction sequences. Existing methods predominantly either: 1) optimize models to learn such item relations from fixed datasets at significant training cost, or 2) employ generative models to adaptively learn only interaction patterns, which lack interpretability and cannot guarantee effective data quality enhancement. In this paper, we pioneer a relation-guided dataset augmentation and regeneration framework for sequential recommendation called \textbf{RaSR}. This framework can significantly improve model performance on original datasets while maintaining training efficiency without modifying the model architecture. Specifically, we first preprocess user interactions to construct standardized sequential data and extract semantic representations via a Large Language Model (LLM). We then build a multi-relation graph with manually predefined metrics and semantic representations to generate augmented datasets. Finally, a relation-aware generator can produce regenerated datasets with both the multi-relation graph and the augmented dataset. To verify the effectiveness of RaSR, we conduct experiments on various backbone models and datasets, and achieve significant performance improvement compared to training the model only on the original dataset.

Xuetao Li, Wenke Huang, Nengyuan Pan, Kaiyan Zhao, Songhua Yang, Yiming Wang, Mengde Li, Mang Ye, Jifeng Xuan, Miao Li

Humanoid robots exhibit significant potential in executing diverse human-level skills. However, current research predominantly relies on data-driven approaches that necessitate extensive training datasets to achieve robust multimodal decision-making capabilities and generalizable visuomotor control. These methods raise concerns due to the neglect of geometric reasoning in unseen scenarios and the inefficient modeling of robot-target relationships within the training data, resulting in a significant waste of training resources. To address these limitations, we present the Recurrent Geometric-prior Multimodal Policy (RGMP), an end-to-end framework that unifies geometric-semantic skill reasoning with data-efficient visuomotor control. For perception capabilities, we propose the Geometric-prior Skill Selector, which infuses geometric inductive biases into a vision language model, producing adaptive skill sequences for unseen scenes with minimal spatial common sense tuning. To achieve data-efficient robotic motion synthesis, we introduce the Adaptive Recursive Gaussian Network, which parameterizes robot-object interactions as a compact hierarchy of Gaussian processes that recursively encode multi-scale spatial relationships, yielding dexterous, data-efficient motion synthesis even from sparse demonstrations. Evaluated on both our humanoid robot and desktop robot, the RGMP framework achieves 87% task success in generalization tests and exhibits 5× greater data efficiency than the state-of-the-art model. This performance underscores its superior cross-domain generalization, paving the way for more versatile and data-efficient robotic systems.

Xiaoxiang Li, Xihe Xie, Hai Wan, Xibin Zhao

Graph-level anomaly detection (GLAD), which identifies rare or atypical graphs within a graph set, is crucial for applications such as image analysis, industrial defect inspection and fraud detection. However, existing GLAD approaches typically rely on the in-distribution hypothesis while lacking generalization capability for out-of-distribution (OOD) scenarios (e.g., different graph sizes), which largely limits the application in the real world. For the first time, we formulate the OOD generalization problem for GLAD, where testing graph data exhibit significant distributional shifts from training data. To tackle two common types of distributional shifts, domain generalization and subpopulation shift, we propose the Fine-Grained Subpopulation Graph-Level Anomaly Detection (FGS-GLAD). First, we propose a Graph Information Bottleneck-based Anomaly Detection Module (GIB4AD) that implements graph reverse distillation and graph information bottleneck on the graph to enhance task-relevant feature extraction for domain generalization. Second, We propose a Fine-Grained Subpoulation Inference Module (FGSI) to predict fine-grained subpopulations and focus on critical inter-subpopulation features through a supervised contrastive mechanism. Experiments on seven benchmark datasets and ten baselines demonstrate our model's superiority in handling domain generalization and subpopulation shift.

Xiaojing Li, Bin Wang, Xiaochun Yang, Meng Luo

Learnable sparse retrieval (LSR) models encode texts into high-dimensional sparse representations, supporting token-level expansion beyond the original text and addressing the vocabulary mismatch problem in traditional bag-of-words retrieval. However, in the absence of representation-level supervision, these representations usually overemphasize irrelevant tokens while neglecting truly relevant ones. We term this phenomenon the Representation Hallucination problem in LSR models, a critical bottleneck impeding accurate retrieval. To address this challenge, we introduce SiRe, a self-improving training framework for sparse retrieval that integrates two core strategies: Heuristic Representation Refinement and Representation-Focused Learning. Specifically, SiRe first identifies and corrects representation hallucinations in the outputs of the current LSR model using heuristic methods. The resulting representations serve as the primary supervision signals, guiding a pretrained language model (e.g., BERT) to mitigate the problem directly at the representation level. This process can be iterated, enabling progressive model improvement. Extensive experiments on both in-domain and out-domain benchmarks show that SiRe produces higher-quality sparse representations, significantly enhancing retrieval performance over strong baselines.