Video-guided 3D animation holds immense potential for content creation, offering intuitive and precise control over dynamic assets. However, practical deployment faces a critical yet frequently overlooked hurdle: the pose misalignment dilemma. In real-world scenarios, the initial pose of a user-provided static mesh rarely aligns with the starting frame of a reference video. Naively forcing a mesh to follow a mismatched trajectory inevitably leads to severe geometric distortion or animation failure. To address this, we present Rectified Dynamic Mesh (R-DMesh), a unified framework designed to generate high-fidelity 4D meshes that are “rectified” to align with video context. Unlike standard motion transfer approaches, our method introduces a novel VAE that explicitly disentangles the input into a conditional base mesh, relative motion trajectories, and a crucial rectification jump offset. This offset is learned to automatically transform the arbitrary pose of the input mesh to match the video’s initial state before animation begins. We process these components via a Triflow Attention mechanism, which leverages vertex-wise geometric features to modulate the three orthogonal flows, ensuring physical consistency and local rigidity during the rectification and animation process. For generation, we employ a Rectified Flow-based Diffusion Transformer conditioned on pre-trained video latents, effectively transferring rich spatio-temporal priors to the 3D domain. To support this task, we construct Video-RDMesh, a large-scale dataset of over 500k dynamic mesh sequences specifically curated to simulate pose misalignment. Extensive experiments demonstrate that R-DMesh not only solves the alignment problem but also enables robust downstream applications, including pose retargeting and holistic 4D generation. Code and pre-trained weights will be available at: https://github.com/Tencent-Hunyuan/R-DMesh.
论文检索
输入标题、作者或关键词，从 3,998 篇学术成果中精准定位
The growing demand for personalized 3D avatars calls for efficient reconstruction methods from casual photos. This task remains challenging due to unconstrained viewpoints, partial body visibility, and temporal variations across input images. While some previous methods circumvent these difficulties by adopting generative approaches like score distillation, they struggle to preserve authentic appearance details from source images. To address these limitations, we introduce Visual-Fusion-Avatar (VFAvatar), a novel feed-forward framework that reconstructs 3D avatars by fusing visual cues in just a few seconds. VFAvatar couples a pose-free reconstruction foundation model with a pretrained human generation prior in a mutually reinforcing manner. And we propose a visibility-aware, view-attentive residual aggregation mechanism that routes and fuses per-view updates, allowing partial observations from different images to be assembled into a single coherent avatar. Experiments demonstrate that VFAvatar significantly outperforms state-of-the-art methods in both reconstruction fidelity and efficiency, while enabling shape and pose manipulation. Code is available on: https://github.com/huangshuo200823/VFAvatar.
Global illumination (GI) is essential for realism but remains computationally expensive. While per-scene neural methods lack generalization and screen-space approaches inherently suffer from view inconsistency, prior 3D neural rendering methods face a severe scalability barrier, restricting them to small, object-centric meshes. To overcome these trade-offs, we introduce a generalizable light transport 3D embedding that predicts global illumination directly from 3D scene configurations without rasterized or path-traced illumination cues, per-scene retraining or screen-space limitations. We employ a point-based representation to decouple our embedding from the original scene topology, then utilize a linear-complexity transformer to encode long-range light transport. This design scales to environments with millions of triangles, enabling the first generalizable GI learning on complex, high-fidelity indoor scenes, far beyond prior limits. To achieve this, we enforce a local query mechanism where rendering queries are processed independently under 3D supervision. This ensures constant complexity per pixel relative to scene size, yielding view-consistent and resolution-agnostic rendering without the memory bottlenecks typical of globally coupled attention. We further demonstrate versatility by re-targeting the encoder with limited fine-tuning, presenting preliminary results on spatial-directional radiance field prediction for glossy materials and validating transfer to downstream rendering tasks.
Realistic integration of user-specified textures into scene images is a fundamental task in computer graphics and image editing. While existing material transfer and reference-guided inpainting methods can edit surface appearances, they often fail to address the specific requirements of texture tiling. This task necessitates precisely repeating a reference pattern according to user-defined parameters such as frequency, orientation, and scale. Furthermore, current generative approaches often struggle to maintain the structural fidelity of the reference texture, limited by either destructive pixel-level resampling or the lack of fine-grained spatial information in semantic image encoders, and they frequently fail to preserve the coherent lighting and geometry of the original scene. In this paper, we propose a novel framework for controllable and high-fidelity texture tiling based on Diffusion Transformers. Our approach introduces two key technical innovations to decouple spatial manipulation from content generation. First, we propose a Coordinate-Transformed Rotary Embedding mechanism. By applying 2D affine transformations directly to the relative positional embeddings between the target latent and the image condition, we achieve precise control over tiling patterns without explicit pixel warping, thereby utilizing the full information of the reference condition without degradation. Second, a Disjoint Attention Mask is employed to shield reference features from semantic leakage. This preserves structural integrity while seamlessly blending the synthesized texture with the scene’s original lighting and geometry. Extensive experiments demonstrate that our method outperforms state-of-the-art baselines in both control accuracy and texture fidelity. The code and dataset are publicly accessible at https://github.com/junrongh/ControlTile.
Camera-controlled video generation is valuable for applications ranging from visual design to providing 2D supervision for 4D generation tasks. However, existing approaches are limited to single-trajectory generation, forcing users to process multiple trajectories in separate batches. This serial inference introduces content inconsistencies across viewpoints due to the inherent randomness of diffusion models. Explicit point cloud methods can only partially address this problem, as single-viewpoint back-projection suffers from sparsity and depth estimation errors. We propose CameraSquad, a multi-trajectory camera control framework that supports both single-trajectory and parallel multi-trajectory generation. Our method achieves precise camera control while preserving input video content through decoupled content and camera control mechanisms. To ensure viewpoint consistency in multi-trajectory mode, we design a dual-mode cross-view attention mechanism that maintains consistency across parallel trajectories while guaranteeing camera control precision. Extensive experiments demonstrate that CameraSquad achieves competitive performance in camera control accuracy, consistency maintenance, and generation quality compared to existing approaches. Our project page is available at https://rabberk.github.io/CameraSquad/.
WED-Net: A Weather-Effect Disentanglement Network with Causal Augmentation for Urban Flow Prediction
Urban spatio-temporal prediction under extreme conditions (e.g., heavy rain) is challenging due to event rarity and dynamics. Existing data-driven approaches that incorporate weather as auxiliary input often rely on coarse-grained descriptors and lack dedicated mechanisms to capture fine-grained spatio-temporal effects. Although recent methods adopt causal techniques to improve out-of-distribution generalization, they typically overlook temporal dynamics or depend on fixed confounder stratification. To address these limitations, we propose WED-Net (Weather-Effect Disentanglement Network), a dual-branch Transformer architecture that separates intrinsic and weather-induced traffic patterns via self- and cross-attention, enhanced with memory banks and fused through adaptive gating. To further promote disentanglement, we introduce a discriminator that explicitly distinguishes weather conditions. Additionally, we design a causal data augmentation strategy that perturbs non-causal parts while preserving causal structures, enabling improved generalization under rare scenarios. Experiments on taxi-flow datasets from three cities demonstrate that WED-Net delivers robust performance under extreme weather conditions, highlighting its potential to support safer mobility, disaster preparedness, and urban resilience in real-world settings. The code is publicly available at https://github.com/HQ-LV/WED-Net.
Accurate enterprise power consumption forecasting is not only a core component of optimized green energy management but also a key support for promoting the coordinated development of a sustainable society and the digital economy. The temporal fluctuations in power consumption reflect an enterprise's production activity and operational resilience, while credit assessment combined with Web data reveals a two-way coupling relationship between it and energy use: credit changes influence financing and power consumption strategies, while energy anomalies may become early signals of credit risk. However, existing methods still have shortcomings in modeling the co-evolution of Web data and power data. Most models only focus on static or unidirectional correlations, making it difficult to capture the dynamic feedback between credit risk and power consumption; traditional multi-task learning frameworks often rely on parameter sharing or simple attention mechanisms, lacking consistency constraints across time scales and network structures. To address this, this paper proposes CPDGL, a credit-electricity co-evolution framework based on dynamic graph learning, which simultaneously performs power forecasting and credit risk assessment within a unified multi-task system. Its co-evolution path interaction module explicitly models the feedback loop between credit dynamics and power behavior, learning bidirectional causal relationships through an adaptive influence matrix; the semantic path aggregation module integrates static and dynamic features, strengthening cross-modal expression and global reasoning capabilities. Large-scale experiments conducted in a real-world enterprise environment of one of the world's largest power suppliers demonstrate that CPDGL achieves state-of-the-art performance in both power forecasting and credit assessment tasks. The results validate its broad applicability in multi-source Web data fusion scenarios, significantly improving forecasting accuracy and dispatch efficiency in clean energy management, and showcasing practical value and social impact in smart cities and sustainable development.
User interactions on e-commerce platforms are inherently diverse, involving behaviors such as clicking, favoriting, adding to cart, and purchasing. The transitions between these behaviors offer valuable insights into user-item interactions, serving as a key signal for understanding evolving preferences. Consequently, there is growing interest in leveraging multi-behavior data to better capture user intent. Recent studies have explored sequential modeling of multi-behavior data, many relying on transformer-based architectures with polynomial time complexity. While effective, these approaches often incur high computational costs, limiting their applicability in large-scale industrial systems with long user sequences. To address this challenge, we propose the Transition-Aware Graph Attention Network (TGA), a linear-complexity approach for modeling multi-behavior transitions. Unlike traditional transformers that treat all behavior pairs equally, TGA constructs a structured sparse graph by identifying informative transitions from three perspectives: (a) item-level transitions, (b) category-level transitions, and (c) neighbor-level transitions. Built upon the structured graph, TGA employs a transition-aware graph Attention mechanism that jointly models user-item interactions and behavior transition types, enabling more accurate capture of sequential patterns while maintaining computational efficiency. Experiments show that TGA outperforms all state-of-the-art models while significantly reducing computational cost. Notably, TGA has been deployed in a large-scale industrial production environment, where it leads to impressive improvements in key business metrics.
The proliferation of highly realistic deepfake videos threatens public trust and the integrity of digital information. However, detecting sophisticated deepfakes requires analysis beyond surface-level visual artifacts. We propose Harmonizing Action Units with Temporal-contextual Embeddings (HAUTE), integrating physiological muscle dynamics with holistic semantic context through adaptive attention mechanisms. HAUTE captures temporal Action Unit coordination patterns and high-level contextual embeddings, enabling the model to reveal synthesis-induced inconsistencies imperceptible to isolated modalities. Extensive experiments demonstrate state-of-the-art performance with strong cross-dataset adaptability, particularly on commercial tool-based high-quality deepfakes, advancing trustworthy content verification for web ecosystems.
Attention mechanisms are essential to the success of Large Language Models (LLMs). In practice, models often overemphasize semantically low-value tokens, forming attention sinks while failing to capture truly informative tokens. Existing inference-time optimization methods mainly rely on static adjustments or attention redistribution, which often disrupt the correspondence between attention distribution and the actual semantics of the input, leading to a loss of semantic consistency and degraded performance. To address this problem, we propose PAOSC, a plug-and-play attention optimization model designed to maintain semantic consistency by dynamically adjusting attention. PAOSC employs a generator to identify informative tokens and a discriminator to optimize the generator via policy gradients based on confidence changes and loss fluctuations. Experiments on eight LLMs show up to a 9.68% improvement in the F1 score. On the constructed HTTP-RL dataset, PAOSC eliminates 18% of low-value tokens, improving inference efficiency while maintaining semantic consistency. Our code is available at https://github.com/ChangLi000/PAOSC.
Personalized search ranking systems are critical for driving engagement and revenue in modern e-commerce platforms. Existing methods primarily model users' broad interests from historical behaviors but often fail to explicitly align these with real-time intent expressed in user queries. In this paper, we propose DiffusionGS, a scalable generative framework that treats user queries as explicit intent anchors to extract user interests from long-term, noisy behavior histories. Specifically, we formulate interest extraction as a conditional denoising task, where the user's query guides a conditional diffusion process to produce a robust, user intent-aware representation from their behavioral sequence. A User-aware Denoising Layer (UDL) further refines attention distribution using user-specific profiles. By reframing queries as intent priors and leveraging diffusion-based denoising, our method provides a powerful mechanism for capturing dynamic user interest shifts. Extensive offline and online experiments demonstrate the superiority of DiffusionGS over state-of-the-art methods.
Generative inference with Large Language Models (LLMs) is the cornerstone of web-scale AI search, where queries are answered using vast, heterogeneous documents retrieved via Retrieval-Augmented Generation (RAG). This paradigm is critically bottlenecked by the cost of self-attention mechanism on long context. The sheer diversity of retrieved web content (multi-sourced, multi-lingual, multi-faceted) makes simple Key-Value (KV) cache optimizations with pre-fixed subsets ineffective, demanding a dynamic, content-aware approach. This challenge, however, introduces a classic chicken-and-egg problem: the model cannot foresee the necessary KV entries for attention without first inferring on the content, yet doing so on the full context is prohibitively expensive. This paper introduces P&F, a unified framework that resolves this dilemma through a core ''probe-and-fetch'' mechanism, which ingeniously integrates with speculative decoding -- an acceleration approach already adopted in web-scale AI search. The probe step repurposes the speculative draft model: while generating candidate tokens, it simultaneously probes the context to predict the most salient KV entries the large model will need for attention. The fetch step immediately acts on this prediction, asynchronously fetching these sparse entries from memory. This synergistic design piggybacks the probing step onto the drafting process, allowing the expensive gathering of a sparse KV cache to be fully masked. Crucially, this co-design breaks the sequential dependency bottleneck that cripples naive integrations of speculative decoding and prefetching due to synchronization issues. Extensive experiments show P&F significantly outperforms state-of-the-art methods in throughput and scalability, offering a practical, drop-in solution. Extensive offline evaluations across various settings and datasets demonstrate that P&F yields superior throughput and scalability compared to advanced baselines, while maintaining model quality across diverse models and scales. In online settings, P&F delivers substantial gains in throughput improvements while preserving response quality, making it well-suited for large-scale industrial deployment in real-time AI Search services.
In recent years, the success of large language models (LLMs) has driven the exploration of scaling laws in recommender systems. However, models that demonstrate scaling laws are actually challenging to deploy in industrial settings for modeling long sequences of user behaviors, due to the high computational complexity of the standard self-attention mechanism. Despite various sparse self-attention mechanisms proposed in other fields, they are not fully suited for recommendation scenarios. This is because user behaviors exhibit personalization and temporal characteristics: different users have distinct behavior patterns, and these patterns change over time, with data from these users differing significantly from data in other fields in terms of distribution. To address these challenges, we propose SparseCTR, an efficient and effective model specifically designed for long-term behaviors of users. To be precise, we first segment behavior sequences into chunks in a personalized manner to avoid separating continuous behaviors and enable parallel processing of sequences. Based on these chunks, we propose a three-branch sparse self-attention mechanism to jointly identify users' global interests, interest transitions, and short-term interests. Furthermore, we design a composite relative temporal encoding via learnable, head-specific bias coefficients, better capturing sequential and periodic relationships among user behaviors. Extensive experimental results show that SparseCTR not only improves efficiency but also outperforms state-of-the-art methods. More importantly, it exhibits an obvious scaling law phenomenon, maintaining performance improvements across three orders of magnitude in FLOPs. In online A/B testing, SparseCTR increased CTR by 1.72% and CPM by 1.41%. Our source code is available at https://github.com/laiweijiang/SparseCTR.
Generative Recommenders (GRs), exemplified by the Hierarchical Sequential Transduction Unit (HSTU), have emerged as a powerful paradigm for modeling long user interaction sequences. However, we observe that their ''flat-sequence'' assumption overlooks the rich, intrinsic structure of user behavior. This leads to two key limitations: a failure to capture the temporal hierarchy of session-based engagement, and computational inefficiency, as dense attention introduces significant noise that obscures true preference signals within semantically sparse histories, which deteriorates the quality of the learned representations. To this end, we propose a novel framework named HPGR (Hierarchical and Preference-aware Generative Recommender), built upon a two-stage paradigm that injects these crucial structural priors into the model to handle the drawback. Specifically, HPGR comprises two synergistic stages. First, a structure-aware pre-training stage employs a session-based Masked Item Modeling (MIM) objective to learn a hierarchically-informed and semantically rich item representation space. Second, a preference-aware fine-tuning stage leverages these powerful representations to implement a Preference-Guided Sparse Attention mechanism, which dynamically constrains computation to only the most relevant historical items, enhancing both efficiency and signal-to-noise ratio. Empirical experiments on a large-scale proprietary industrial dataset from APPGallery and an online A/B test verify that HPGR achieves state-of-the-art performance over multiple strong baselines, including HSTU and MTGR.
Modeling asynchronous event sequences is crucial in numerous real-world applications such as healthcare monitoring, financial transaction analysis, and so on. Traditional temporal point processes, including Hawkes Processes, often fail to capture complex dependencies due to their parametric limitations. While neural approaches like RNNs and Transformers have improved flexibility, they struggle with computational inefficiency, and attention saturation. In this paper, we introduce the Mamba Hawkes Process (MHP), the first framework to integrate selective state space model (Mamba) with temporal point processes. MHP leverages time-varying state transitions and input-dependent gating to efficiently encode event history and capture long-term dependencies with linear complexity. Importantly, we provide theoretical guarantees showing that MHP generalizes both classical multi-exponential Hawkes processes and exponential-decay gated RNNs, underscoring its expressive power and theoretical soundness. To address the inherent constraints of pure state space models in handling heterogeneous event interactions, we further develop Adaptive Mamba Hawkes Process (A-MHP) that incorporates two novel mechanisms: a Time-Scaling Mechanism that adaptively weights time intervals based on event type and history, and a Dual-Channel State Transition that adaptively processes event content and temporal dynamics for more refined state updates. Extensive experiments on synthetic and real-world datasets demonstrate that MHP and A-MHP consistently outperform state-of-the-art baselines in event prediction tasks, particularly in long-sequence scenarios. Our work establishes a scalable and theoretically grounded paradigm for event sequence modeling, with practical implications for predictive maintenance, anomaly detection, and dynamic system analysis. The code is available at https://github.com/Ethan-Shen-Individual-Lab/Mamba-Hawkes-Process.
The strategic planning and reliability of modern web services, from cloud infrastructures to e-commerce platforms, increasingly hinge on accurate long-term forecasting of high-dimensional time series. A fundamental challenge within this task is modeling the intricate point-to-point dependencies that span across both time and variable dimensions. However, many existing methods face restricted direction modeling and computational inefficiency due to their reliance on localized paradigms and Transformer architectures. To address these, we propose replacing Self-Attention with autocorrelation, achieving two key innovations: 1) We propose calculating autocorrelation across both variable and time dimensions, which is a global paradigm, to model point-to-point dependencies. 2) Our proposed Spectral Product Mechanism (SPM) optimizes traditional autocorrelation into a data-driven form suitable for deep learning. Moreover, SPM reformulates autocorrelation as spectral product and reduces the complexity from O(N2) to O(NlogN), while its Hadamard product-based correlation score matrix further reduces core computation to O(N) compared to Self-Attention's O(N2) matrix multiplication. We further propose a Generalized Spectral Product Mechanism (GSPM), which extends traditional autocorrelation by mapping input into distinct feature representations, enabling modeling of complex dependencies through cross-feature correlations. SPM and GSPM surpass current state-of-the-art (SOTA) methods on 14 authoritative benchmarks, collectively securing the top rank on 22 out of 28 metrics, while ranking 1st in time, 2nd in memory, and 2nd in parameter overhead. Source code is available at: https://github.com/lxy-PhD2022/SPM.
Time series forecasting is crucial for the development of sophisticated web technologies, driving smarter, more responsive, and data-driven web applications. A key to accurate forecasting lies in effectively capturing the intricate dependencies among different variables (channels). While existing channel-dependent methods have shown strong performance by explicitly modeling inter-channel relationships, they face two critical challenges when applied to high-dimensional datasets with thousands of channels. First, the computational complexity of them grows quadratically with the number of channels, leading to significant scalability issues. Second, attention weights reveal that inter-channel dependencies exhibit both local clusters and global structures, yet current methods fail to disentangle these heterogeneous patterns, resulting in mutual interference and degraded forecasting accuracy. To address these challenges, we propose a novel Channel Reordering-Aligned group Fusion Transformer (CRAFT) for high-dimensional time series forecasting. Specifically, we design an energy-based channel reordering mechanism that reorganizes channels into a minimal-energy state, preserving inherent local-global structures. Building on reordered structure, we introduce a group fusion Transformer that explicitly separates local and global dependencies, significantly reducing computational complexity while enhancing representational clarity. Experiments on high-dimensional datasets demonstrate that CRAFT consistently outperforms baselines, achieving higher forecasting accuracy with lower computational overhead.
The proliferation of misinformation on video-sharing platforms demands robust detection of video fake news. Existing methods struggle to integrate external world knowledge with internal multimodal cues, limiting their generalization and robustness. In this work, we propose TrueLens, a new framework for video fake news detection that gathers and consolidates dual-level evidence \zznotethrough three primary components, \ie, External Precedent Retriever, Adversarial Contrastor, and Internal Evidential Logic Fusion. At the external level, the External Precedent Retriever first decomposes the query video into textual, visual, and audio queries while leveraging multimodal large language models (MLLMs) to enhance the overall semantic representation. It then applies an entropy-guided multimodal retrieval mechanism to identify the two most similar reference videos from a gallery of real and fake samples, \ie, one real and one fake video. The Adversarial Contrastor integrates these references with the input video through contrastive attention, enhancing contextual reasoning. At the internal level, our Evidential Logic Fusion module aggregates multimodal signals from the Adversarial Contrastor to produce consistent, robust predictions. Extensive experiments on three benchmarks show that the proposed TrueLens consistently surpasses competitive baselines under both temporal and event settings by a clear margin, yielding up to a +21.60% F1 improvement under the event setting and achieving 93.73%, 90.64%, and 98.83% accuracy on the FakeSV, FakeTT, and FVC datasets under the temporal setting. The code for our project is available at https://github.com/JunyiChen-ai/TrueLens.
Event sequences are integral to domains such as e-commerce, social networks, and healthcare. Traditional point process models, like Poisson and Hawkes processes, are foundational but limited by rigid parametric assumptions, constraining their flexibility in complex real-world scenarios. Neural point processes offer a more adaptable alternative, but typically perform implicit sequence modeling, which does not fully exploit critical event interaction patterns and limits transparency. To address these challenges, we introduce the Explicit Interaction-aware Attention Network (EIAN), a novel model that enhances event modeling by explicitly capturing both intra-type and cross-type event interactions. Specifically, EIAN employs two key components: an intra-type temporal encoder that preserves the unique temporal dynamics within each event type, and a cross-type interaction decoder that highlights interactions across event types. Furthermore, two temporal encoding mechanisms are integrated into the interaction decoder to handle irregular inter-event intervals in diverse temporal scenarios. Extensive experiments show that EIAN consistently outperforms existing models in predictive performance and provides deeper insights into event interaction patterns, advancing both flexibility and interpretability. Our code is available at https://github.com/CGCL-codes/EIAN.git.
Attribute-Specific Fashion Retrieval (ASFR) aims to improve fine-grained image retrieval by focusing on specific attributes. However, existing patch-based attention and Transformer methods often misalign with irregular attribute regions and are prone to background noise, limiting their ability to capture subtle, pixel-level microstructures. To tackle these challenges, we propose Super Fashion. , the first ASFR framework that adopts superpixel tokens within a Transformer architecture. Super Fashion initially employs an attribute-guided attention mechanism to extract attribute-related features, which in turn guide the cropping of semantically meaningful image regions. Superpixel segmentation is then leveraged on these regions to generate compact, semantically coherent superpixel tokens. By incorporating modality-specific embeddings for both attribute and superpixel tokens, the superpixel token-based Transformer facilitates adaptive interaction and fusion, thereby enhancing attribute localization and discrimination. Extensive experiments on FashionAI, DARN, and DeepFashion demonstrate relative overall MAP improvements of 1.84%, 9.27%, and 9.35% over prior SOTA. Super Fashion offers a new solution for web-based image retrieval.