We propose a system to optimize parametric designs subject to radiation pressure, i.e., the effect of light on the motion of objects. This is most relevant in the design of spacecraft, where radiation pressure presents the dominant non-conservative forcing mechanism, which is the case beyond approximately 800 km altitude. Despite its importance, the high computational cost of high-fidelity radiation pressure modeling has limited its use in large-scale spacecraft design, optimization, and space situational awareness applications. We enable this by offering three innovations in the simulation, in representation and in optimization: First, a practical computer graphics-inspired Monte-Carlo (MC) simulation of radiation pressure. The simulation is highly parallel, uses importance sampling and next-event estimation to reduce variance and allows simulating an entire family of designs instead of a single spacecraft as in previous work. Second, we introduce neural networks as a representation of forces from design parameters. This neural proxy model, learned from simulations, is inherently differentiable and can query forces orders of magnitude faster than a full MC simulation. Third, and finally, we demonstrate optimizing inverse radiation pressure designs, such as finding geometry, material or operation parameters that minimizes travel time, maximizes proximity given a desired end-point, minimize thruster fuel, trains mission control policies or allocated compute budget in extraterrestrial compute.
论文检索
输入标题、作者或关键词，从 2,893 篇学术成果中精准定位
Following the success of deep neural networks in 2D pose estimation, reconstruction-based approaches have significantly advanced multi-person 3D pose estimation from sparse multi-view images. These methods typically detect 2D poses independently in each view and then associate them for 3D reconstruction. However, despite strong progress, recent state-of-the-art methods still face critical limitations: 1) They often depend on global optimization over a large and complex set of multi-view 2D joints to jointly infer 3D poses for all individuals, making the process highly complex and prone to suboptimal solutions; 2) Their tight coupling with the bottom-up detector OpenPose hinders the use of more advanced top-down or single-stage 2D pose estimators and restricts the integration of richer instance-level cues learned by these models. To address these limitations, we propose TwinPose, a novel framework that alleviates the complexity of global pose inference by optimizing within person-specific 3D pose subspaces, while fully supporting diverse 2D pose detectors and effectively leveraging pose-instance cues. The key idea is to introduce a twin pose — a 3D counterpart of each 2D pose — that inherits its instance representation and aggregates geometrically consistent 2D joints from other views. All twin poses are unified in a common 3D space, where those belonging to the same individual naturally share a number of bones. This structural property enables association by counting shared bones, forming person-specific subspaces from which each individual's 3D pose can be inferred independently in an efficient and robust manner. Extensive experiments demonstrate that TwinPose achieves state-of-the-art performance in both accuracy and efficiency across multiple public and proprietary datasets. Importantly, it is fully detector-agnostic, allowing seamless integration with current and future advances in 2D pose estimation while remaining highly robust to noisy or imperfect 2D predictions. Project page with code and additional resources: https://github.com/zgspose/TwinPose
We describe a method for computing signed distance to point clouds that allows fast pointwise evaluation at arbitrary spatial resolution. As input, our method takes a point cloud with normals; as output, it provides an analytical parameterization that allows queries of signed distance to the approximate underlying surface at arbitrary points — simultaneously providing reconstruction and distance. Our key idea is to reconstruct shapes by locally fitting point clouds with tori, which have closed-form signed distance functions. Tori are fitted in a feed-forward manner, using a pre-trained network to output per-point curvature and shift parameters. Importantly, our method does not require costly global optimization or spatial discretization, and is easily parallelizable. Underlying our method is a new theory that unifies signed distance with the classic reconstruction methods of winding numbers and Poisson surface reconstruction. We use our method to compute signed distance to point clouds arising from photogrammetry, meshes, 3D Gaussians, and neural implicits. Our method allows point clouds to be used directly in applications, without explicit surface reconstruction: as examples, we take offsets of point clouds, apply morphological and Boolean operations, and directly visualize offset surfaces using sphere tracing.
Anisotropic friction is a critical source of propulsion for efficient locomotion of many terrestrial animals. The interplay between animal morphology, control, and anisotropically frictional contact makes designing optimal anisotropic friction for terrestrial locomotion intriguing and challenging. We propose a computational pipeline for co-designing anisotropic friction and controllers of terrestrial robots with diverse morphologies. Our pipeline presents a co-design algorithm that alternates between optimizing direction of anisotropic friction and training a neural network controller to improve the locomotion performance of a given robot morphology. Based on the intuition that controller’s performance does not change significantly when the frictional force differs slightly, we introduce the concept of trust-region into robot co-design, allowing the controller network to continue training from the previous iteration. Our evaluation on various morphologies show that anisotropic friction is critical for terrestrial robot locomotion, and our pipeline is statistically better than current state-of-the-art methods. Furthermore, we reveal that large language models (LLM) constitute a strong baseline for this kind of co-design problems, worth receiving more attention. We demonstrate that co-designing anisotropic friction and control unlocks effective locomotion in various downstream tasks, including locomotion on uneven terrain, navigation in a maze, and object manipulation. To validate our pipeline in the real world, we design and 3D print a variety of scales and systematically measure their anisotropic friction coefficients. Then we construct a multi-link robot with anisotropic scales designed by our pipeline and compare its performance with isotropic scales. Our real-world experiments confirm that isotropic scales are insufficient to support terrestrial robots’ locomotion abilities, and computationally co-designing friction and control enables robots to perform tasks including turning, slithering, and other non-trivial locomotion tasks.
We present a training algorithm to mitigate optimization instabilities in small neural networks, like those used in real-time neural shading applications. While large, overparameterized models exhibit predictable convergence, smaller architectures often suffer from high optimization variance: differently initialized models converge to disparate local minima. To reduce these training instabilities, we introduce an optimization approach that utilizes an ensemble of network instances during training. We prune underperforming instances and dynamically resize training batches to maintain wall-clock timings comparable to—or faster than—single-instance training. This strategy allows efficiently exploring the weight space yielding a well-performing model with significantly higher likelihood than optimizing a single instance only. We develop and analyze the algorithm in the context of learning reflectance functions for neural shading. This is a challenging task for small neural models due to the high dynamic range of the target function. In addition to our multi-instance training method, we also revisit the choices of loss functions, activation functions, and input parameterization to further improve quality and training robustness.
Low-discrepancy points have proven to be a useful source of samples for numerical integration via quasi-Monte Carlo integration. Low-discrepancy points are typically uniformly distributed in the unit hypercube. It is thus common to transform the input uniform points into target non-uniform samples via mapping, just like how uniform random numbers are used for generating non-uniform samples. Unfortunately, this common approach does not guarantee to carry over the low-discrepancy nature of the input points into the output non-uniform samples, giving the resulting non-uniform samples suboptimal discrepancy. The existing methods for generating low-discrepancy points focus only on uniform points, thus generating non-uniform and low-discrepancy samples has been an open problem. We introduce an optimization framework to approximate non-uniform low-discrepancy samples by numerically minimizing the discrepancy metric for non-uniform samples. We propose a differentiable numerical estimator of the generalized non-uniform discrepancy to enable gradient-based optimization. Our differentiable discrepancy estimator turns a challenging problem of generating non-uniform low-discrepancy samples into a simple numerical optimization problem. Since optimization can generate only one set of samples, we further propose a neural network architecture to encode the sample generation process, allowing us to generate multiple sets of samples from conditional targets without runtime optimization. We experimentally validate that our method reduces discrepancy below prior methods at least up to a certain sample size and consequently achieves lower integration error on numerical integration problems.
Web recommendations provide personalized items from massive catalogs for users, which rely heavily on retrieval stages to trade off the effectiveness and efficiency of selecting a small relevant set from billion-scale candidates in online digital platforms. As one of the largest Chinese search engine and news feed providers, Baidu resorts to Deep Neural Network (DNN) and graph-based Approximate Nearest Neighbor Search (ANNS) algorithms for accurate relevance estimation and efficient search for relevant items. However, current retrieval at Baidu fails in comprehensive user-item relational understanding due to dissected interaction modeling, and performs inefficiently in large-scale graph-based ANNS because of suboptimal traversal navigation and the GPU computational bottleneck under high concurrency. To this end, we propose a GPU-accelerated Multi-relational Parallel Graph Retrieval (GMP-GR) framework to achieve effective yet efficient retrieval in web-scale recommendations. First, we propose a multi-relational user-item relevance metric learning method that unifies diverse user behaviors through multi-objective optimization and employs a self-covariant loss to enhance pathfinding performance. Second, we develop a hierarchical parallel graph-based ANNS to boost graph retrieval throughput, which conducts breadth-depth-balanced searches on a large-scale item graph and cost-effectively handles irregular neural computation via adaptive aggregation on GPUs. In addition, we integrate system optimization strategies in the deployment of GMP-GR in Baidu. Extensive experiments demonstrate the superiority of GMP-GR in retrieval accuracy and efficiency. Deployed across more than twenty applications at Baidu, GMP-GR serves hundreds of millions of users with a throughput exceeding one hundred million requests per second.
Traffic prediction plays a pivotal role in contemporary web technologies, motivating various intelligent web services such as route planning and remote traffic management. Many recent proposals that target deep learning for traffic prediction solely leverage historical traffic observations to predict future ones. However, traffic prediction is always susceptible to different factors such as road networks and social events, exhibiting different modalities. Most existing methods focus on a single modality, failing to capture the comprehensive traffic patterns among various factors, resulting in sub-optimal performance. Web-sourced geo-images, e.g., satellite imagery, encompass comprehensive contextual information and offer an effective way to represent diverse modalities. To unleash the power of such geo-images, we propose VisionST, a Vision-augmented Spatial-Temporal Neural Network, which coordinates cross-modal traffic prediction with interactive geo-image encoding. To bolster resilience against highly intricate and overlapping traffic patterns, VisionST features a visual semantic extraction mechanism and a pattern-guided aggregation mechanism. The former extracts node-level visual tokens and node-to-node visual relation patterns from geo-referenced images. The latter generates relation patterns that encompass visual, spatial, and temporal aspects, constraining nodes to interact with these relation patterns for contextual information interaction. Extensive experiments on real large-scale datasets offer insight into the effectiveness of the proposed solutions, showing that VisionST consistently outperforms state-of-the-art baselines.
Rapid development of web services has led to an explosion of multimodal content, making multimodal recommender systems (MRSs) vital tools for mitigating information overload. Current MRSs have achieved remarkable progress by incorporating advanced technologies such as Graph Neural Networks (GNNs) and Large Language Models (LLMs). However, these studies still suffer from the semantic shift problem. Generally, item's multimodal content usually contain multiple objects, including target object (core content of item) and auxiliary objects (decorations of item). Existing MRSs overlooked this distinction, failing to prevent auxiliary objects from dominating the representation, leading to biased item representation. To address this issue, we propose a model-agnostic framework ''TargetMR''. Concretely, TargetMR comprises two core modules, including Object Disentangler and Object Identifier. The Object Disentangler decouples item text and image into multiple objects via text syntactic parsing and image segmentation. The Object Identifier performs knowledge distillation based on LLMs to efficiently identify the target text object. It then identifies the target image object through cross-modal semantic evaluation. Moreover, this module refines the representation of image target object by optimizing the semantic correlation. Owing to the model-agnostic design of TargetMR, it can be integrated into various backbone MRSs. Extensive experiments on three benchmark datasets show that TargetMR consistently improves the performance of five backbone MRSs, with an average improvement of 12.26%. Our codes are available at https://github.com/gutang-97/TargetMR/.
Embedded AI applications usually require compact on-device models that can continually adapt to new tasks. However, recent studies have revealed that neural networks trained on non-stationary data streams gradually lose their ability to adapt to new tasks, a phenomenon known as plasticity loss. Moreover, to enable neural networks to run on resource-constrained embedded devices, model pruning is commonly applied for model compression, which may further affect their plasticity. To conduct efficient model adaptation on new tasks on embedded devices, we propose Plasticity-aware Continual Pruning (PaCP), a novel framework that operates in two stages. First, a pre-deployment stage uses a plasticity-aware strategy to prune the model while optimizing its initial structure for future adaptability. Second, during continual learning, the model's capacity is temporarily expanded at task boundaries to efficiently learn new information, before plasticity-aware pruning restores its compact form. Extensive experiments on multiple continual learning benchmarks demonstrate that PaCP significantly outperforms existing plasticity-maintenance methods and, remarkably, even surpasses non-pruned models lacking explicit plasticity preservation.
Scholarly web is a vast network of knowledge connected by citations. However, this system is increasingly compromised by miscitation, where references do not support or even contradict the claims they are cited for. Current miscitation detection methods, which primarily rely on semantic similarity or network anomalies, struggle to capture the nuanced relationship between a citation's context and its place in the wider network. While large language models (LLMs) offer powerful capabilities in semantic reasoning for this task, their deployment is hindered by hallucination risks and high computational costs. In this work, we introduce LLM-Augmented Graph Learning-based Miscitation Detector (LAGMiD), a novel framework that leverages LLMs for deep semantic reasoning over citation graphs and distills this knowledge into graph neural networks (GNNs) for efficient and scalable miscitation detection. Specifically, LAGMiD introduces an evidence-chain reasoning mechanism, which uses chain-of-thought prompting, to perform multi-hop citation tracing and assess semantic fidelity. To reduce LLM inference costs, we design a knowledge distillation method aligning GNN embeddings with intermediate LLM reasoning states. A collaborative learning strategy further routes complex cases to the LLM while optimizing the GNN for structure-based generalization. Experiments on three real-world benchmarks show that LAGMiD achieves state-of-the-art miscitation detection with significantly reduced inference cost.
The explosive growth of multimodal web data demands communication that transmits meaning rather than raw bits. Existing semantic-communication systems often fail under noise, missing modalities, and distribution shifts because they optimize surface features instead of modality-invariant knowledge. We present Grasp, a knowledge-centric framework for cross-modal communication. Grasp segments streams into semantic blocks and builds a graph over them; a lightweight Graph Neural Networks (GNN) produces schedulable, importance-weighted representations. At its core is knowledge purification : we minimize a conditional mutual information upper bound to perform a three-way disentanglement—strongly related, weakly related, and task-irrelevant components—so that only essential semantics are transmitted while non-essential factors are suppressed. To maintain synchrony, we introduce one-to-two temporal contrastive learning to achieve triple alignment of video, audio, and text despite sampling asynchrony. For efficient transmission, Grasp uses a cross-modal shared vector-quantization codebook—a discrete knowledge codebook —updated by multimodal attention. At the receiver, a soft-recovery mechanism leverages this shared knowledge to robustly reconstruct semantics under low signal-to-noise ratio (SNR) or missing modalities, yielding graceful degradation. Across web tasks—including cross-modal retrieval and missing-modality inference—Grasp improves knowledge consistency, semantic fidelity, and downstream performance over strong baselines while maintaining low latency. These results show that communication structured around purified knowledge is key to building robust, semantic-aware systems for the modern web.
Graph Neural Networks (GNNs) have demonstrated remarkable performance across web-centric applications, yet they are vulnerable to sophisticated adversarial attacks, particularly node injection attacks. The success of such attacks heavily relies on their stealthiness, the ability to blend in with the original graph and evade detection. However, existing methods often achieve stealthiness by relying on indirect proxy metrics, lacking consideration for the fundamental characteristics of the injected content, or focusing only on imitating local structures, which leads to the problem of local myopia. To overcome these limitations, we propose a dual-constraint stealthy node injection framework, called Joint Alignment of Nodal and Universal Structures (JANUS). At the local level, we introduce a local feature manifold alignment strategy to achieve geometric consistency in the feature space. At the global level, we incorporate structured latent variables and maximize the mutual information with the generated structures, ensuring the injected structures are consistent with the semantic patterns of the original graph. We model the injection attack as a sequential decision process, which is optimized by a reinforcement learning agent. Experiments on multiple standard datasets demonstrate that the JANUS framework significantly outperforms existing methods in terms of both attack effectiveness and stealthiness. Our code and scripts are available at https://github.com/jiahaozhang121/JANUS-WWW2026.
Space-based Parameter Evolving with Lightweight Optimization for Graph Adaptation to Evolving Shifts
Adapting graph neural networks to evolving domain shifts presents a fundamental challenge: how to acquire new knowledge while preventing catastrophic forgetting. Existing continual learning methods often rely on memory replay or complex regularization schemes, incurring significant computational overhead. We propose STEM (State-based Parameter Evolving with Lightweight Optimization), a replay-free framework that transforms continual adaptation into controlled parameter space evolution via a controller-worker architecture. At its core is a Test-time Evolving State Space (TESS) controller with a selective gating mechanism that recursively updates its hidden state by integrating compact summaries of the current graph domain. Unlike traditional linear state space models, TESS enables nonlinear, input-dependent state transitions that capture temporal dynamics of domain evolution. A lightweight parameter generator decodes this evolving state into domain-specific adapter parameters injected into a frozen base GNN. We employ unsupervised Information Maximization and parameter space stability regularization that penalizes adapter changes across time steps, with theoretical guarantees of forgetting mitigation and stable convergent adaptation. Extensive experiments validate that our method achieves state-of-the-art performance while maintaining minimal computational overhead. The code is available at https://github.com/miaomiao1220/stem
Temporal Betweenness Centrality (TBC) measures how often a node appears on optimal temporal paths, reflecting its importance in temporal networks. However, exact computation is highly expensive, and real-world TBC distributions are extremely imbalanced, causing learning-based models to overfit to zero-centrality nodes and fail to identify truly central nodes. Existing graph neural networks (GNNs) either ignore temporal dependencies or cannot handle such extreme imbalance. To address these issues, we propose CLGNN, a scalable and inductive contrastive learning-based GNN for accurate TBC prediction. CLGNN preserves temporal path validity through an instance graph and encodes structural, path-time aware dependencies via dual aggregation. To mitigate imbalance, a stability-based clustering-guided contrastive module separates nodes of different centrality levels in representation space, while a regression head estimates TBC values. Extensive experiments on diverse benchmarks demonstrate that CLGNN is scalable, generalizable, and effective.
Ensuring the trustworthiness of graph neural networks (GNNs), which are often treated as black-box models, requires effective explanation techniques. Existing GNN explanations typically apply input perturbations to identify subgraphs that are responsible for the occurrence of the final output of GNNs. However, such approaches lack finer-grained, layer-wise analysis of how intermediate representations contribute to the final result, capabilities that are crucial for model diagnosis and architecture optimization. This paper introduces SliceGX, a novel GNN explanation approach that generates explanations at specific GNN layers in a progressive manner. Given a GNN model M, a set of selected intermediate layers, and a target layer, SliceGX slices M into layer blocks (''model slice'') and discovers high-quality explanatory subgraphs within each block that elucidate how the model output arises at the target layer. Although finding such layer-wise explanations is computationally challenging, we develop efficient algorithms and optimization techniques that incrementally construct and maintain these subgraphs with provable approximation guarantees. Extensive experiments on synthetic and real-world benchmarks demonstrate the effectiveness and efficiency of SliceGX, and illustrate its practical utility insupporting model debugging.
Explaining the decision-making process of Graph Neural Networks (GNNs) is essential for improving their transparency and reliability. However, real-world graphs are often heterogeneous and subject to structural noise, posing severe challenges to the robustness of existing explanation methods. To address these issues, we propose RoHeX, a Robust Heterogeneous GNN Explainer that enhances explanation quality under noisy conditions. RoHeX begins with a theoretical analysis revealing how different heterogeneous GNN architectures amplify structural perturbations through message passing. Building on this insight, we design a denoising variational inference framework that filters noisy structures and learns robust latent graph representations. Furthermore, we incorporate relation-aware heterogeneous semantics into the explanation generation process, formulating explanation as an optimization problem under the graph information bottleneck principle. This formulation enables RoHeX to balance fidelity and compactness, producing explanations that are both semantically meaningful and structurally stable. Comprehensive experiments on multiple real-world heterogeneous graphs demonstrate that RoHeX consistently surpasses state-of-the-art baselines in explanation fidelity, robustness to structural perturbations, and explainability.
Identification of Influential Node Group in Attributed Graph through Explaining Graph Neural Network
Identification of influential groups of nodes in attributed graphs has applications in a wide range of real-world problems, for instance, collecting important proceedings in citation networks, or identifying essential genes for diagnosing disease in Protein-Protein Interaction networks. Previous approaches for influence maximization manipulated on the graph structure, despite their proliferation, neglect the node attribute information containing additional knowledge. In this work, we introduce Global Graph UNderstanding (GGUN), a perturbation-based framework leveraging the explanatory power of Graph Neural Networks. It takes into account the entire graph structure and node attributes simultaneously and fuses knowledge through GNN layers. Following the perturbation-based explanation, GGUN fills the gap between Deep Neural Network gradient-based feature importance analysis and discrete structure in the graph, which is formulated as a combinatorial optimization problem. Moreover, GGUN obtains an efficient solution by relaxing the infeasible combinatorial optimization problem with performance guaranteed. Evaluations of synthetic and real-world datasets show that GGUN outperforms baselines on both quantitative metrics and human-intelligible analysis.
Graph Anomaly Detection (GAD) is critical in applications such as fraud prevention, cybersecurity, and social governance. While Graph Neural Networks (GNNs) have achieved remarkable success in detecting anomalies on homogeneous graphs, they face fundamental challenges in real-world heterogeneous settings involving diverse node types and imbalanced semantic richness. In heterogeneous graphs, nodes often vary significantly in semantic richness, with anomalies potentially spanning multiple types and emerging implicitly through cross-type dependencies. We identify two core limitations of existing methods: (i) the ineffective propagation of discriminative anomaly cues from informative to sparse nodes due to semantic imbalance, and (ii) conflicting optimization objectives arising from joint detection across multiple node types. To address these issues, we propose CSA-MTHGAD, a novel framework that integrates smoothness-guided cross-type semantic alignment with dynamic multi-task learning. It selectively propagates anomaly-sensitive features across types and harmonizes task-specific gradients through adaptive projection and weighting.To facilitate research, we employ two real-world heterogeneous benchmarks in the domain of social governance. Extensive experiments demonstrate that CSA-MTHGAD achieves superior performance over state-of-the-art baselines in accuracy, robustness, and generalization for multi-type anomaly detection.
Many real-world systems can be modeled as graphs, where nodes store and consume entities, actively produce them, or have them emerge naturally, and edges transport them between nodes. This paper studies the networked control problem on such large-scale systems, aiming to decide production and transportation over time to maximize long-term profits, subject to node or edge capacity constraints. Existing SOTAs either fail to guarantee feasibility or cannot scale to large-scale systems. We propose a two-stage policy that integrates a constrained optimization layer after a neural network to explicitly enforce constraints and ensure feasibility. By leveraging the problem structure to obtain expert actions and designing a decision-focused and differentiable loss to enable imitation learning, our method significantly improves efficiency and scalability. In small-scale systems with action dimensions in the order of 10, our method achieves 60x sample efficiency over SOTAs on average. In large-scale systems with action dimensions ranging from 100 to 100000, where SOTAs fail to train, our method converges quickly and outperforms non-learning-based baselines significantly.