A new arXiv preprint examines recurrent graph neural networks, which repeatedly run message passing until it converges. The authors note that existing logical characterizations of these models rely on multi-set aggregation, counting-based graded logics, and halting or acceptance conditions that cannot be verified from the network's parameters.
A new arXiv preprint addresses node classification on evolving graphs, where classifiers must generalize to newly arriving nodes despite distribution shift while also providing calibrated uncertainty. The authors propose an online Bayesian approach aimed at inductive settings where safety-sensitive applications require trustworthy confidence estimates. The work targets a gap left by standard graph neural networks, which typically assume static graphs and offer limited uncertainty quantification.
A new arXiv paper proposes using geometric flow techniques to enhance graph coarsening, a pooling step used in graph convolutional networks to cut computational cost. The approach aims to reduce the expense of graph pooling operations that mirror pooling in standard convolutional networks. The work is a preprint and reports on method design rather than deployed results.
A new arXiv preprint introduces HGTO, a unified graph-based, physics-informed framework for density-based structural topology optimization. The approach reframes the usual nested loop of material updates, structural analysis, and sensitivity computation, drawing on neural density parameterization and dual-field physics-informed methods that require no labeled data. The abstract positions the work as a data-free alternative to conventional optimization pipelines.
A revised arXiv paper argues that neural algorithmic reasoning, which typically trains networks to imitate classic algorithms via supervised learning, is limited by its reliance on post-processing to produce valid outputs. The authors propose reformulating the task as a reinforcement learning problem, aiming to let models build correct solutions directly rather than repairing them afterward. The work appears in the cs.LG and cs.AI listings as a replacement submission.
A new arXiv preprint proposes deformable graph neural networks built on positional encoding to tackle long-standing limits of message passing over first-order neighbors. The authors note that conventional GNNs struggle as depth increases, leading to over-smoothing and difficulty capturing long-range structure. The work targets graph-structured data in real-world settings where deeper models are needed.
A new arXiv paper addresses a limitation in graph neural networks used for spatio-temporal traffic prediction, which often rely only on sensor proximity or road-network topology. The authors propose a method that injects general semantic knowledge into the model to improve forecasting accuracy. The work is a revised submission (v2) to arXiv's machine learning category.
Researchers introduce LiftGCN, a graph learning method designed to predict finite element stress fields that contain sharp gradients near holes, notches and load points. The approach uses a Joukowski spectral lifting transform to preserve energy and retain high-frequency graph components that standard graph neural networks tend to smooth away. The work is posted as an arXiv preprint in computer science categories.
A revised arXiv preprint presents a graph neural network approach for predicting links in temporal signed networks, which capture how cooperative and adversarial relationships evolve over time. The authors motivate the work with applications including social media analysis, trust and reputation systems, and financial transaction networks. The paper's abstract excerpt focuses on the method's design for handling dynamic, sign-aware graph structure.
A new arXiv paper proposes a molecular representation learning approach that combines multiple views rather than relying on a single one. It pairs hierarchical graph modeling of atom-bond topology with contextualized molecular fingerprints to improve property prediction. The goal is representations that generalize from limited labeled data to structurally novel compounds.
A new arXiv paper examines how persistent Gaussian noise affects recurrent graph neural networks. While such perturbations are known to keep a positive stationary Dirichlet energy and thus avoid asymptotic oversmoothing, the authors show that this global bound alone does not ensure individual nodes stay distinguishable. The work analyzes finite-time separation between node representations under these random perturbations.
A new arXiv paper introduces HiGFRL, a method that fuses hierarchical graph representations with reinforcement learning to schedule tasks with dependencies across heterogeneous cloud clusters. The authors target the difficulty of jointly handling DAG structure and multi-dimensional resource limits in online settings, which they say existing deep RL schedulers address only partially. The work is a preprint and has not yet been peer reviewed.
A new arXiv paper proposes an unsupervised graph neural network approach to the Minimum Dominating Set problem, an NP-hard combinatorial task. The method is aimed at applications such as influence maximization in social networks, viral marketing, and public health interventions. The work frames dominating set selection as a learning problem that does not require labeled optimal solutions.
A new arXiv preprint introduces a method for semantic image communication that uses channel-adaptive region adjacency graphs as carriers instead of dense latent tensors or grid-aligned semantic layouts. The approach aims to explicitly encode relationships between image regions so that task-relevant scene structure survives tight channel budgets. The abstract frames the work as addressing a gap in how region-level relations are represented under limited bandwidth.
A new arXiv preprint introduces an instance-aware graph modeling approach for detecting and classifying cells in pathology images. The method aims to capture complex cellular interactions within the tumor microenvironment rather than relying only on visual appearance. Accurate cell detection matters for diagnostic accuracy and treatment planning.
A new arXiv paper examines how different pre-training approaches affect graph transformer performance on biochemistry tasks. The authors report that pre-training with supervision, using computed molecular properties as labels, outperformed the other strategies tested. The finding comes from a set of comparative experiments run in that domain.
A new arXiv paper proposes a hierarchical reinforcement learning method that uses graph attention to schedule workflows in cloud environments. The approach targets three competing goals at once: meeting deadlines, improving container utilization, and lowering energy use. It also accounts for unpredictable task runtimes, communication costs that depend on where tasks are placed, and the need to decide task assignment and container selection together.
A new arXiv paper proposes GNN4PPM, which applies relational graph convolutional networks to predictive process monitoring. The method targets several predictions at once, such as the next event in a running process, the time remaining until a trace finishes, and its eventual outcome. The authors argue that existing techniques typically address only one of these targets; the posted abstract is truncated before any experimental results are described.
A new arXiv paper introduces GTFD, a graph-transformer approach for spotting fraud in corporate transaction networks. The method combines self-supervised pretraining with conformal risk control to handle coordinated fraud that rule-based systems and per-transaction models miss. It is positioned as an improvement over classical models that analyze each transaction in isolation.
Researchers have developed a graph neural operator that estimates full mid-plane stress tensor fields in concrete during projectile penetration, connecting the material's mesoscale structure to complete field-level predictions. The surrogate was trained on data from a detailed aggregate-resolved LS-DYNA simulation and is designed to generalize across different impact velocities, offering a faster stand-in for expensive finite-element computations.
Researchers present a dual-channel graph neural architecture that predicts which exact solver will perform best on a given maximum clique problem instance. Since no single solver dominates across all types of graphs, the approach learns from graph characteristics to make per-instance algorithm choices. The paper appears on arXiv in both the AI and machine learning categories.
A new arXiv paper presents an efficient graph neural network method for optimizing multicarrier wideband hybrid beamforming, a key technique for highly directional 6G links. The work targets beam squint, a distortion that grows as 6G systems use much wider frequency bands and that traditionally requires costly true-time-delay filters. The authors position the learning-based approach as a more efficient alternative for next-generation wireless systems.
A revised preprint examines how the modular structure of relational graphs interacts with the depth of neural networks when learning from graph-structured data. The author situates the work within graph-based machine learning, including graph neural networks and reinforcement learning, and analyzes how graph structure shapes learning outcomes. The version posted is an update to an earlier draft.
A research paper introduces WaveGraphNet, a graph-based approach that localizes damage in composite plates from sparse networks of piezoelectric sensors using guided waves. By jointly training inverse and forward models under physics-consistency constraints, the method aims to overcome the weak supervision that limits standard approaches when only a few sensor measurements are available.