# graph theory

Graph theory is a branch of discrete combinatorial mathematics that studies the properties of graphs and has applications including computer science.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## On the Informativeness of Structural Stability Bounds for GNNs: GINs versus GATs

DevFeed: [On the Informativeness of Structural Stability Bounds for GNNs: GINs versus GATs](<https://devfeed.tech/articles/on-the-informativeness-of-structural-stability-bounds-for-gnns-gins-versus-gats-145076.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25603>)

Author: Alan Gany, Paul Landrier, Bogdan Cautis, Laks V. S. Lakshmanan, Silviu Maniu

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [model architecture](<https://devfeed.tech/topics/model-architecture.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [arxiv](<https://devfeed.tech/tags/arxiv.md>), [attention](<https://devfeed.tech/tags/attention.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-lo](<https://devfeed.tech/tags/cs-lo.md>), [graph](<https://devfeed.tech/tags/graph.md>), [graph-isomorphism](<https://devfeed.tech/tags/graph-isomorphism.md>), [graph-neural-networks](<https://devfeed.tech/tags/graph-neural-networks.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [node](<https://devfeed.tech/tags/node.md>)

### AI overview

The structural stability bound is informative for GINs under a condition involving perturbation size, network depth, and maximum degree; it is never tighter than a graph-independent output-diameter bound for single-head GATv1. The authors derive bounds for deterministic edge perturbations that need not be small. Their condition holds with a wide margin on Cora but fails beyond a few layers on low-degree MUTAG graphs. For GATv1, the result motivates analyzing stability through attention masses.

### Source excerpt

arXiv:2611.25603v1 Announce Type: new Abstract: Stability bounds for graph neural networks limit how much node representations can drift when the graph changes, but they are rarely compared with the simplest alternative: a bound on the distance between any two outputs, whatever the graph. We derive structural stability bounds for Graph Isomorphism Networks (GINs) and single-head GATv1 networks under deterministic edge perturbations that need not be small, and compare each with such a perturbation-independent diameter bound. For GINs, we characterize exactly when the stability bound is tighter, through a condition linking perturbation size, depth and maximum degree; it holds with a wide margin on Cora, but fails beyond a few layers on the low-degree MUTAG graphs. For GATv1, the bound is never tighter, which calls for analyses based on attention masses.

## AdsNet: A Graph Neural Network for Direct Adsorption Energy Prediction

DevFeed: [AdsNet: A Graph Neural Network for Direct Adsorption Energy Prediction](<https://devfeed.tech/articles/adsnet-a-graph-neural-network-for-direct-adsorption-energy-prediction-145085.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25646>)

Author: Changzhi Ai, Frank Abild-Pedersen, Johannes Voss

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [3D](<https://devfeed.tech/topics/3d.md>), [Hyperparameter optimization](<https://devfeed.tech/topics/hyperparameter-optimization.md>), [concept-learning](<https://devfeed.tech/topics/concept-learning.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [algorithm](<https://devfeed.tech/tags/algorithm.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [automatic](<https://devfeed.tech/tags/automatic.md>), [catalyst](<https://devfeed.tech/tags/catalyst.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [computational-complexity](<https://devfeed.tech/tags/computational-complexity.md>), [cond-mat-mtrl-sci](<https://devfeed.tech/tags/cond-mat-mtrl-sci.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [contributions](<https://devfeed.tech/tags/contributions.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [dft](<https://devfeed.tech/tags/dft.md>), [energy](<https://devfeed.tech/tags/energy.md>), [fast](<https://devfeed.tech/tags/fast.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gradient](<https://devfeed.tech/tags/gradient.md>), [gradient-descent](<https://devfeed.tech/tags/gradient-descent.md>), [graph](<https://devfeed.tech/tags/graph.md>), [neural](<https://devfeed.tech/tags/neural.md>), [neural-network](<https://devfeed.tech/tags/neural-network.md>), [physics-comp-ph](<https://devfeed.tech/tags/physics-comp-ph.md>), [prediction](<https://devfeed.tech/tags/prediction.md>)

### AI overview

AdsNet predicts adsorption energies in one forward pass using an equivariant message-passing network over adsorbate atoms and nearby surface atoms within a cutoff. Its local graph uses distances and directional vectors, so computation is decoupled from total system size while retaining sensitivity to local geometry. On 33,920 DFT structures covering 26 transition metals and six adsorbates, it reports an energy RMSE of 0.032 eV and force RMSE of 0.047 eV/Å. Automatically differentiated forces support projected-gradient relaxation with adaptive line search, and predicted energies can be decomposed into individual local-graph atom contributions. These results are reported for the described dataset; the excerpt does not provide further validation details.

### Source excerpt

arXiv:2611.25646v1 Announce Type: cross Abstract: Fast and accurate prediction of adsorption energies is critical for high-throughput computational catalyst screening, existing methods compromise either on evaluation speed (full-graph machine learning potentials) or geometric sensitivity (coordination-based models like the alpha-scheme). Here, we present AdsNet, an equivariant graph neural network that directly predicts adsorption energies in a single forward pass, which operates on a local graph restricted only to adsorbate atoms and its immediate surface neighbors within a cutoff radius. This local graph formulation decouples computational complexity from total system size while encoding continuous geometric features (interatomic distances and directional vectors) enabling sensitivity to subtle variations in local bonding environments. AdsNet adopts a message-passing architecture with scalar and vector node features over multiple interaction layers. Trained on AdsTraj, a dataset of 33,920 density functional theory (DFT) calculated structures spanning 26 transition metals and 6 adsorbates (H*, C*, O*, CO*, OH*, HOCO*), the model achieves an adsorption energy root-mean-square error (RMSE) of 0.032 eV and a force RMSE of 0.047 eV/A. Analytic forces obtained via automatic differentiation enable structure relaxation through a projected gradient descent algorithm with adaptive line search, eliminating entire geometry optimizations. Additionally, AdsNet uniquely decomposes predicted adsorption energies into contributions from individual atoms within the local adsorbate graph - an interpretability feature unattainable with conventional machine-learning potentials. AdsNet thus establishes an accurate, highly efficient, and interpretable framework for large-scale adsorption energy prediction and local interface relaxation across various material surfaces.

## Scone: Symmetry-aware Neural Combinatorial Optimization for Graph Coloring

DevFeed: [Scone: Symmetry-aware Neural Combinatorial Optimization for Graph Coloring](<https://devfeed.tech/articles/scone-symmetry-aware-neural-combinatorial-optimization-for-graph-coloring-145032.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25273>)

Author: Paolo Pellizzoni, Melanie Weber

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Combinatorial optimization](<https://devfeed.tech/topics/combinatorial-optimization.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [combinatorial-optimization](<https://devfeed.tech/tags/combinatorial-optimization.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [general](<https://devfeed.tech/tags/general.md>), [graph](<https://devfeed.tech/tags/graph.md>), [graph-coloring](<https://devfeed.tech/tags/graph-coloring.md>), [inference](<https://devfeed.tech/tags/inference.md>), [information](<https://devfeed.tech/tags/information.md>), [learning](<https://devfeed.tech/tags/learning.md>), [model](<https://devfeed.tech/tags/model.md>), [neural-network](<https://devfeed.tech/tags/neural-network.md>), [node](<https://devfeed.tech/tags/node.md>), [np-hard](<https://devfeed.tech/tags/np-hard.md>), [permutations](<https://devfeed.tech/tags/permutations.md>), [policy](<https://devfeed.tech/tags/policy.md>), [quality](<https://devfeed.tech/tags/quality.md>), [space](<https://devfeed.tech/tags/space.md>), [structure](<https://devfeed.tech/tags/structure.md>), [symmetry](<https://devfeed.tech/tags/symmetry.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [times](<https://devfeed.tech/tags/times.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

Scone uses a color-equivariant graph neural network to build graph colorings sequentially. It exploits the fact that permuting color labels does not change coloring validity, reducing the effective state space, while propagating earlier assignments so the model can break node symmetries when adjacent nodes need different colors. The abstract reports perfect solution quality with low inference times across several tasks, including sudoku_extreme and register allocation, and generalization to graph families absent from training. It provides no benchmark details or quantitative comparisons.

### Source excerpt

arXiv:2611.25273v1 Announce Type: new Abstract: Graph coloring, which entails assigning colors to nodes such that no adjacent nodes have the same color, is a fundamental combinatorial optimization problem. In general, however, finding a node coloring with no monochromatic edges given a fixed palette is NP-hard. This motivates learning-based methods that can learn to exploit recurring structure across instances to efficiently produce high-quality colorings. Crucially, the validity of a coloring is invariant to permutations of the colors, and we show that exploiting this symmetry reduces the size of the effective state space over which the model has to learn its policy. At the same time, the coloring algorithm must be able to break some symmetries, for example to assign different colors to two adjacent nodes belonging to the same automorphism orbit. In this paper, we introduce Scone, a model that uses a color-equivariant graph neural network to construct colorings sequentially, allowing information from previous assignments to propagate and break node symmetries. Scone achieves perfect solution quality with low inference times across several graph coloring tasks, including on sudoku_extreme and register allocations, and showcases generalization to graph families not encountered during training.

## From Sequence Uncertainty to Phenotype Hypotheses: Uncertainty-Conditioned Inference over Biomedical Knowledge Graphs

DevFeed: [From Sequence Uncertainty to Phenotype Hypotheses: Uncertainty-Conditioned Inference over Biomedical Knowledge Graphs](<https://devfeed.tech/articles/from-sequence-uncertainty-to-phenotype-hypotheses-uncertainty-conditioned-inference-over-biomedical-knowledge-graphs-145053.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25422>)

Author: Huixin Zhan, Jason H. Moore

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [база знаний](<https://devfeed.tech/topics/tag-c51cc4bf76e5.md>), [Inference](<https://devfeed.tech/topics/inference.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [biomedical](<https://devfeed.tech/tags/biomedical.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [distributions](<https://devfeed.tech/tags/distributions.md>), [embedding](<https://devfeed.tech/tags/embedding.md>), [energy](<https://devfeed.tech/tags/energy.md>), [entropy](<https://devfeed.tech/tags/entropy.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [framework](<https://devfeed.tech/tags/framework.md>), [hypothesis](<https://devfeed.tech/tags/hypothesis.md>), [inference](<https://devfeed.tech/tags/inference.md>), [knowledge-graphs](<https://devfeed.tech/tags/knowledge-graphs.md>), [masks](<https://devfeed.tech/tags/masks.md>), [probability](<https://devfeed.tech/tags/probability.md>), [q-bio-gn](<https://devfeed.tech/tags/q-bio-gn.md>), [query](<https://devfeed.tech/tags/query.md>), [recall](<https://devfeed.tech/tags/recall.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [sequence](<https://devfeed.tech/tags/sequence.md>), [supervision](<https://devfeed.tech/tags/supervision.md>), [support](<https://devfeed.tech/tags/support.md>), [temperature](<https://devfeed.tech/tags/temperature.md>), [uncertainty](<https://devfeed.tech/tags/uncertainty.md>)

### AI overview

The framework turns genomic sequence predictions and calibrated uncertainty into probabilities over graph-supported phenotype hypotheses, without patient-level phenotype supervision. It scores admissible gene-disease-phenotype paths with a learned energy, uses query uncertainty to set the target temperature, and aggregates path probabilities while preserving traceable support. On AlzKB and Hetionet, phenotype entropy tracked upstream uncertainty (Spearman 0.951 and 0.878). Uncertainty-dependent temperature training improved held-out recall@10 over a comparable constant temperature. In gene hold-out, recall@10 exceeded trained DistMult on 24 of 30 graph masks (0.183 vs. 0.123), but the difference was driven mainly by three genes with many variants and was not established for all genes.

### Source excerpt

arXiv:2611.25422v1 Announce Type: new Abstract: Genomic foundation models provide variant-level evidence, but connecting that evidence to phenotype hypotheses requires accounting for uncertainty throughout structured reasoning. We introduce an uncertainty-conditioned path-energy framework that converts sequence predictions and calibrated predictive uncertainty into explicit distributions over graph-supported phenotype hypotheses, without patient-level phenotype supervision. An interpretable learned energy scores admissible gene--disease--phenotype routes, query uncertainty controls the target temperature, and path probabilities are aggregated into phenotype hypotheses. This formulation makes evidence quality an explicit determinant of hypothesis breadth while retaining traceable multi-path support. Exact normalization over enumerable candidate sets provides a reference for evaluating a Hastings-corrected learned revision operator. On AlzKB and Hetionet, phenotype-level entropy tracks upstream uncertainty, with Spearman correlations of 0.951 and 0.878, respectively; ablations identify the contribution of uncertainty conditioning. In masked-edge recovery, the model exceeds chance and degree-matched probability-permutation controls. Training the energy under the uncertainty-dependent temperature yields higher held-out recall@10 than training under a constant temperature of comparable magnitude. Under gene hold-out, recall@10 exceeds that of a trained DistMult embedding on 24 of 30 graph masks (0.183 versus 0.123); this difference is attributable mainly to three genes with many variants and is not established for all genes. These results establish an uncertainty-aware interface from genomic predictions to graph-supported phenotype hypotheses that combines interpretable probability allocation with held-out association recovery of explicitly stated scope.

## A Sparse Low-Rank Biclique Decomposition for Graphs

DevFeed: [A Sparse Low-Rank Biclique Decomposition for Graphs](<https://devfeed.tech/articles/a-sparse-low-rank-biclique-decomposition-for-graphs-145107.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25826>)

Author: Antoine Vialle, Sergei Gerasimov, Aref Einizade, Antonio Ortega, Fragkiskos D. Malliaros, Jhony H. Giraldo

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [algorithm](<https://devfeed.tech/tags/algorithm.md>), [block](<https://devfeed.tech/tags/block.md>), [compression](<https://devfeed.tech/tags/compression.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [greedy](<https://devfeed.tech/tags/greedy.md>), [low-rank](<https://devfeed.tech/tags/low-rank.md>)

### AI overview

Signed biclique decomposition (SBD) represents an integer-weighted graph as a sum of sparse signed binary rank-one factors plus a sparse signed residual, keeping the low-rank structure discrete and sparse. The paper formulates fixed-rank and greedy objectives and derives a gain-based greedy algorithm with theoretical guarantees. The abstract reports recovery of block structure in noisy graphs and, on real graphs with strong shared-neighborhood structure, compression factors above 10 and sparse matrix-matrix multiplication speedups up to 6x over cuSPARSE. These results suggest SBD as an alternative to dense low-rank decompositions for graphs; the supplied abstract does not specify broader performance conditions.

### Source excerpt

arXiv:2611.25826v1 Announce Type: new Abstract: Low-rank plus sparse decompositions, such as robust PCA, typically model the low-rank component as dense and the sparse component as corruption. This viewpoint is poorly suited to graph adjacency matrices, where sparsity is structural and dense low-rank factors destroy the combinatorial and computational properties of the graph. We propose signed biclique decomposition (SBD), a sparse low-rank decomposition for integer-weighted graphs that represents the structured component as a sum of sparse signed binary rank-one factors, together with a sparse signed residual. The resulting representation remains discrete, interpretable, sparse, and compatible with standard linear algebra. We formulate fixed-rank and greedy SBD objectives and derive a gain-based greedy algorithm with theoretical guarantees. Empirically, SBD recovers block structure in noisy graphs, while its factorized representation enables efficient sparse matrix-matrix multiplication (SpMM). On real graphs with strong shared-neighborhood structure, SBD achieves compression factors above $10$ and SpMM speedups up to $6\times$ over cuSPARSE, suggesting a practical alternative to dense low-rank decompositions for graphs. Code is available at https://github.com/IAntoineV/sbd.

## Multilevel Graph Partitioning on GPU Supercomputers

DevFeed: [Multilevel Graph Partitioning on GPU Supercomputers](<https://devfeed.tech/articles/multilevel-graph-partitioning-on-gpu-supercomputers-142978.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24264>)

Author: Michael S. Gilbert, Erik G. Boman, Kamesh Madduri

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [Multi-GPU](<https://devfeed.tech/topics/multi-gpu.md>), [High-Performance Computing](<https://devfeed.tech/topics/high-performance-computing.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>), [GPU](<https://devfeed.tech/topics/gpu.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [graph](<https://devfeed.tech/tags/graph.md>), [multi-gpu](<https://devfeed.tech/tags/multi-gpu.md>), [optimize](<https://devfeed.tech/tags/optimize.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [partitioning](<https://devfeed.tech/tags/partitioning.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [scaling](<https://devfeed.tech/tags/scaling.md>), [supercomputers](<https://devfeed.tech/tags/supercomputers.md>)

### AI overview

The work presents a distributed multi-GPU graph partitioner based on the single-GPU Jet partitioner, aiming to improve scalability while preserving solution quality. It uses CUDA-aware MPI, one-sided communication, and a bulk-synchronous parallel structure. The authors report solution quality comparable to Jet and weak scaling up to 256 GPUs, allowing it to partition instances larger than a single-GPU partitioner can handle.

### Source excerpt

arXiv:2611.24264v1 Announce Type: new Abstract: Graph partitioning is a key enabler in computational science, with applications to sparse linear algebra, load-balanced parallel computation, and graph computations. In recent years, graphics processing units (GPUs) have become the preferred accelerator on the world's top supercomputers. Despite this, most existing distributed graph partitioners do not utilize GPUs. In this work, we develop a new multi-GPU distributed partitioner based on Jet, a prior single-GPU partitioner. We seek to optimize scalability without sacrificing quality. Our partitioner utilizes CUDA-aware MPI, one-sided communication routines, and an overall bulk-synchronous parallel structure. We demonstrate solution quality comparable to Jet. Our experiments show excellent weak scaling up to 256 GPUs, enabling partitioning of instances significantly larger than a single-GPU partitioner can handle.

## GLIDE: Scalable and Event-Aware Inductive Forecasting over Evolving Logistics Graphs

DevFeed: [GLIDE: Scalable and Event-Aware Inductive Forecasting over Evolving Logistics Graphs](<https://devfeed.tech/articles/glide-scalable-and-event-aware-inductive-forecasting-over-evolving-logistics-graphs-142729.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23942>)

Author: Zhi Sheng, Chuyue Wang, Jiahao Ji, Yile Liang, Xuetao Ding, Jinghua Hao, Jie Feng, Yong Li

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [delivery](<https://devfeed.tech/tags/delivery.md>), [event](<https://devfeed.tech/tags/event.md>), [event-driven](<https://devfeed.tech/tags/event-driven.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [inductive](<https://devfeed.tech/tags/inductive.md>), [logistics](<https://devfeed.tech/tags/logistics.md>)

### AI overview

GLIDE targets order-volume forecasting for scale-effect hotspots in on-demand food delivery, where logistics graphs continually change, range from roughly 100 to 10,000 nodes, and exhibit sparse, long-tailed demand. The abstract says these conditions challenge standard spatiotemporal models, while holidays, weather, and academic calendars can shift demand beyond what purely data-driven backbones anticipate. GLIDE's Graph-Guided Latent Aggregation projects changing physical subgraphs into a fixed-size, semantically stable latent space using differentiable sparse assignment. The excerpt presents the framework as designed for scalability and event-aware adaptation, but does not provide evaluation results or further method details.

### Source excerpt

arXiv:2611.23942v1 Announce Type: new Abstract: Accurate forecasting of order volume at scale-effect hotspots (SEHs), spatially coherent groups of logistics flows that serve as distributed optimization units for real-time many-to-one dispatch, is a prerequisite for efficient order pooling in on-demand food delivery (OFD) platforms such as Meituan. However, real-world logistics networks present unique challenges that render standard spatiotemporal prediction model ineffective. (i) Dynamic Topology: Unlike static road networks, logistics graphs are ephemeral, with hotspot emerging and vanishing continuously; (ii) Scale--Volume Heterogeneity: Graph sizes vary from $10^2$ to $10^4$, while zero-inflated, long-tailed volumes create computational bottlenecks and bias predictions toward inactivity; and (iii) Event-Driven Disruption: Exogenous events (holidays, weather, academic calendars) induce systematic demand shifts that purely data-driven backbones cannot anticipate. In this paper, we present GLIDE, a production-ready framework that models evolving logistics graphs in a scalable latent space while adapting to heterogeneous demand regimes and event-driven shifts. GLIDE introduces a Graph-Guided Latent Aggregation mechanism, which dynamically projects varying physical subgraphs into a fixed-size, semantically stable latent space via a differentiable sparse assignment...

## FairPGT: Partitioned Graph Transformers for Dual-Level Fair Graph Clustering

DevFeed: [FairPGT: Partitioned Graph Transformers for Dual-Level Fair Graph Clustering](<https://devfeed.tech/articles/fairpgt-partitioned-graph-transformers-for-dual-level-fair-graph-clustering-142683.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23631>)

Author: Siamak Ghodsi, Amjad Seyedi, Sahar Tahmasebi, Eirini Ntoutsi, S\\"oren Auer

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [clustering](<https://devfeed.tech/topics/clustering.md>), [Transformers](<https://devfeed.tech/topics/transformers.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [attention](<https://devfeed.tech/tags/attention.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [bias](<https://devfeed.tech/tags/bias.md>), [clustering](<https://devfeed.tech/tags/clustering.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [distillation](<https://devfeed.tech/tags/distillation.md>), [fairness](<https://devfeed.tech/tags/fairness.md>), [graph](<https://devfeed.tech/tags/graph.md>), [information](<https://devfeed.tech/tags/information.md>), [parity](<https://devfeed.tech/tags/parity.md>), [partitions](<https://devfeed.tech/tags/partitions.md>), [ratio](<https://devfeed.tech/tags/ratio.md>), [routing](<https://devfeed.tech/tags/routing.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [transformers](<https://devfeed.tech/tags/transformers.md>)

### AI overview

FairPGT combines partitioned graph attention with two fairness controls for graph clustering. It limits attention to METIS partitions linked by routing tokens, reducing quadratic attention cost by the number of partitions. A mutual-information term targets demographic parity in cluster assignments, while nCGAR-based self-distillation guides attention toward group neutrality; the authors prove a neutrality correction bound and a condition against overshooting. Across seven benchmarks, it reports the best clustering performance and statistical parity on the three largest graphs, with a statistical-parity gap about ten times lower than DMoN, and is Pareto-dominated on one dataset. In controlled stochastic block models with aligned communities and protected groups, it reaches 97% of a derived NMI upper bound for exactly fair clustering. The abstract also reports that either fairness control alone can leave a different failure: assignment fairness can make attention group-heterophilous, while attention fairness alone does not guarantee fair clusters.

### Source excerpt

arXiv:2611.23631v1 Announce Type: cross Abstract: Clustering on attributed graphs can produce demographically biased clusters, particularly when communities align with protected groups. Graph transformers can amplify this bias through same-group attention, while global attention limits scalability. We propose FairPGT, a partitioned graph transformer for fair graph clustering. FairPGT restricts attention to METIS partitions and connects them through routing tokens, which divides the quadratic cost of attention by the number of partitions. A graph-aware self-training target guides clustering, with a provably bounded deviation from the standard DEC target. Fairness is imposed at two levels: a mutual-information term promotes demographic parity in cluster assignments, while a normalized cross-group attention ratio (nCGAR) guides attention toward group neutrality through self-distillation. We prove that the nCGAR correction moves attention toward neutrality and establish a condition preventing overshooting. Across seven benchmarks, FairPGT achieves the best clustering performance and statistical parity on the three largest graphs (Credit, Pokec-n, Pokec-z), with a statistical-parity gap approximately ten times lower than DMoN, and is Pareto-dominated on only one dataset. On controlled stochastic block models, FairPGT attains 97% of the derived upper bound on NMI for exactly fair clustering when communities and protected groups coincide. These graphs further show that, under strong alignment, assignment-level fairness alone makes the attention group-heterophilous, whereas attention-level fairness alone improves neutrality without ensuring fair clusters. Code for reproducibility is available at https://github.com/SiamakGhodsi/FairPGT.

## Improving LLM Final Representations with Inter-Layer Geometry

DevFeed: [Improving LLM Final Representations with Inter-Layer Geometry](<https://devfeed.tech/articles/improving-llm-final-representations-with-inter-layer-geometry-143245.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2603.22665>)

Author: Tom Ulanovski, Eyal Blyachman, Maya Bechler-Speicher

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [few-shot](<https://devfeed.tech/topics/few-shot.md>), [Fine-tuning](<https://devfeed.tech/topics/fine-tuning.md>), [lora](<https://devfeed.tech/topics/lora.md>)

Tags: [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [diameter](<https://devfeed.tech/tags/diameter.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [encoder](<https://devfeed.tech/tags/encoder.md>), [explainability](<https://devfeed.tech/tags/explainability.md>), [few-shot](<https://devfeed.tech/tags/few-shot.md>), [fully-connected](<https://devfeed.tech/tags/fully-connected.md>), [gnn](<https://devfeed.tech/tags/gnn.md>), [intermediate](<https://devfeed.tech/tags/intermediate.md>), [layers](<https://devfeed.tech/tags/layers.md>), [llm](<https://devfeed.tech/tags/llm.md>), [lora](<https://devfeed.tech/tags/lora.md>), [multiple](<https://devfeed.tech/tags/multiple.md>), [overfitting](<https://devfeed.tech/tags/overfitting.md>), [performance](<https://devfeed.tech/tags/performance.md>), [topology](<https://devfeed.tech/tags/topology.md>)

### AI overview

Using intermediate LLM layers can improve downstream prediction over relying only on the final-layer representation. The work first applies a lightweight GNN over a fully connected graph of layers, then introduces Cayley-Encoder, which uses a sparse, regular, low-diameter Cayley graph over SL(2, Zn) to make aggregation more efficient and constrain its structure. Across 13 tasks and 13 LLMs, Cayley-Encoder reportedly outperforms baselines, with accuracy gains up to 48 percentage points and added parameters capped at 0.1% of LLM size. The abstract also reports effectiveness in few-shot settings and better performance than LoRA fine-tuning while keeping the LLM frozen. Its explainability analysis finds meaningful contributions from multiple layers. These are abstract-level claims; the supplied material gives no task-specific results or experimental details.

### Source excerpt

arXiv:2603.22665v4 Announce Type: replace-cross Abstract: The standard in LLM-based prediction is to use the final-layer representation as the input to a downstream predictor. However, intermediate layers may encode complementary task-relevant signals. Existing approaches therefore either search for the best layer for each task or apply expensive attention-based mechanisms to learn inter-layer aggregation. In this work, we first show that such complexity is unnecessary: a lightweight Graph Neural Network over a fully connected graph of LLM layers is more efficient and achieves significantly stronger predictive performance than existing approaches. We then introduce the Cayley-Encoder, which further improves both efficiency and predictive performance by replacing the fully connected graph with a Cayley graph over SL(2, Zn). These Cayley graphs provide a mathematically grounded topology that is sparse, regular by construction, and has low diameter. This enables effective communication across layers while constraining the aggregation structure and reducing the risk of GNN overfitting. In an evaluation of Cayley-Encoder across 13 tasks and 13 LLMs, Cayley-Encoder consistently outperforms baselines, achieving improvements of up to 48 percentage points in accuracy, while introducing at most 0.1% additional parameters relative to the LLM size. We further show that Cayley-Encoder is effective in few-shot regimes. Finally, we show that Cayley-Encoder outperforms LoRA fine-tuning while operating on the frozen LLM. We conclude with an explainability analysis showing that multiple layers contribute meaningfully to the final prediction, supporting our hypothesis.

## Evaluating invariance training with certified answer-changing controls in neural graph algorithms

DevFeed: [Evaluating invariance training with certified answer-changing controls in neural graph algorithms](<https://devfeed.tech/articles/evaluating-invariance-training-with-certified-answer-changing-controls-in-neural-graph-algorithms-142721.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23907>)

Author: Zhe Dong, Fang Qin

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [algorithms](<https://devfeed.tech/tags/algorithms.md>), [controls](<https://devfeed.tech/tags/controls.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [graph](<https://devfeed.tech/tags/graph.md>), [graph-algorithms](<https://devfeed.tech/tags/graph-algorithms.md>), [invariance](<https://devfeed.tech/tags/invariance.md>), [neural](<https://devfeed.tech/tags/neural.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

On synthetic minimum-spanning-tree tasks, invariance training reduced a graph transformer's errors on answer-preserving bridge attachments but also weakened its response to answer-changing two-edge attachments. The study certifies both kinds of edits: a one-bridge attachment preserves original labels, while two oracle-selected edges can change a label; lowering an internal edge weight provides another changing control. On 16-vertex graphs, three invariance objectives cut bridge-attachment violations from 2.91% to 0.52-2.02% of pairs with essentially unchanged accuracy, while two-edge response fell from 97.6% to 66.0-79.0%; response to internal weight edits stayed near 99.5%. At 64 vertices, accuracy also declined. A message-passing network showed some loss of attachment response at low penalty weights but little invariance gain. Applying the conventionally trained transformer separately to each biconnected block eliminated observed violations while retaining accuracy and both control responses, suggesting the trade-off was avoidable in this setup. The excerpt recommends testing invariance with certified answer-changing edits that attach material in the same way as training edits.

### Source excerpt

arXiv:2611.23907v1 Announce Type: new Abstract: Invariance training aims to make neural algorithmic reasoners robust to answer-preserving input changes and is judged by accuracy and by how rarely such edits change predictions. We ask whether this evaluation detects lost sensitivity to answer-changing edits on minimum spanning trees of synthetic graphs, where both directions can be certified. Attaching a subgraph through one bridge provably preserves every original label. Attaching it through two edges, oracle-selected to change a label, and lowering one weight inside the graph serve as answer-changing controls; a response is a changed prediction on an edge whose label changes. A reproduced graph transformer (GDT) within its published accuracy range changes a provably fixed label on 53.3-65.9% of the 64-vertex graphs it solves exactly after a 64-vertex bridge attachment; removing the connecting edge changes little. With 16-vertex graphs and attachments, three invariance objectives lower GDT violations from 2.91% to 0.52-2.02% of pairs at essentially unchanged accuracy. Training also lowers the response to two-edge attachments from 97.6% to 66.0-79.0%, while the response to in-base weight edits stays near 99.5%. At 64 vertices accuracy also falls, but on 16-vertex bases only the attachment control reveals the loss. A message-passing network loses some attachment response at small penalty weights but gains little invariance. Running the conventionally trained GDT on each biconnected block separately removes all observed violations and keeps accuracy and both responses, so the task does not force the trade-off. Sensitivity checks for trained invariance should include certified answer-changing edits that attach material as the training edits do.

## A Metric for Semantic Graph Matching of No-Code Workflows

DevFeed: [A Metric for Semantic Graph Matching of No-Code Workflows](<https://devfeed.tech/articles/a-metric-for-semantic-graph-matching-of-no-code-workflows-142122.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.23233>)

Author: Bart{\\l}omiej Matejczyk, Joao Tiago Aparicio

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [n8n](<https://devfeed.tech/topics/n8n.md>)

Tags: [cs-se](<https://devfeed.tech/tags/cs-se.md>), [graph](<https://devfeed.tech/tags/graph.md>), [matching](<https://devfeed.tech/tags/matching.md>), [metric](<https://devfeed.tech/tags/metric.md>), [no-code](<https://devfeed.tech/tags/no-code.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

The proposed distance compares workflows as attributed directed graphs, using partial one-to-one node matches to score operation substitutions and labelled-edge disagreements, with a positive cost for unmatched nodes. Under stated fixed representations and metric-compatible primitive distances, it is a metric up to attributed isomorphism; a padded Hungarian method gives a feasible upper bound rather than the exact objective. Tests on public n8n exports found near-zero distance for surface-preserving changes and higher distance for major type or subgraph changes; edit counts were positively associated with distance, and all tested triangle orientations had non-negative slack. The measure compares workflow artifacts and does not establish behavioral equivalence under external state or runtime data.

### Source excerpt

arXiv:2611.23233v1 Announce Type: new Abstract: No-code and agentic systems increasingly produce executable workflows, but workflows cannot be compared adequately as strings because typed operations, parameters, and directed connections jointly determine the artifact. We define a semantic graph-matching distance for finite attributed workflow graphs. A partial one-to-one node matching determines node-substitution and directed labelled-edge disagreement, while unmatched nodes incur a positive null cost. Under a fixed type hierarchy, encoder, field universe, and metric-compatible primitive distances, the objective is a metric on workflow graphs up to attributed isomorphism. A padded Hungarian procedure provides a scalable feasible upper bound distinct from the exact objective. We evaluate the metric on 4,705 public n8n exports, representing 2,282 distinct normalized-graph hashes, 73,057 nodes, and 469 observed node types. Across 32,093 branch-and-bound-complete evaluations, the mean raw distance is near zero under surface-preserving transformations ($1.21 \times 10^{-5}$) and rises to $0.584$ under major type or subgraph changes. On a complete 796-hash panel, accumulated edit count has Spearman $\rho = 0.520$ with distance, while the median within-workflow correlation is $0.975$. One- and three-perturbation systems have marginal means of $0.307$ and $0.763$, and all 117,312 evaluated triangle orientations have non-negative slack. Weight ablations preserve the positive edit-count association across tested configurations. These tests support numerical consistency, near-invariance to surface-preserving transformations, graded sensitivity to semantic and structural change, and robust discrimination across controlled modifications. The metric measures artifacts and does not by itself assert behavioural equivalence under external state or runtime data.

## Spatiotemporal and Multi-Robot Motion Planning on Graphs of Space-Time Convex Sets

DevFeed: [Spatiotemporal and Multi-Robot Motion Planning on Graphs of Space-Time Convex Sets](<https://devfeed.tech/articles/spatiotemporal-and-multi-robot-motion-planning-on-graphs-of-space-time-convex-sets-139786.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2607.00444>)

Author: Jingtao Tang, Zining Mao, Lufan Yang, Hang Ma

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Robot Navigation](<https://devfeed.tech/topics/robot-navigation.md>), [Combinatorial optimization](<https://devfeed.tech/topics/combinatorial-optimization.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [continuous](<https://devfeed.tech/tags/continuous.md>), [convex](<https://devfeed.tech/tags/convex.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [motion](<https://devfeed.tech/tags/motion.md>), [motion-planning](<https://devfeed.tech/tags/motion-planning.md>), [planning](<https://devfeed.tech/tags/planning.md>), [pruning](<https://devfeed.tech/tags/pruning.md>), [quality](<https://devfeed.tech/tags/quality.md>), [robot](<https://devfeed.tech/tags/robot.md>), [robots](<https://devfeed.tech/tags/robots.md>), [route](<https://devfeed.tech/tags/route.md>), [scalability](<https://devfeed.tech/tags/scalability.md>), [search](<https://devfeed.tech/tags/search.md>), [simulations](<https://devfeed.tech/tags/simulations.md>), [space](<https://devfeed.tech/tags/space.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [time](<https://devfeed.tech/tags/time.md>), [timing](<https://devfeed.tech/tags/timing.md>), [velocity](<https://devfeed.tech/tags/velocity.md>)

### AI overview

The framework plans continuous robot motion in space and time by representing collision-free regions as convex sets, jointly optimizing the route and piecewise-linear trajectory timing under velocity limits. Exact Convex Decomposition removes swept space occupied by moving obstacles and previously planned robots while preserving a convex decomposition for later queries. A best-first solver uses temporal reachability for dominance pruning and admissible heuristics; heuristic bounds can be reused after reservation updates. The authors combine these methods with prioritized planning and dynamically adjusted coordination windows to limit graph growth. Experiments report improved scalability and competitive solution quality against optimization-based and sampling-based planners; simulations coordinate up to 100 robots within a minute, with real-robot demonstrations also reported.

### Source excerpt

arXiv:2607.00444v2 Announce Type: replace-cross Abstract: Decoupled multi-robot motion planning avoids the high-dimensional joint configuration space but requires repeated single-robot planning among moving obstacles, where route choice and timing are coupled. We develop a framework based on graphs of space-time convex sets (ST-GCSs) for continuous spatiotemporal planning and multi-robot coordination. An ST-GCS represents collision-free space-time by convex sets, allowing the sequence of regions and the geometry and timing of piecewise-linear trajectories to be optimized jointly under velocity limits. To incorporate moving obstacles and previously planned robots, we develop Exact Convex Decomposition (ECD), which removes their swept occupancies while maintaining a convex decomposition for subsequent planning queries. Building on path-based GCS search, we develop a best-first solver tailored to the ST-GCS motion model, exploiting temporal reachability for dominance pruning and using admissible heuristics with separate or joint search-key evaluation. Precomputed heuristic bounds can also be reused across reservation-updated graphs. We integrate ST-GCS planning and ECD into prioritized multi-robot planning and introduce windowed coordination with dynamic window adjustment to limit reservation-induced graph growth. Experiments demonstrate improved scalability and competitive solution quality against optimization-based and sampling-based planners on challenging spatiotemporal and multi-robot problems. Large-scale simulations coordinate up to $100$ robots within a minute, while real-robot demonstrations showcase the practical applicability of the framework. Project homepage: https://sites.google.com/view/stgcs.

## When Does Discrete Ricci Curvature Help? An Edge-Level Analysis of Curvature-Enhanced GCNs for PPI Prediction

DevFeed: [When Does Discrete Ricci Curvature Help? An Edge-Level Analysis of Curvature-Enhanced GCNs for PPI Prediction](<https://devfeed.tech/articles/when-does-discrete-ricci-curvature-help-an-edge-level-analysis-of-curvature-enhanced-gcns-for-ppi-prediction-141685.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21849>)

Author: Clara Joswig, Elham Gholamzadeh, Ghazal Mostofi, Nico Scherf

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [clustering](<https://devfeed.tech/topics/clustering.md>)

Tags: [ablation](<https://devfeed.tech/tags/ablation.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [clustering](<https://devfeed.tech/tags/clustering.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [communities](<https://devfeed.tech/tags/communities.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [edge](<https://devfeed.tech/tags/edge.md>), [graph](<https://devfeed.tech/tags/graph.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [protein](<https://devfeed.tech/tags/protein.md>)

### AI overview

Across four protein-protein interaction datasets, curvature-enhanced graph convolutional networks (CGCN) improve aggregate AUC, AUPR, and F1 over a standard GCN, but the gains are concentrated on low-degree edges. Performance on low-curvature bottleneck edges declines in three of four datasets, contrary to the proposed over-squashing explanation. The authors' analyses support an alternative account: curvature weighting sharpens cluster separation, helping within-cluster predictions while hurting predictions between clusters. Their guidance on choosing CGCN or a standard GCN depends on whether intra- or inter-cluster predictions matter more.

### Source excerpt

arXiv:2611.21849v1 Announce Type: new Abstract: Curvature-enhanced graph convolutional networks (CGCN) incorporate Ollivier-Ricci curvature (ORC), grounded in optimal transport theory, into message passing, motivated by the claim that curvature-weighted edges alleviate "over-squashing" at structural bottlenecks. But does CGCN's improvement actually come from mitigating over-squashing? We investigate this on four real-world protein-protein interaction (PPI) datasets, comparing CGCN against a standard GCN at the level of individual edges rather than aggregate metrics alone. While CGCN improves aggregate AUC, AUPR, and F1 across all four datasets, this gain is concentrated almost entirely on low-degree edges, while performance on low-curvature (bottleneck) edges systematically degrades in three of four datasets, the opposite of what the over-squashing motivation predicts. Examining the learned edge-weight function directly, together with clustering-validity metrics, edge-distance ratios, and a layer-depth ablation across four depths, we find evidence for an alternative account: curvature-weighting sharpens the separation between clusters, which improves intra-cluster predictions at the cost of inter-cluster (bottleneck) ones. We test this account against topologically defined communities rather than the classification labels used elsewhere. Additionally, we offer practical guidance on when to prefer CGCN over standard GCN, depending on whether intra- or inter-cluster predictions matter most.

## HOPSE: Scalable Higher-Order Positional and Structural Encoder for Combinatorial Representations

DevFeed: [HOPSE: Scalable Higher-Order Positional and Structural Encoder for Combinatorial Representations](<https://devfeed.tech/articles/hopse-scalable-higher-order-positional-and-structural-encoder-for-combinatorial-representations-141520.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2505.15405>)

Author: Guillermo Bern\\'ardez, Marco Montagna, Louis Van Langendonck, Martin Carrasco, Amirreza Akbari, Louisa Cornelis, Mathilde Papillon, Pere Barlet-Ros, Nina Miolane, Lev Telyatnikov

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [model architecture](<https://devfeed.tech/topics/model-architecture.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [cellular](<https://devfeed.tech/tags/cellular.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [encoder](<https://devfeed.tech/tags/encoder.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [graph-neural-networks](<https://devfeed.tech/tags/graph-neural-networks.md>), [message](<https://devfeed.tech/tags/message.md>), [order](<https://devfeed.tech/tags/order.md>), [performance](<https://devfeed.tech/tags/performance.md>), [scalability](<https://devfeed.tech/tags/scalability.md>)

### AI overview

HOPSE addresses the cost of higher-order message passing in topological deep learning with a framework that has no learnable message-passing layers. It uses Hasse graph decompositions to produce positional and structural encodings over higher-order combinatorial domains. The abstract states that training and inference scale linearly with representation size while retaining the expressive power and permutation equivariance of HOMP approaches. On molecular and topological benchmarks, HOPSE reportedly matches or exceeds state-of-the-art performance and is consistently faster than HOMP-based models.

### Source excerpt

arXiv:2505.15405v4 Announce Type: replace Abstract: While Graph Neural Networks (GNNs) have proven highly effective at modeling relational data, pairwise connections cannot fully capture multi-way relationships naturally present in complex real-world systems. In response to this, Topological Deep Learning (TDL) leverages more general combinatorial representations--such as simplicial or cellular complexes--to accommodate higher-order interactions. Existing TDL methods often extend GNNs through Higher-Order Message Passing (HOMP), but face critical scalability challenges due to the steep complexity overhead of propagating messages through combinatorial structures. To overcome this limitation, we propose HOPSE (Higher-Order Positional and Structural Encoder), a framework free of learnable message passing layers that uses Hasse graph decompositions to derive efficient and expressive encodings over arbitrary higher-order domains. Notably, HOPSE training and inference scale linearly with the size of combinatorial representations while preserving the expressive power and permutation equivariance of the HOMP approaches. Experiments on molecular and topological benchmarks show that it matches or surpasses state-of-the-art performance while consistently achieving speedups over HOMP-based models, opening a new path for scalable TDL. The code is available at https://github.com/geometric-intelligence/topobench.

## SiGMA: Learning under Graph-Level Heterogeneity with Gated Message-Passing and Attention

DevFeed: [SiGMA: Learning under Graph-Level Heterogeneity with Gated Message-Passing and Attention](<https://devfeed.tech/articles/sigma-learning-under-graph-level-heterogeneity-with-gated-message-passing-and-attention-141519.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2503.00547>)

Author: Raphael Pellegrin, Lukas Fesser, Melanie Weber

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Model Development](<https://devfeed.tech/topics/model-development.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [architecture](<https://devfeed.tech/tags/architecture.md>), [attention](<https://devfeed.tech/tags/attention.md>), [classification](<https://devfeed.tech/tags/classification.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [graph](<https://devfeed.tech/tags/graph.md>), [graph-neural-networks](<https://devfeed.tech/tags/graph-neural-networks.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [identity](<https://devfeed.tech/tags/identity.md>), [language](<https://devfeed.tech/tags/language.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [layer](<https://devfeed.tech/tags/layer.md>), [learning](<https://devfeed.tech/tags/learning.md>), [local](<https://devfeed.tech/tags/local.md>), [message](<https://devfeed.tech/tags/message.md>), [performance](<https://devfeed.tech/tags/performance.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [regression](<https://devfeed.tech/tags/regression.md>), [sigma](<https://devfeed.tech/tags/sigma.md>), [transformer](<https://devfeed.tech/tags/transformer.md>), [transformers](<https://devfeed.tech/tags/transformers.md>)

### AI overview

SiGMA addresses differences in prediction difficulty across graphs by combining message-passing and attention heads in each layer, with learned node-wise sigmoid gates regulating which mechanisms are active. The authors study graph-level heterogeneity in classification and regression and report that SiGMA matches or improves on competitive graph transformer and hybrid baselines across graph-level benchmarks. They say its design can emulate expressive message-passing backbones, provide near-identity paths that avoid forced over-smoothing, and enable non-local communication intended to mitigate over-squashing.

### Source excerpt

arXiv:2503.00547v2 Announce Type: replace Abstract: In graph-level learning, graph neural networks are typically applied with a fixed computational template: the same depth, local message-passing operator, and global attention mechanism are used for every input graph. However, different graphs within the same dataset can vary substantially in prediction difficulty and may require different mixtures of local and non-local computation. We study this phenomenon as graph-level heterogeneity in both classification and regression tasks. To address it, we propose SiGMA (Sigmoid Gated Message-Passing and Attention), a gated hybrid graph neural network architecture that combines message-passing and attention heads within each layer. Inspired by the growing use of gated attention and gated computation in large language models and vision transformers, each head in SiGMA is modulated by a learned node-wise sigmoid gate, allowing the model to adaptively regulate which local or global mechanisms are active for different nodes, layers, and input graphs. The resulting architecture can emulate expressive message-passing backbones, admits near-identity computation paths that are not forced to over-smooth, and enables direct non-local communication beyond message-passing neighborhoods to mitigate over-squashing. Empirically, SiGMA achieves strong performance across graph-level benchmarks, consistently matching or improving over competitive graph transformer and hybrid baselines. Our results suggest that adapting computation to graph-level heterogeneity is a promising direction for building more flexible and effective graph learning architectures.

## Zero-Dimensional Attention Topology for Linguistic Acceptability

DevFeed: [Zero-Dimensional Attention Topology for Linguistic Acceptability](<https://devfeed.tech/articles/zero-dimensional-attention-topology-for-linguistic-acceptability-141765.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22312>)

Author: Saeed AlMarri (Khalifa University)

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [bert](<https://devfeed.tech/topics/bert.md>), [Transformer](<https://devfeed.tech/topics/transformer.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [Feature Engineering](<https://devfeed.tech/topics/feature-engineering.md>)

Tags: [attention](<https://devfeed.tech/tags/attention.md>), [bert](<https://devfeed.tech/tags/bert.md>), [classification](<https://devfeed.tech/tags/classification.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [persistent-homology](<https://devfeed.tech/tags/persistent-homology.md>), [sampling](<https://devfeed.tech/tags/sampling.md>), [topology](<https://devfeed.tech/tags/topology.md>)

### AI overview

The fixed H₀ attention representation did not show a predictive advantage over simpler attention statistics for linguistic acceptability. It summarizes minimum-spanning-tree merge values from 144 symmetrized BERT attention graphs into 1,440 features. In the retrospective evaluation, its out-of-domain Matthews correlation was 0.438, versus 0.508 for attention statistics and 0.545 for BERT. The analysis found that the representation retains token-count information while discarding some directional attention differences. The authors also report that uncertainty estimates from resampling BERT predictions depend on the sampling unit. Attention-feature scores are retained aggregate measurements, while BERT predictions were rescored and length controls refitted from accompanying records.

### Source excerpt

arXiv:2611.22312v1 Announce Type: cross Abstract: Topological summaries of Transformer attention provide sentence representations for linguistic acceptability classification. We examine a zero-dimensional persistent-homology ($H_0$) representation that summarizes minimum-spanning-tree merge values from 144 symmetrized BERT attention graphs into 1,440 features. The retrospective evaluation uses 8,517 downstream training sentences and 521 in-domain and 497 out-of-domain CoLA validation sentences after data-integrity filtering. The recorded out-of-domain Matthews correlation coefficients are 0.438 for the topological representation and 0.508 for attention statistics, compared with 0.545 for BERT. Attention-feature results are retained aggregate measurements; BERT predictions are rescored and length controls are refitted from the accompanying records. Analytical results show that the representation retains token-count information and discards a class of directional attention differences. Source resampling of BERT predictions also demonstrates the sensitivity of uncertainty estimates to the sampling unit. The reported results do not demonstrate a predictive advantage for this fixed topological representation over simpler attention summaries. They identify representation properties and evaluation controls relevant to subsequent studies of attention topology.

## Markov-Equivalent Conditional Independence Graphs on Path Spaces

DevFeed: [Markov-Equivalent Conditional Independence Graphs on Path Spaces](<https://devfeed.tech/articles/markov-equivalent-conditional-independence-graphs-on-path-spaces-139104.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20877>)

Author: Simon Kempf, Niki Kilbertus, Georg Manten

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [abstract](<https://devfeed.tech/tags/abstract.md>), [added](<https://devfeed.tech/tags/added.md>), [addition](<https://devfeed.tech/tags/addition.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [allows](<https://devfeed.tech/tags/allows.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [but](<https://devfeed.tech/tags/but.md>), [can](<https://devfeed.tech/tags/can.md>), [canonical](<https://devfeed.tech/tags/canonical.md>), [central](<https://devfeed.tech/tags/central.md>), [characterization](<https://devfeed.tech/tags/characterization.md>), [class](<https://devfeed.tech/tags/class.md>), [classes](<https://devfeed.tech/tags/classes.md>), [components](<https://devfeed.tech/tags/components.md>), [conditional](<https://devfeed.tech/tags/conditional.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [markov](<https://devfeed.tech/tags/markov.md>), [spaces](<https://devfeed.tech/tags/spaces.md>)

### AI overview

The paper introduces W-separation, an asymmetric criterion for directed mixed graphs that conditions separately on past and future components. For a broad class of stochastic differential equations with latent confounding, it satisfies a global Markov property, connecting graph relations to conditional independences among coordinate processes that are testable from data. The paper also characterizes W-Markov equivalence classes: each has a unique greatest graph as a canonical representative and potential learning target. It gives necessary and sufficient conditions for adding directed or bidirected edges without changing the implied independence model.

### Source excerpt

arXiv:2611.20877v1 Announce Type: new Abstract: Causal discovery for continuous-time stochastic processes with latent confounding typically faces a trade-off between (i) approaches that are highly informative about the causal structure but are difficult to test statistically, and (ii) approaches that admit practical tests from data but yield more coarse-grained conclusions, potentially not leveraging the direction of time. We introduce W-separation, an asymmetric graphical criterion for directed mixed graphs defined as a quaternary relation that allows distinct conditioning on past and future components. This added flexibility yields a highly informative graphical independence model while preserving testability. W-separation satisfies a global Markov property for a broad class of stochastic differential equations with latent confounding, linking it to conditional independence relations among coordinate processes that are testable from data. A central result of this work is the full characterization of W-Markov equivalence classes: we show that every equivalence class admits a unique greatest element that serves as a canonical representative and a principled target for learning algorithms. In addition, we provide necessary and sufficient conditions under which directed or bidirected edges can be added without altering the induced independence model.

## Minimum cores in lattice graphs: zonotope structure and quartic growth

DevFeed: [Minimum cores in lattice graphs: zonotope structure and quartic growth](<https://devfeed.tech/articles/minimum-cores-in-lattice-graphs-zonotope-structure-and-quartic-growth-137310.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18633>)

Author: Navid Mehrdad

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>)

Tags: [certificates](<https://devfeed.tech/tags/certificates.md>), [condition](<https://devfeed.tech/tags/condition.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [convex](<https://devfeed.tech/tags/convex.md>), [core](<https://devfeed.tech/tags/core.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-si](<https://devfeed.tech/tags/cs-si.md>), [graph](<https://devfeed.tech/tags/graph.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [growth](<https://devfeed.tech/tags/growth.md>), [lattice](<https://devfeed.tech/tags/lattice.md>), [math-co](<https://devfeed.tech/tags/math-co.md>), [neighbours](<https://devfeed.tech/tags/neighbours.md>), [packing](<https://devfeed.tech/tags/packing.md>), [reduction](<https://devfeed.tech/tags/reduction.md>), [structure](<https://devfeed.tech/tags/structure.md>), [translation](<https://devfeed.tech/tags/translation.md>), [vertex](<https://devfeed.tech/tags/vertex.md>)

### AI overview

For the graph on integer lattice points joined when their sup-norm distance is at most n, every finite core has a convex hull constrained by an explicit lattice zonotope. In the plane, the zonotope's width grows as 6ζ(3)n³/π²; its lattice points form a core, and a shell construction plus packing bounds establish minimum core size Θ(n⁴) and coordinate widths Θ(n³). The paper determines minima 12 and 96 for n=1 and n=2, with uniqueness up to translation, and gives certificates for further minima restricted to the prescribed zonotope hull, distinguishing these from global optimality. It also proves NP-completeness on general finite graphs by reduction from Clique.

### Source excerpt

arXiv:2611.18633v1 Announce Type: cross Abstract: A core of a graph is a nonempty vertex set in which every member has at least as many neighbours inside the set as outside. We study the minimum size $c_{m,n}$ of a finite core in the graph on $\Z^m$ joining points at sup-norm distance at most $n$. The local majority condition imposes a global geometric constraint: the convex hull of every core has an explicit lattice zonotope $H_n^{(m)}$ as a Minkowski summand. In the plane, this zonotope has width asymptotic to $6\zeta(3)n^3/\pi^2$. We prove that all its lattice points form a core and construct a shell with $O(n^4)$ points. Together with a packing argument, this gives $c_{2,n}=\Theta(n^4)$ and coordinate widths $\Theta(n^3)$ for every minimum core. We determine $c_{2,1}=12$ and $c_{2,2}=96$ by analytic arguments and prove uniqueness up to translation in both cases. Explicit forced-set and deficit certificates determine further minima within the prescribed zonotope hull. We distinguish these restricted results from global optimality and state the remaining geometric conjectures. We also give a direct reduction from \textsc{Clique} showing that the core problem is NP-complete on general finite graphs, which settles a conjecture of Hassanpour in the terms in which it was posed.

## Canonical activation patterns and ReLU pattern varieties

DevFeed: [Canonical activation patterns and ReLU pattern varieties](<https://devfeed.tech/articles/canonical-activation-patterns-and-relu-pattern-varieties-139030.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20346>)

Author: Bella Finkel, Yaoying Fu, Chenxi Wu

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>)

Tags: [activation](<https://devfeed.tech/tags/activation.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [bounded](<https://devfeed.tech/tags/bounded.md>), [canonical](<https://devfeed.tech/tags/canonical.md>), [combinatorics](<https://devfeed.tech/tags/combinatorics.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dimension](<https://devfeed.tech/tags/dimension.md>), [edge](<https://devfeed.tech/tags/edge.md>), [equality](<https://devfeed.tech/tags/equality.md>), [full](<https://devfeed.tech/tags/full.md>), [functional](<https://devfeed.tech/tags/functional.md>), [graph](<https://devfeed.tech/tags/graph.md>), [hierarchy](<https://devfeed.tech/tags/hierarchy.md>), [map](<https://devfeed.tech/tags/map.md>), [maps](<https://devfeed.tech/tags/maps.md>), [math-ag](<https://devfeed.tech/tags/math-ag.md>), [math-co](<https://devfeed.tech/tags/math-co.md>), [network](<https://devfeed.tech/tags/network.md>), [neuron](<https://devfeed.tech/tags/neuron.md>), [pattern](<https://devfeed.tech/tags/pattern.md>), [patterns](<https://devfeed.tech/tags/patterns.md>), [relu](<https://devfeed.tech/tags/relu.md>)

### AI overview

The paper connects the combinatorics of canonical activation patterns in unbiased ReLU networks to the geometry of their pattern varieties. For shallow networks, it characterizes pattern stability and combinatorial stability; when the wall graph is connected and includes an edge for every neuron label, it explicitly determines the variety's defining ideal, resolving a problem of Alexandr and Montúfar. It also lays out progressively weaker equivalences among labeled combinatorial structure, canonical pattern sets, and their varieties. Functional dimension is bounded by pattern variety dimension at every depth; for nondegenerate shallow networks the authors give an exact formula, and for every shallow architecture the two dimensions agree on an open dense set of full Lebesgue measure.

### Source excerpt

arXiv:2611.20346v1 Announce Type: cross Abstract: A ReLU pattern variety records the polynomial relations among the coefficients of the affine maps associated with a prescribed set of activation patterns of a ReLU network. We study how the combinatorics of the canonical set of activation patterns of an unbiased ReLU network is reflected in the geometry of its pattern variety. In particular, we introduce the notion of pattern stability and characterize pattern stability and combinatorial stability for shallow networks. When the associated wall graph is connected and contains an edge of every neuron label, we resolve a problem of Alexandr and Mont\'ufar by determining the defining ideal of the associated pattern variety explicitly. We establish a hierarchy of progressively weaker equivalence relations relating labeled combinatorial equivalence, equality and permutation equivalence of canonical pattern sets, and equivalences of the associated varieties. Finally, we relate functional dimension to the differential of the pattern map and show that it is bounded above by the dimension of the canonical pattern variety at every depth. For nondegenerate shallow networks, we give an exact formula for the functional dimension; for every shallow architecture, it equals the dimension of the canonical pattern variety on an open dense set of full Lebesgue measure.

## When Facts Change, Which Decisions Break? Dependency-Aware Context Graphs for Agents

DevFeed: [When Facts Change, Which Decisions Break? Dependency-Aware Context Graphs for Agents](<https://devfeed.tech/articles/when-facts-change-which-decisions-break-dependency-aware-context-graphs-for-agents-137503.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19859>)

Author: Paul Deepakraj Retinraj

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [context window](<https://devfeed.tech/topics/context-window.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>)

Tags: [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [agents](<https://devfeed.tech/tags/agents.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [context-graphs](<https://devfeed.tech/tags/context-graphs.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [dependencies](<https://devfeed.tech/tags/dependencies.md>), [dependency](<https://devfeed.tech/tags/dependency.md>), [directory](<https://devfeed.tech/tags/directory.md>), [domains](<https://devfeed.tech/tags/domains.md>), [effects](<https://devfeed.tech/tags/effects.md>), [enhancement](<https://devfeed.tech/tags/enhancement.md>), [expression](<https://devfeed.tech/tags/expression.md>), [field](<https://devfeed.tech/tags/field.md>), [gpt](<https://devfeed.tech/tags/gpt.md>), [gpt-6-1-sol](<https://devfeed.tech/tags/gpt-6-1-sol.md>), [gpt-6-luna](<https://devfeed.tech/tags/gpt-6-luna.md>), [graph](<https://devfeed.tech/tags/graph.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [impact](<https://devfeed.tech/tags/impact.md>), [imperative](<https://devfeed.tech/tags/imperative.md>)

### AI overview

In controlled executable workflows, field-level dependency graphs extracted from agent read traces reduce the decisions reviewed and estimated total cost, but can miss alternative supports or repair context and thereby lower complete-plan correctness. Across 120 workflows, the study compares these graphs with full recomputation, rolling summaries, lexical retrieval, and verified dependency upper bounds; it also tests edge and context ablations. In a fresh set of 60 workflows, adding a record ID/field directory during graph extraction--without exposing unread values or reference edges--matched full-context correctness for gpt-6.1-sol (100%) with an estimated 4.0% cost reduction, while gpt-6-luna reached 84.8% versus 99.6%. The authors distinguish potential dependency impact from actual decision invalidity and caution that results on explicit procedural workflows do not establish generalization to real-world agent memory.

### Source excerpt

arXiv:2611.19859v1 Announce Type: new Abstract: When an observed fact changes, an agent must identify and repair the decisions that depend on it while preserving unaffected outputs. We investigate the reliability-cost trade-off of field-level dependency graphs extracted from actual agent read traces. A controlled benchmark spans three executable workflow domains and four successive updates, with independent imperative and expression-based references. Across 120 workflows, two models and two repeated runs, we compare full recomputation, rolling summary memory, lexical retrieval, trace-extracted dependencies and verified dependency upper bounds. Trace-derived graphs reduce reviewed decisions and estimated total cost, but omitted alternative supports and missing repair context can reduce complete-plan correctness. Paired ablations isolate the effects of edge deletion, spurious edges, transitive propagation and context scope. An exploratory enhancement supplies the existing record ID/field directory during extraction without revealing unread values or reference edges. We evaluate this enhancement on 60 fresh workflows with two repeats per model. On this fresh set, the directory-aware graph matches observed full-context complete-plan correctness for gpt-6.1-sol (100%) with an estimated 4.0% total-cost reduction, but reaches only 84.8% versus 99.6% for gpt-6-luna. The results distinguish potential dependency impact from actual decision invalidity and show why selective repair must be evaluated with both executable correctness and full memory-processing cost. The evidence concerns explicit procedural workflows; broad real-world agent-memory generalization remains untested.

## Improving fine-grained expressivity of graph neural networks

DevFeed: [Improving fine-grained expressivity of graph neural networks](<https://devfeed.tech/articles/improving-fine-grained-expressivity-of-graph-neural-networks-137526.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20020>)

Author: Yonatan Sverdlov

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>), [Representation learning](<https://devfeed.tech/topics/representation-learning.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>), [model architecture](<https://devfeed.tech/topics/model-architecture.md>)

Tags: [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [graph](<https://devfeed.tech/tags/graph.md>), [graph-neural-networks](<https://devfeed.tech/tags/graph-neural-networks.md>), [message](<https://devfeed.tech/tags/message.md>), [metric](<https://devfeed.tech/tags/metric.md>), [neural-networks](<https://devfeed.tech/tags/neural-networks.md>), [performance](<https://devfeed.tech/tags/performance.md>), [range](<https://devfeed.tech/tags/range.md>), [set](<https://devfeed.tech/tags/set.md>), [structured-data](<https://devfeed.tech/tags/structured-data.md>), [symmetry](<https://devfeed.tech/tags/symmetry.md>), [theory](<https://devfeed.tech/tags/theory.md>)

### AI overview

The thesis presents four contributions to expressive, symmetry-aware neural networks for structured data: it characterizes generic expressivity for geometric message-passing networks on sparse graphs and introduces EGenNet to achieve those guarantees; uses representation theory to characterize linear layers equivariant to multiple permutations, including wreath-product symmetries; establishes dimensional bounds and impossibility results for permutation-invariant set representations that preserve containment, and introduces MASNet; and proposes FSW-GNN, combining expressivity and bi-Lipschitz stability guarantees with reported strong performance on long-range graph-learning tasks.

### Source excerpt

arXiv:2611.20020v1 Announce Type: cross Abstract: This thesis investigates the expressive power of neural networks for structured data, focusing on graph neural networks and symmetry-aware architectures. We present four main contributions. First, we characterize the generic expressive power of geometric message-passing neural networks on sparse graphs and introduce EGenNet, an architecture achieving the corresponding guarantees. Second, we use representation theory to characterize linear layers equivariant to multiple permutations, including wreath-product symmetries. Third, we study permutation-invariant set representations that preserve containment, establish dimensional bounds and impossibility results, and introduce MASNet. Finally, we propose FSW-GNN, a message-passing architecture combining strong expressivity and bi-Lipschitz stability guarantees with strong empirical performance on long-range graph-learning tasks. Together, these results provide a broader understanding of how neural architectures can preserve the geometric, algebraic, order-theoretic, and metric structure of their inputs.

## Learning Graph-Conditioned Predictors in Weight Space

DevFeed: [Learning Graph-Conditioned Predictors in Weight Space](<https://devfeed.tech/articles/learning-graph-conditioned-predictors-in-weight-space-138847.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19250>)

Author: Zhewei Chen, Hao Zhu, Ahad N. Zehmakan

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [model architecture](<https://devfeed.tech/topics/model-architecture.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [lora](<https://devfeed.tech/topics/lora.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>)

Tags: [classification](<https://devfeed.tech/tags/classification.md>), [context](<https://devfeed.tech/tags/context.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [distillation](<https://devfeed.tech/tags/distillation.md>), [generator](<https://devfeed.tech/tags/generator.md>), [global](<https://devfeed.tech/tags/global.md>), [graph](<https://devfeed.tech/tags/graph.md>), [graphs](<https://devfeed.tech/tags/graphs.md>), [inference](<https://devfeed.tech/tags/inference.md>), [learning](<https://devfeed.tech/tags/learning.md>), [matrix](<https://devfeed.tech/tags/matrix.md>), [mechanism](<https://devfeed.tech/tags/mechanism.md>), [message](<https://devfeed.tech/tags/message.md>), [models](<https://devfeed.tech/tags/models.md>), [node](<https://devfeed.tech/tags/node.md>), [nodes](<https://devfeed.tech/tags/nodes.md>), [structure](<https://devfeed.tech/tags/structure.md>), [weight](<https://devfeed.tech/tags/weight.md>)

### AI overview

SCOPE adapts how node attributes are processed to each node's structural context. A small generator maps cached structural fingerprints to coefficients over a shared low-rank basis, yielding context-specific updates to a global weight matrix while the predictor, basis, and generator remain shared. After fingerprints are cached, inference needs only a forward pass and no online message passing. On two large OGB node-classification graphs, conditioning weights with raw positional signals yielded clear gains over concatenating those same signals to the inputs. Results with structure-transductive held-out nodes and teacher-free graph classification suggest the mechanism extends beyond one distillation setting.

### Source excerpt

arXiv:2611.19250v1 Announce Type: new Abstract: The same node attributes can imply different labels in different structural contexts. Yet fast graphless models typically treat topology as additional evidence supplied to a globally shared predictor. This assumes that structure augments the available information, rather than determining how node attributes should be interpreted. We introduce SCOPE (Structure-COnditioned PrEdictor), in which a small generator maps each node's cached structural fingerprint to coefficients over a shared low-rank basis, producing a context-specific update to a global weight matrix. The base predictor, basis, and generator remain shared across all nodes, while the resulting transformation adapts how attributes are processed in each relational environment. Once fingerprints are cached, inference requires only a forward pass and no online message passing. This design combines context-specific specialization with global parameter sharing, compact low-rank adaptation, and message-passing-free inference. On two large OGB node-classification graphs, conditioning the weights with raw positional signals yields clear gains over supplying the same signals through input concatenation. Further results under a structure-transductive held-out-node protocol and in teacher-free graph classification show that the mechanism extends beyond a particular distillation setting. These findings establish graph structure not only as predictive evidence, but as context that determines the predictor used to interpret that evidence.

## SERAG: SEmantical and Relational approach for Anomaly detection on knowledge Graphs

DevFeed: [SERAG: SEmantical and Relational approach for Anomaly detection on knowledge Graphs](<https://devfeed.tech/articles/serag-semantical-and-relational-approach-for-anomaly-detection-on-knowledge-graphs-138759.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18640>)

Author: Ludovic Moncla, Antoine Salazar, Yassir Lairgi, Khalid Benabdeslem, R\\'emy Cazabet

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Graph Mining](<https://devfeed.tech/topics/graph-mining.md>), [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [база знаний](<https://devfeed.tech/topics/tag-c51cc4bf76e5.md>)

Tags: [anomaly-detection](<https://devfeed.tech/tags/anomaly-detection.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [knowledge-graphs](<https://devfeed.tech/tags/knowledge-graphs.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [synthetic](<https://devfeed.tech/tags/synthetic.md>)

### AI overview

SERAG detects synthetic anomalies in knowledge graphs by fusing two plausibility scores: a semantic score from separately encoded subject, relation, and object labels, and a relational score from a relational graph convolutional network. The scores are standardized and calibrated before fusion. On a subset of ICEWS18 with vocabulary-constrained LLM-generated anomalies at 10% prevalence, SERAG achieved test AUPRCs of 0.321 with pooled splits and 0.258 with chronological splits, versus 0.159 and 0.146 for the strongest adapted baseline. Fusion improved mean test AUPRC over either branch, with a larger gain in the pooled protocol and a smaller validation-selected gain in the chronological protocol; additional anomaly supervision also improved performance. The evidence is limited to these synthetic anomalies and evaluated protocols.

### Source excerpt

arXiv:2611.18640v1 Announce Type: new Abstract: Knowledge graphs constructed from text can contain facts that are individually plausible yet inconsistent with their relational context. We introduce SERAG, a supervised anomaly detector that combines two complementary plausibility estimates: a semantic score computed from separately encoded subject, relation, and object labels, and a relational score obtained through a relational graph convolutional network. The scores are standardized, calibrated, and fused. We evaluate the method on a subset of ICEWS18 with vocabulary-constrained synthetic anomalies generated by an LLM, using pooled and chronological splits. At 10% anomaly prevalence, SERAG achieves test AUPRCs of 0.321 and 0.258, respectively, compared with 0.159 and 0.146 for the strongest adapted baseline. Fusion improves mean test AUPRC over either branch alone, with a larger gain in the pooled protocol and a smaller validation-selected gain in the chronological protocol. Additional anomaly supervision improves performance. These results support the complementarity of textual and relational signals for detecting synthetic anomalies under the evaluated protocols.

## Diagnosis as Percolation on an Evidence Graph: Root-Cause Recoverability in Network Fault Diagnosis

DevFeed: [Diagnosis as Percolation on an Evidence Graph: Root-Cause Recoverability in Network Fault Diagnosis](<https://devfeed.tech/articles/diagnosis-as-percolation-on-an-evidence-graph-root-cause-recoverability-in-network-fault-diagnosis-137028.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19584>)

Author: Jerome Henry, Swadhin Pradhan, Amine Choukir

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Networking](<https://devfeed.tech/sources/arxiv-networking.md>)

Topics: [graph theory](<https://devfeed.tech/topics/graph-theory.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>), [Wi-Fi](<https://devfeed.tech/topics/wi-fi.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [after](<https://devfeed.tech/tags/after.md>), [alarm](<https://devfeed.tech/tags/alarm.md>), [alternatives](<https://devfeed.tech/tags/alternatives.md>), [analyze](<https://devfeed.tech/tags/analyze.md>), [apply](<https://devfeed.tech/tags/apply.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [association](<https://devfeed.tech/tags/association.md>), [availability](<https://devfeed.tech/tags/availability.md>), [better](<https://devfeed.tech/tags/better.md>), [but](<https://devfeed.tech/tags/but.md>), [can](<https://devfeed.tech/tags/can.md>), [capture](<https://devfeed.tech/tags/capture.md>), [cause](<https://devfeed.tech/tags/cause.md>), [challenges](<https://devfeed.tech/tags/challenges.md>), [cs-ni](<https://devfeed.tech/tags/cs-ni.md>), [diagnosis](<https://devfeed.tech/tags/diagnosis.md>), [graph](<https://devfeed.tech/tags/graph.md>), [network](<https://devfeed.tech/tags/network.md>), [percolation](<https://devfeed.tech/tags/percolation.md>), [root](<https://devfeed.tech/tags/root.md>)

### AI overview

The evidence-graph framework separates whether evidence links a possible cause to a symptom, whether that cause can be distinguished from alternatives, and whether the resulting diagnosis is trustworthy. In simulations with average degree four, a giant component appears at about 25% evidence retention, while a particular cause-alarm pair reaches 50% connectivity only above half retention. For multi-LLM diagnosis of IEEE 802.11 packet captures, a reconciliation pass raises evidence-grounding weighted F1 from 0.912 to 0.964, mainly by removing isolated citations. Connectivity alone does not reliably indicate diagnostic quality after statistical controls; directed and weighted edges help in some cases, while simpler overlap measures or domain-specific methods sometimes do better. The supplied text is an abstract, so it does not provide full methodological detail.

### Source excerpt

arXiv:2611.19584v1 Announce Type: new Abstract: Diagnosing failures from incomplete observations presents three challenges: determining whether sufficient evidence connects a possible cause to a symptom, distinguishing the true cause from plausible alternatives, and assessing when a diagnosis can be trusted. We introduce an evidence-graph framework that separates these challenges and uses percolation theory to analyze how missing information affects cause-symptom connectivity. In controlled simulations with an average degree of four, a giant connected component emerges when approximately 25% of evidence is retained, but a specific cause-alarm pair reaches 50% connectivity only after slightly more than half is retained. Even connectivity does not guarantee correct cause identification. We apply this framework to multi-LLM diagnosis of IEEE 802.11 packet captures. A final reconciliation pass improves evidence-grounding weighted F1 from 0.912 to 0.964, primarily by removing isolated citations rather than connecting previously separate evidence chains, consistently across three graph constructions. However, connectivity is not a reliable confidence measure: its association with diagnostic quality becomes statistically insignificant after controlling for capture length, citation count, within-capture dependence, and multiple comparisons. Further experiments show that directed and weighted edges help selectively, while simpler overlap measures or domain-specific methods sometimes perform better. These findings establish evidence graphs as a framework for separating evidence availability, causal distinguishability, and confidence reliability, while demonstrating that connected evidence alone cannot certify diagnostic correctness.

[Next page](<https://devfeed.tech/topics/graph-theory.md?cursor=WyIyMDI2LTExLTE3VDA1OjAwOjAwKzAwOjAwIiwgIjg5Yjk4MTU1LWM1MDYtNGY2Yi1hMGYwLWU2NDM1Mzc2NjliMyJd>)