Aug 2026· Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2· pp. 1851-1861· 0 citations· 21 references
Abstract
Text-attributed graphs are widely used across domains, offering rich opportunities for zero-shot learning via graph-text alignment. However, existing methods struggle with tasks requiring fine-grained pattern recognition, particularly on heterophilic graphs. Through empirical and theoretical analysis, we identify an over-abstraction problem: current approaches operate at excessively large hyperbolic radii, compressing multi-scale structural information into uniform high-level abstractions. This abstraction-induced information loss obscures critical local patterns essential for accurate predictions. By analyzing embeddings in hyperbolic space, we demonstrate that optimal graph learning requires faithful preservation of fine-grained structural details, better retained by representations positioned closer to the origin. To address this, we propose H4G, a framework that systematically reduces embedding radii using learnable block-diagonal scaling matrices and Möbius matrix multiplication. This approach restores access to fine-grained patterns while maintaining global receptive ability with minimal computational overhead. Experiments show H4G achieves state-of-the-art zero-shot performance with 12.8% improvement on heterophilic graphs and 8.4% on homophilic graphs, confirming that radius reduction enables faithful multi-scale representation for advancing zero-shot graph learning.
It is found that SAE activation sets do not recover human category boundaries or within-category typicality more faithfully than dense embeddings or residual-stream states, but instead track model-internal similarity structure.
Nikolai Bolik, Lennart Stöpler, Artur Andrzejak· 0 citations
This work introduces Topoformer, a lightweight and scalable framework for graph representation learning that encodes topological structure into attention-friendly sequences by decomposing a graph into a short, ordered sequence of topological tokens by slicing over node or edge filtrations.
Md Joshem Uddin, Astrit Tola, C. Akcora et al.· arXiv.org· 1 citation
This framework introduces a dual-encoder architecture that separately processes structural and attribute information, incorporates node positional encoding to approximate Neighborhood Identity Distribution (NID), and employs dual reconstruction tasks for both edges and node attributes.
Graph Neural Networks exhibit a puzzling numerical fragility under mixed-precision training, failing significantly more often than MLPs or CNNs. This failure is inherently tied to graph structure, with heterophilic graphs and high-degree nodes being particularly vulnerable. We identify the root cause as catastrophic cancellation during neighborhood aggregation. When neighboring node embeddings point in opposite directions, their sum collapses toward zero and amplifies floating-point errors by orders of magnitude. We formalize this through the cancellation ratio ?, proving that it is fundamentally governed by graph topology, including heterophily, node degree, and network depth. Consequently, we propose Aggregation-Aware Representation Learning (AARL) to learn numerically stable and cancellation-resistant representations without sacrificing expressiveness. Unlike naive approaches that enforce neighbor alignment and destroy discriminative power, AARL maintains representation diversity while ensuring numerically safe aggregation. Experiments on diverse benchmarks demonstrate that AARL substantially improves training stability under low precision while preserving or improving classification accuracy.
Jiawei Gu, Ziyue Qiao· Proceedings of the 32nd ACM...· 0 citations
A theoretical analysis positioning CoCo with respect to related objectives shows that the new proposed loss benefits from closer initialization to the optimal configuration, more informative gradients, and stronger incentives for class-wise representation collapse.
Blanca Cano-Camarero, 'Angela Fern'andez-Pascual, José R. Dorronsoro· arXiv.org· 0 citations
This work proposes Hyperspherical Representation Equilibrium Shift (HyRES), a theoretically grounded, unsupervised framework that redefines node influence as a geometric displacement within the latent space, and leverages Linear Response Theory to derive a closed-form approximation of the global representation shift caused by a node's removal.
Yantuan Xian, Chunping Li, Hongbin Wang et al.· Proceedings of the 32nd ACM...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
