betaFeedback

WorksComputer Science

Learning Pseudo-Riemannian Manifolds for Heterophilic Graphs via Graph Signature

Abstract

Several widely used families of Graph Neural Networks carry a positive-definite geometric prior. Graph convolution read as heat diffusion needs a positive semi-definite operator, hyperbolic models place features on positive-definite Riemannian manifolds, and dot-product and GAT-style attention scores use positive-definite inner products. On heterophilic graphs, where edges often signify dissimilarity or structural repulsion, that prior is a mismatch: no positive-definite form can assign a negative squared distance to a direction the labels separate. To measure the mismatch, we introduce the Graph Signature Index (sGs_{\mathcal{G}}), the inertia of a class-compatibility matrix built from label topology alone, and we prove that no value of the edge or node homophily index in (0,1)(0,1) rules out a negative eigenvalue. We then propose the Pseudo-Riemannian Attention Network (PRAT), whose attention logit is a signed squared pseudo-distance under a learnable diagonal metric initialized from sGs_{\mathcal{G}}. The design is diagnose-then-learn: the signature supplies a sign structure that unconstrained training keeps outside near-zero directions, and the projection learns the rest of the metric around it. Across fifteen node-classification benchmarks PRAT outperforms or matches the strongest heterophily-adapted models. A controlled ablation ties that performance to the sign structure itself: with architecture and protocol fixed, forbidding the metric to be indefinite costs accuracy on the graphs sGs_{\mathcal{G}} marks as indefinite and leaves the others unchanged.

Introduction

Graph Neural Networks (GNNs) are the standard tool for relational learning [4]. Several of the most widely used families rest on a geometric prior rooted in Riemannian manifolds, typically Euclidean space, in which the metric tensor is strictly positive definite and connected nodes are therefore “close” and expected to share representations. The prior is explicit in three places. Graph convolution read as heat diffusion requires a positive semi-definite operator [5]; hyperbolic architectures such as HGCN [6] and HAT [47] embed features on positive-definite Riemannian manifolds; and dot-product [38] and GAT-style [39] attention use Euclidean inner products of projected features, which admit no negative squared distance. The assumption is not universal. Signed attention [1] and sheaf models [2] are natural counterexamples, and precisely because they work, the geometry deserves to be treated as a modeling choice rather than a default.

While this Riemannian inductive bias serves well for homophilic graphs, it creates a fundamental geometric mismatch when applied to heterophilic graphs, where edges often denote dissimilarity or structural repulsion [27] [49]. Heterophily-adapted GNNs address this architecturally with high-frequency spectral filters [11, 7, 14] or signed aggregation weights [1, 44, 41]. These are empirically effective, but the space they act in stays positive definite, so repulsion is expressed by distorting an embedding built for attraction [36].

We read heterophily geometrically, whereas the prevailing view is statistical. Non-Euclidean GNNs [6, 42] adapt curvature to hierarchical structure. We argue that some heterophilic graphs call instead for an indefinite metric, which admits negative squared distances [17] and so can sharpen across one class pair while smoothing across another. We diagnose the geometry the labels induce and then learn the metric inside it. The diagnostic is the Graph Signature Index (sGs_G), a spectral invariant of the label topology. The model is the Pseudo-Riemannian Attention Network (PRAT), whose attention logit is a signed squared pseudo-distance under a diagonal metric initialized from sGs_G and learned thereafter.

Theoretical Framework: The Geometry of Heterophily

Graph Signature Index

Previous studies proposed measures of graph homophily, including the node homophily index hVh_V [27] and edge homophily index hEh_E [50] described in Appendix § 8.5. Their values, ranging between 0 and 1, quantify the likelihood that nodes with identical labels are connected by edges (see Appendix § 8.§ 24). The graph signature index refines these indices by recording the signature of the eigenvalues of a real symmetric matrix constructed from a graph with node labels.

Definition 2.1. Let G:=(V,E)\mathcal{G} := (\mathcal{V}, \mathcal{E}) be an undirected graph without self-loops. Let l:V→L={1,2,⋯ ,∣L∣}l : \mathcal{V} \to L = \{1, 2, \cdots, |L|\} be the ground-truth node label function assigning each node to one of ∣L∣|L| classes. The graph signature matrix SGS_G is the ∣L∣×∣L∣|L| \times|L| real symmetric matrix whose (i,j)(i,j)-th entry is given as

si,j:=∣{(v,w)∈E∣l(v)=i,l(w)=j}∣∣{v∈V∣l(v)=i}∣⋅∣{v∈V∣l(v)=j}∣(1)s_{i,j} := \frac{|\{(v,w) \in\mathcal{E} \mid l(v) = i,l(w) = j\}|}{|\{v \in\mathcal{V} \mid l(v) = i\}| \cdot|\{v \in\mathcal{V} \mid l(v) = j\}|} \tag*{(1)}

The graph signature index of G\mathcal{G}, denoted as sGs_G, is the 3-tuple consisting of the number of positive, zero, and negative eigenvalues of SGS_G.

Because SGS_G is a real symmetric matrix, every eigenvalue of SGS_G is real. Sylvester’s law of inertia shows that the graph signature index is invariant under any invertible linear change of coordinates on R∣L∣\mathbb{R}^{|L|}.

Theorem 2.2 ([34]). Let S∈Rn×nS \in\mathbb{R}^{n \times n} be a real symmetric matrix. Then there exists a diagonal matrix D∈Rn×nD \in\mathbb{R}^{n \times n} and an invertible matrix W∈Rn×nW \in\mathbb{R}^{n \times n} such that S=WTDWS = W^{T} D W, where (⋅)T(\cdot)^T denotes the transpose of a square matrix. The numbers of positive, zero, and negative eigenvalues of SS are invariant under the change of coordinates.

Each (i,j)(i,j)-th entry of SGS_G is the expected probability that a node with a preassigned label jj lies in the depth 1 unfolding tree rooted at a node with label ii; see Appendix § 8.2. The signature of the eigenvalues of SGS_G can be used to determine whether graphs with discrete labels are heterophilic or homophilic. For example, graphs with two discrete node labels and E≠∅\mathcal{E} \ne\varnothing have three types of sGs_G; see Appendix § 8.3. Appendix § 8.4 provides heuristics on the density of these three graph types.

The following theorem relates the edge homophily level hEh_E to the graph signature index sGs_G.

Theorem 2.3 (Critical Homophily Threshold). Assume class ii contains nin_i nodes and N=∑i=1∣L∣niN = \sum_{i=1}^{|L|} n_i. Under a proportional allocation model, the expected graph signature matrix SGS_G has eigenvalues:

λ1>0,λ2=⋯=λ∣L∣=2∣E∣(hEN2−∑k=1∣L∣nk2)∑k=1∣L∣nk2⋅(N2−∑k=1∣L∣nk2).(2)\lambda_1 > 0,\qquad\lambda_2 = \cdots= \lambda_{|L|} = \frac{2|\mathcal{E}|\left(h_E N^2 - \sum_{k=1}^{|L|} n_k^2\right)}{\sum_{k=1}^{|L|} n_k^2 \cdot\left(N^2 - \sum_{k=1}^{|L|} n_k^2\right)}. \tag*{(2)}

The critical homophily threshold is given by: hE∗=∑k=1∣L∣nk2N2h_E^* = \frac{\sum_{k=1}^{|L|} n_k^2}{N^2}. For balanced classes where ni=nn_i = n for all ii, this reduces to hE∗=1/∣L∣h_E^* = 1/|L|.

While Theorem 2.3 reveals a correlation between edge homophily and graph signature under idealized conditions, homophily indices cannot fully characterize the spectral properties of SGS_{\mathcal{G}}. We formalize this limitation by showing that no value of hEh_E or hVh_V in (0,1)(0,1) rules out a negative eigenvalue of SGS_{\mathcal{G}}.

Theorem 2.4 (Limitation of Homophily Indices). For every rational h∈(0,1)h \in(0,1), there exists a graph G:=(V,E)\mathcal{G} := (\mathcal{V},\mathcal{E}) with discrete node labels such that hE=hV=hh_E = h_V = h and SGS_{\mathcal{G}} has a negative eigenvalue.

Limitations of Riemannian GNNs

We reformulate the problem of analyzing heterophilic graphs using pseudo-Riemannian manifolds. A pseudo-Riemannian manifold (M,g)(\mathcal{M},g) is a smooth manifold endowed with a metric tensor gp:TpM×TpM→Rg_p : T_p\mathcal{M} \times T_p\mathcal{M} \to\mathbb{R} which defines a non-degenerate symmetric bilinear form over the tangent space TpMT_p\mathcal{M} at each point p∈Mp \in\mathcal{M}. A pseudo-Riemannian manifold is Riemannian if the tensor gpg_p is positive-definite for every point p∈Mp \in\mathcal{M}. We keep this terminology for a symmetric bilinear form that is allowed to be degenerate. This is deliberate: the form on the hidden space is a pullback of rank at most ∣L∣|L|, hence degenerate whenever the hidden width exceeds the number of classes, and it depends on a hidden vector only through its image in L\mathcal{L}. Non-degeneracy returns on L\mathcal{L} itself whenever sGs_{\mathcal{G}} has no zero eigenvalue, and Theorem 2.5 counts only negative eigenvalues, so degeneracy does not affect it. We focus on the hh-dimensional manifold M\mathcal{M} endowed with a metric tensor gp:=S∈Rh×hg_p := S \in\mathbb{R}^{h \times h} for some fixed real symmetric matrix SS at every point p∈Mp \in\mathcal{M}, which defines the following quadratic form on tangent vectors w1,w2∈TpMw_1,w_2 \in T_p\mathcal{M}: gp(w1,w2):=w1TSw2g_p(w_1,w_2) := w_1^{\mathsf{T}} S w_2 [24, 37].

Let G:=(V,E)\mathcal{G} := (\mathcal{V},\mathcal{E}) be a graph with continuous node attributes X:V→RhX : \mathcal{V} \to\mathbb{R}^h and embedded node labels l:V→R∣L∣l : \mathcal{V} \to\mathbb{R}^{|L|}. We set the metric tensor over R∣L∣\mathbb{R}^{|L|} to be the graph signature matrix SGS_{\mathcal{G}} of the given graph G\mathcal{G}. By doing so, we regard the graph G\mathcal{G} as a subspace of some smooth manifold M\mathcal{M} admitting a smooth function f:M→L≅R∣L∣f : \mathcal{M} \to\mathcal{L} \cong\mathbb{R}^{|L|}, whose metric tensor is the pullback of metric tensor f∗SGf^*S_{\mathcal{G}}. Given a metric tensor gg on L\mathcal{L}, its pullback f∗gf^*g on M\mathcal{M} is defined as

(f∗g)p(v,w):=gf(p)(Tpf(v),Tpf(w)).(f^*g)_p(v,w) := g_{f(p)}(T_p f(v),T_p f(w)).

Because SGS_{\mathcal{G}} can have negative eigenvalues, M\mathcal{M} is a pseudo-Riemannian manifold, thereby covering a class of manifolds which are not Riemannian. The theorem below, proved in Appendix § 8.§ 25, makes this observation explicit.

Theorem 2.5 (Geometric Mismatch). Let M\mathcal{M} and L\mathcal{L} be differentiable manifolds and f:M→Lf : \mathcal{M} \to\mathcal{L} be a smooth map. If SGS_{\mathcal{G}} defined on L\mathcal{L} is indefinite and the pushforward Tpf:TpM→Tf(p)LT_p f : T_p\mathcal{M} \to T_{f(p)}\mathcal{L} is surjective for every p∈Mp \in\mathcal{M}, then there exists a diffeomorphism ϕ:L→L\phi: \mathcal{L} \to\mathcal{L} such that the pullback of the Euclidean metric II, the identity form, satisfies (ϕ∘f)∗I=f∗SG(\phi\circ f)^*I = f^*S_{\mathcal{G}}.

Theorem 2.5 establishes a geometric mismatch between the Riemannian prior and graphs whose SGS_{\mathcal{G}} is indefinite. It proves that this geometric mismatch is intrinsic and cannot be bridged by any smooth transformation of the Riemannian space. Surjectivity of the pushforward is what carries the indefiniteness through to the pullback: the image of TpfT_p f is then all of Tf(p)LT_{f(p)}\mathcal{L}, so a negative direction of SGS_{\mathcal{G}} has a preimage. Injectivity alone would not suffice, since an injective map can pull an indefinite form back to a degenerate one. In our architecture TpfT_p f is the projection WW of Equation (4), surjective exactly when rank⁡W=∣L∣\operatorname{rank} W = |L|. Every configuration we report has h′>∣L∣h' > |L|, so the hypothesis holds for almost every WW. Appendix A.8 discusses further implications for heterophilic graphs.

Method: Pseudo-Riemannian Attention Network (PRAT)

Pseudo-Riemannian Edge Attention using continuous features

References

  1. [1]Deyu Bo, Xiao Wang, Chuan Shi, and Huawei Shen. Beyond low-frequency information in graph convolutional networks. In AAAI. AAAI Press, 2021.
  2. [2]Cristian Bodnar, Francesco Di Giovanni, Benjamin Paul Chamberlain, Pietro Liò, and Michael M. Bronstein. Neural sheaf diffusion: A topological perspective on heterophily and oversmoothing in GNNs. In Advances in Neural Information Processing Systems, volume 35, pages 18527–18541, 2022.
  3. [3]Shaked Brody, Uri Alon, and Eran Yahav. How attentive are graph attention networks? In International Conference on Learning Representations, 2022. URL https://openreview.net/forum?id=F72ximsx7C1.arxiv.org/abs/2105.14491
  4. [4]Michael M. Bronstein, Joan Bruna, Yann LeCun, Arthur Szlam, and Pierre Vandergheynst. Geometric deep learning: Going beyond euclidean data. IEEE Signal Processing Magazine, 34 (4):18–42, 2017. doi: 10.1109/MSP.2017.2693418.DOI
  5. [5]Benjamin Paul Chamberlain, James Rowbottom, Maria Goronova, Stefan Webb, Emanuele Rossi, and Michael M Bronstein. Grand: Graph neural diffusion. Proceedings of the 38th International Conference on Machine Learning (ICML) 2021, 18-24 July 2021, Virtual Event, 2021.arxiv.org/abs/2106.10934
  6. [6]Ines Chami, Zhitao Ying, Christopher Ré, and Jure Leskovec. Hyperbolic graph convolutional neural networks. In Advances in Neural Information Processing Systems, pages 4869–4880, 2019.arxiv.org/abs/1910.12933
  7. [7]Eli Chien, Jianhao Peng, Pan Li, and Olgaic Milenkovic. Adaptive universal generalized pagerank graph neural network. In International Conference on Learning Representations, 2021. URL https://openreview.net/forum?id=n6j17fLxrP.arxiv.org/abs/2006.07988
  8. [8]Yun Young Choi, Sun Woo Park, Youngho Woo, and U Jin Choi. Cycle to clique (cy2c) graph neural network: A sight to see beyond neighborhood aggregation. In The eleventh international conference on learning representations, 2022.
  9. [9]Yun Young Choi, Sun Woo Park, Minho Lee, and Youngho Woo. Topology-informed graph transformer. In Proceedings of the Geometry-grounded Representation Learning and Generative Modeling Workshop (GRaM), volume 251 of Proceedings of Machine Learning Research, pages 20–34. PMLR, 2024.arxiv.org/abs/2402.02005
  10. [10]Lun Du, Xiaozhou Shi, Qiang Fu, Xiaojun Ma, Hengyu Liu, Shi Han, and Dongmei Zhang. GBK-GNN: Gated bi-kernel graph neural networks for modeling both homophily and heterophily. In Proceedings of the ACM Web Conference 2022 (WWW), 2022.arxiv.org/abs/2110.15777
  11. [11]Johannes Gasteiger, Aleksandar Bojchevski, and Stephan Günnemann. Predict then propagate: Graph neural networks meet personalized pagerank. In International Conference on Learning Representations, 2019. URL https://openreview.net/forum?id=H1gL-2A9Ym.arxiv.org/abs/1810.05997
  12. [12]William L. Hamilton, Rex Ying, and Jure Leskovec. Inductive representation learning on large graphs. In NIPS, 2017.arxiv.org/abs/1706.02216
  13. [13]Allen Hatcher. Algebraic Topology. Cambridge University Press, 2002.
  14. [14]Mingguo He, Zhewei Wei, Zengfeng Huang, and Hongteng Xu. Bernnet: Learning arbitrary graph spectral filters via bernstein approximation. In NeurIPS, 2021.arxiv.org/abs/2106.10994
  15. [15]Feng Ji, Yanan Zhao, Kai Zhao, Hanyang Meng, Jielong Yang, and Wee Peng Tay. Rethinking graph neural networks from a geometric perspective of node features. In The Thirteenth International Conference on Learning Representations, 2025. URL https://openreview.net/forum?id=1BMRmw59Lk.
  16. [16]Thomas N. Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. In International Conference on Learning Representations, 2017. URL https://openreview.net/forum?id=SJU4ayYgl.arxiv.org/abs/1609.02907
  17. [17]Marc Law and Jos Stam. Ultrahyperbolic representation learning. In H. Larochelle, M. Ranzato, R. Hadsell, M.F. Balcan, and H. Lin, editors, Advances in Neural Information Processing Systems, volume 33, pages 1668–1678. Curran Associates, Inc., 2020. URL https://proceedings.neurips.cc/paper_files/paper/2020/file/123b7f02433572a0a560e620311a469c-Paper.pdf.arxiv.org/abs/2007.00211
  18. [18]Viet Quan Le and Cuong Van Ta. Pseudo-riemannian graph transformer. In Advances in Neural Information Processing Systems, volume 38, pages 15951–15983, 2025.DOI
  19. [19]Xiang Li, Renyu Zhu, Yao Cheng, Caihua Shan, Siqiang Luo, Dongsheng Li, and Weining Qian. Finding global homophily in graph neural networks when meeting heterophily. In International Conference on Machine Learning (ICML), 2022.arxiv.org/abs/2205.07308
  20. [20]Qincheng Lu, Jiaq i Zhu, Sitao Luan, and Xiao-Wen Chang. Representation learning on heterophilic graph with directional neighborhood attention, 2024. URL https://arxiv.org/abs/2403.01475.
  21. [21]Sitao Luan, Chenqing Hua, Qincheng Lu, Jiaq i Zhu, Mingde Zhao, Shuyuan Zhang, Xiao-Wen Chang, and Doina Precup. Revisiting heterophily for graph neural networks. In Advances in Neural Information Processing Systems, 2022.arxiv.org/abs/2210.07606
  22. [22]Liheng Ma, Soumyasundar Pal, Yingxue Zhang, Philip H. S. Torr, and Mark Coates. Plain transformers can be powerful graph learners. Transactions on Machine Learning Research, 2026. URL https://openreview.net/forum?id=bEmDvPOfdv.arxiv.org/abs/2504.12588
  23. [23]Sunil Kumar Maurya, Xin Liu, and Tsuyoshi Murata. Simplifying approach to node classification in Graph Neural Networks. Journal of Computational Science, 62:101695, 2022.
  24. [24]John Douglas Moore. Lectures on differential geometry. Lecture notes, University of California, Santa Barbara. https://web.math.ucsb.edu/~moore/riemanniangeometry.pdf, 2009.
  25. [25]Sun Woo Park, Yun Young Choi, Dosang Joe, U Jin Choi, and Youngho Woo. The pwlr graph representation: A persistent weisfeiler-lehman scheme with random walks for graph classification. In Topological, Algebraic and Geometric Learning Workshops 2022, pages 287–297. PMLR, 2022.arxiv.org/abs/2208.13427
  26. [26]Sun Woo Park, Yun Young Choi, U Jin Choi, and Youngho Woo. A functorial formulation of neighborhood aggregating deep learning. arXiv preprint arXiv:2604.24672, 2026.arxiv.org/abs/2604.24672
  27. [27]Hongbin Pei, Bingzhe Wei, Kevin Chen-Chuan Chang, Yu Lei, and Bo Yang. Geom-gcn: Geometric graph convolutional networks. In International Conference on Learning Representations, 2020. URL https://openreview.net/forum?id=S1e2agrFvS.arxiv.org/abs/2002.05287
  28. [28]Oleg Platonov, Denis Kuznedelev, Michael Diskin, Artem Babenko, and Liudmila Prokhorenkova. A critical look at the evaluation of GNNs under heterophily: Are we really making progress? In International Conference on Learning Representations, 2023.arxiv.org/abs/2302.11640
  29. [29]Ladislav Rampášek, Michael Galkin, Vijay Prakash Dwivedi, Anh Tuan Luu, Guy Wolf, and Dominique Beaini. Recipe for a general, powerful, scalable graph transformer. Advances in Neural Information Processing Systems, 35:14501–14515, 2022.arxiv.org/abs/2205.12454
  30. [30]Benedek Rozemberczki, Carl Allen, and Rik Sarkar. Multi-scale attributed node embedding. Journal of Complex Networks, 9(2):cnab014, 2021.
  31. [31]Patrick Rubin-Delanchy, Joshua Cape, Minh Tang, and Carey E. Priebe. A statistical interpretation of spectral embedding: the generalised random dot product graph. Journal of the Royal Statistical Society Series B, 84(4):1446–1473, 2022.
  32. [32]Prithviraj Sen, Galileo Namata, Mustafa Bilgic, Lise Getoor, Brian Gallagher, and Tina Eliassi-Rad. Collective classification in network data. AI magazine, 29(3):93–93, 2008.DOI
  33. [33]Yunsheng Shi, Zhengjie Huang, Shikun Feng, Hui Zhong, Wenjing Wang, and Yu Sun. Masked label prediction: Unified message passing model for semi-supervised classification. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21, pages 1548–1554. International Joint Conferences on Artificial Intelligence Organization, 2021. doi: 10.24963/ijcai.2021/214.DOI
  34. [34]J. J. Sylvester. Xix. a demonstration of the theorem that every homogeneous quadratic polynomial is reducible by real orthogonal substitutions to the form of a sum of positive and negative squares. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 4(23):138–142, 1852.DOI
  35. [35]Jie Tang, Jimeng Sun, Chi Wang, and Zi Yang. Social influence analysis in large-scale networks. In Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 807–816, 2009.DOI
  36. [36]Jake Topping, Francesco Di Giovanni, Benjamin Paul Chamberlain, Xiaowen Dong, and Michael M. Bronstein. Understanding over-squashing and bottlenecks on graphs via curvature. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum?id=7UmjRGzp-A.arxiv.org/abs/2111.14522
  37. [37]Loring Tu. Differential Geometry: Connections, Curvature, and Characteristic Classes. Springer, 2017.DOI
  38. [38]Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017.arxiv.org/abs/1706.03762
  39. [39]Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, and Yoshua Bengio. Graph attention networks. In International Conference on Learning Representations, 2018. URL https://openreview.net/forum?id=rJXMpikCZ.arxiv.org/abs/1710.10903
  40. [40]Xiyuan Wang and Muhan Zhang. How powerful are spectral graph neural networks. In International Conference on Machine Learning (ICML), 2022.arxiv.org/abs/2205.11172
  41. [41]Xudong Wang, Tongxin Li, Chris Ding, and Jicong Fan. Adaptive riemannian graph neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence, 2026.
  42. [42]Bo Xiong, Shichao Zhu, Nico Potyka, Shirui Pan, Chuan Zhou, and Steffen Staab. Pseudo-riemannian graph convolutional networks. In Advances in Neural Information Processing Systems, 2022.arxiv.org/abs/2106.03134
  43. [43]Keyulu Xu, Weihua Hu, Jure Leskovec, and Stefanie Jegelka. How powerful are graph neural networks? In International Conference on Learning Representations, 2019. URL https://openreview.net/forum?id=ryGs6iA5Km.arxiv.org/abs/1810.00826
  44. [44]Yujun Yan, Milad Hashemi, Kevin Swersky, Yaoqing Yang, and Danai Koutra. Two sides of the same coin: Heterophily and oversmoothing in graph convolutional neural networks. In 2022 IEEE International Conference on Data Mining (ICDM), pages 1287–1292. IEEE, 2022.arxiv.org/abs/2102.06462
  45. [45]Zhilin Yang, William Cohen, and Ruslan Salakhudinov. Revisiting semi-supervised learning with graph embeddings. In International conference on machine learning, pages 40–48. PMLR, 2016.arxiv.org/abs/1603.08861
  46. [46]Chengxuan Ying, Tianle Cai, Shengjie Luo, Shuxin Zheng, Guolin Ke, Di He, Yanming Shen, and Tie-Yan Liu. Do transformers really perform badly for graph representation? Advances in neural information processing systems, 34:28877–28888, 2021.
  47. [47]Yiding Zhang, Xiao Wang, Chuan Shi, Xunqiang Jiang, and Yanfang Fanny Ye. Hyperbolic graph attention network. IEEE Transactions on Big Data, 2021.arxiv.org/abs/1912.03046
  48. [48]Yilun Zheng, Sitao Luan, and Lihui Chen. What is missing for graph homophily? disentangling graph homophily for graph neural networks. In Advances in Neural Information Processing Systems, volume 37, pages 68406–68452, 2024.DOI

Paper details

vlnc:W_7rvcv8gs
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Computer Science

Not recorded for this paper

View statement
Looking for related papers…
Contents