Pith. sign in

REVIEW 7 cited by

Specformer: Spectral Graph Neural Networks Meet Transformers

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2303.01028 v1 pith:5XB3M3KW submitted 2023-03-02 cs.LG cs.AIcs.SI

Specformer: Spectral Graph Neural Networks Meet Transformers

classification cs.LG cs.AIcs.SI
keywords spectralgraphspecformerfiltersgnnsdatasetslearnablenetworks
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Spectral graph neural networks (GNNs) learn graph representations via spectral-domain graph convolutions. However, most existing spectral graph filters are scalar-to-scalar functions, i.e., mapping a single eigenvalue to a single filtered value, thus ignoring the global pattern of the spectrum. Furthermore, these filters are often constructed based on some fixed-order polynomials, which have limited expressiveness and flexibility. To tackle these issues, we introduce Specformer, which effectively encodes the set of all eigenvalues and performs self-attention in the spectral domain, leading to a learnable set-to-set spectral filter. We also design a decoder with learnable bases to enable non-local graph convolution. Importantly, Specformer is equivariant to permutation. By stacking multiple Specformer layers, one can build a powerful spectral GNN. On synthetic datasets, we show that our Specformer can better recover ground-truth spectral filters than other spectral GNNs. Extensive experiments of both node-level and graph-level tasks on real-world graph datasets show that our Specformer outperforms state-of-the-art GNNs and learns meaningful spectrum patterns. Code and data are available at https://github.com/bdy9527/Specformer.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Graph Convolutional Attention: A Spectral Perspective on Graph Denoising and Diffusion

    cs.LG 2026-07 conditional novelty 6.5

    Under a denoising objective, linear attention is suboptimal; Graph Convolutional Attention matches idealized spectral attention on SBMs and improves graph denoising and diffusion in proportion to spectral diversity.

  2. Crys-JEPA: Accelerating Crystal Discovery via Embedding Screening and Generative Refinement

    cs.LG 2026-05 unverdicted novelty 6.0

    Crys-JEPA introduces a joint embedding predictive architecture that creates an energy-aware latent space, enabling embedding-based stability screening and a refinement pipeline that yields up to 72.7% gains on the V.S...

  3. Frequency-Corrupt Based Graph Self-Supervised Learning

    cs.LG 2026-04 unverdicted novelty 6.0

    FC-GSSL improves graph SSL by generating high-frequency biased corrupted graphs via low-frequency contribution-based corruption, reconstructing low-frequency features in an autoencoder, and aligning multi-view represe...

  4. SPECTRA: Spectral Domain-Aware Graph Generation for Imbalanced Molecular Property Regression

    cs.LG 2025-11 unverdicted novelty 6.0

    SPECTRA improves molecular property regression on underrepresented targets via spectral graph generation with rarity-aware budgeting and Laplacian interpolation, paired with edge-aware Chebyshev GNNs, yielding competi...

  5. S$^3$GNN: Efficient Global Mixing and Local Message Passing for Long-Range Graph Learning

    cs.LG 2026-05 unverdicted novelty 5.0

    S³GNN mitigates oversquashing in message-passing networks via lightweight global mixing without strong prior assumptions, yielding up to 10x error reduction and 50% fewer parameters across multiple domains.

  6. Fast and Featureless Node Representation Learning with Partial Pairwise Supervision

    cs.LG 2026-05 unverdicted novelty 5.0

    Contrastive FUSE learns node embeddings from partial pairwise supervision and structural signals alone by optimizing a spectral contrastive objective with a lightweight modularity approximation, yielding competitive p...

  7. Composable Crystals: Controllable Materials Discovery via Concept Learning

    cs.LG 2026-05 unverdicted novelty 5.0

    VQ-VAE concept learning enables controllable recombination of crystal motifs to generate structures with reported gains in validity-stability-uniqueness-novelty metrics on MP-20 and Alex-MP-20.