Metraipolis › archive › dc_a88219ceadd3

Revisiting the Generalization of Neural Graph Edit Distance Models — arXiv:2610.04644

Source
arXiv
Original
https://arxiv.org/abs/2610.04644
License (as archived)
CC0-1.0
Archived
2026-10-06T05:30:39.774Z
SHA-256 of the archived text
4d418c13acfea7146803cfcaaad5e6521ff23daa2ad60f8a242d03e7970671fc
Stable URL of this page
https://metraipolis.com/documento/dc_a88219ceadd3

How to cite

Revisiting the Generalization of Neural Graph Edit Distance Models — arXiv:2610.04644 — arXiv, https://arxiv.org/abs/2610.04644 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a88219ceadd3 (Metraipolis archive dc_a88219ceadd3, sha256 4d418c13acfea7146803cfcaaad5e6521ff23daa2ad60f8a242d03e7970671fc)
BibTeX
@misc{dca88219ceadd3,
  title = {Revisiting the Generalization of Neural Graph Edit Distance Models — arXiv:2610.04644},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.04644},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a88219ceadd3; sha256 4d418c13acfea7146803cfcaaad5e6521ff23daa2ad60f8a242d03e7970671fc},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04644. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Neural approaches to Graph Edit Distance (GED) have achieved strong results under standard within-dataset evaluation, but much less is known about how well these models transfer across graph collections. We conduct a systematic study of this problem using exact GED supervision across diverse graph datasets and a broad set of representative learning-based methods. Our results reveal a pronounced gap between within-collection performance and cross-collection transfer. Models that perform well on their training collections often lose this advantage when evaluated on structurally different data. Training on multiple source collections substantially improves zero-shot transfer and provides a better starting point when limited supervision is available for a new target collection. Further analysis shows that transfer behavior varies with the source--target direction and the structural characteristics of the collections involved. These findings suggest that conventional within-collection evaluation provides only a partial view of the generalization behavior of neural GED models and motivate broader evaluation across heterogeneous graph collections.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.04644 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T05:27:05.703Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).