Metraipolis › archive › dc_09ca66f6245c
Interpreting Supervised Pairwise Graph Representations: Local Evidence, Confounding, and Provable Post-hoc Correction — arXiv:2610.32611
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.32611
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-26T20:12:23.603Z
- SHA-256 of the archived text
6428d008e8a483e8e1bbaa748bbbd80e0b6f9d0428625c2cf5f2a3bf91b76865- Stable URL of this page
https://metraipolis.com/documento/dc_09ca66f6245c
How to cite
Interpreting Supervised Pairwise Graph Representations: Local Evidence, Confounding, and Provable Post-hoc Correction — arXiv:2610.32611 — arXiv (en), https://arxiv.org/abs/2610.32611 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_09ca66f6245c (Metraipolis archive dc_09ca66f6245c, sha256 6428d008e8a483e8e1bbaa748bbbd80e0b6f9d0428625c2cf5f2a3bf91b76865)
BibTeX
@misc{dc09ca66f6245c,
title = {Interpreting Supervised Pairwise Graph Representations: Local Evidence, Confounding, and Provable Post-hoc Correction — arXiv:2610.32611},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.32611},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_09ca66f6245c; sha256 6428d008e8a483e8e1bbaa748bbbd80e0b6f9d0428625c2cf5f2a3bf91b76865},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.32611. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Supervised pairwise graph representations convert pairwise predictive relations into sample-specific graphs, but graph-level accuracy does not by itself show that higher-order topology carries the signal. In a controlled Transformer error-detection task, supervised pairwise graphs outperform attention-as-adjacency (ROC-AUC $0.839$ versus $0.559$; the comparable no-confidence pair is $0.687$ versus $0.559$), but their apparent advantage over model confidence disappears after controlling for sequence length, which is almost perfectly recoverable from the pooled hidden state ($R^2=0.9995$). On published HCP Emotion ensemble graphs, signed node aggregation already acts as an ensemble of supervised pairwise predictions. Edge magnitude alone is near chance; topology-aware augmentations change ROC-AUC by at most $0.004$, with only sign-separated degree showing small positive changes across all tested graph sizes, while a validation-tuned three-layer GCN provides no measurable improvement over the signed node-summary baseline. A Fisher post-hoc corrector satisfies its distribution-agnostic guarantees across the evaluated feature representations and remains stable under reduced labelled budgets; in a near-error-free fMRI calibration regime the bound correctly becomes statistically uninformative. A separate cross-attention analysis identifies a redundant reversal circuit and downstream EOS-timing and copy failures. Higher-order topology should therefore be credited only after local supervised evidence and recoverable nuisance variables have been excluded. 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/2610.32611 — 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 2026-10-26T19:49:03.417Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).