Metraipolis › archive › dc_effd4960adc5

Support-Set Target Leakage in Relational Foundation Models during In-Context Learning: Impact, Detection, and Mitigation — arXiv:2609.36384

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.36384
License (as archived)
CC0-1.0
Archived
2026-10-07T16:22:33.178Z
SHA-256 of the archived text
e1dae7b27a98a1b7fe08e65423f6c108754637757898c6c32d7a19e1df4749ea
Stable URL of this page
https://metraipolis.com/documento/dc_effd4960adc5

How to cite

Support-Set Target Leakage in Relational Foundation Models during In-Context Learning: Impact, Detection, and Mitigation — arXiv:2609.36384 — arXiv (en), https://arxiv.org/abs/2609.36384 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_effd4960adc5 (Metraipolis archive dc_effd4960adc5, sha256 e1dae7b27a98a1b7fe08e65423f6c108754637757898c6c32d7a19e1df4749ea)
BibTeX
@misc{dceffd4960adc5,
  title = {Support-Set Target Leakage in Relational Foundation Models during In-Context Learning: Impact, Detection, and Mitigation — arXiv:2609.36384},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.36384},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_effd4960adc5; sha256 e1dae7b27a98a1b7fe08e65423f6c108754637757898c6c32d7a19e1df4749ea},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.36384. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Relational in-context learning (ICL) conditions predictions on the labeled support examples and their linked tables, creating a failure mode when the support set contains target-derived features that are unavailable for the query. We formulate this problem as support-set target leakage, distinct from leakage during dataset construction, temporal splitting, or representation learning. Here, the target-derived (leaker) columns are present only in the labeled support set during relational in-context inference, while queries remain clean. We construct 14 synthetic leaker types, corresponding to 20 columns, spanning proxies with different noise levels, coverage, modalities, semantic transparency, and relational distances. We evaluate a frozen relational encoder with an ICL head on held-out RelBench databases and use Integrated Gradients (IG) to rank and remove suspicious columns. Our results show that the effect of support-set leakage varies across tasks and relational distances. Target-table leakers cause the clearest degradation, while one- and two-hop leakers are not consistently used by the model. IG ranks target-table leakers highly across datasets and partially recovers performance in settings where leakage has the largest effect.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.36384 — 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 2026-10-07T15:54:53.494Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).