Metraipolis › archive › dc_992dd203848c

Coupling Noisy Pairwise Knowledge to the DAG Posterior for Causal Discovery — arXiv:2610.04559

Source
arXiv
Original
https://arxiv.org/abs/2610.04559
License (as archived)
CC0-1.0
Archived
2026-10-06T05:26:19.142Z
SHA-256 of the archived text
5c076d81069b833b633298c771f78df2bf0262310300a0ebe79842b7d0815c0e
Stable URL of this page
https://metraipolis.com/documento/dc_992dd203848c

How to cite

Coupling Noisy Pairwise Knowledge to the DAG Posterior for Causal Discovery — arXiv:2610.04559 — arXiv, https://arxiv.org/abs/2610.04559 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_992dd203848c (Metraipolis archive dc_992dd203848c, sha256 5c076d81069b833b633298c771f78df2bf0262310300a0ebe79842b7d0815c0e)
BibTeX
@misc{dc992dd203848c,
  title = {Coupling Noisy Pairwise Knowledge to the DAG Posterior for Causal Discovery — arXiv:2610.04559},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.04559},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_992dd203848c; sha256 5c076d81069b833b633298c771f78df2bf0262310300a0ebe79842b7d0815c0e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04559. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

External causal reports can improve structure learning from limited observations, but their reliability varies across sources and variable pairs. We introduce HB-NoisyKG, a Bayesian framework that combines observational data with repeated causal reports from sources such as large language models. Each report is a noisy observation of a direct pair state implied by one DAG. A feature-conditioned Beta prior pools information about pair reliability, and a shared error matrix captures systematic mistakes. Alternating inference uses the graph posterior to refine reliability estimates, which determine how reports influence subsequent graph updates. The report likelihood uses only graph pair-state marginals, so the same observation layer supports discrete and continuous likelihoods in graph-only and joint inference. Against an 80-restart no-KG baseline, HB uses at most 80 total restarts and lowers mean SHD from 22.39 to 16.06 on five discrete benchmarks. On a physical light tunnel with random variable IDs and retained descriptions, HB lowers SHD from 39.00 for no-KG to 27.30. On continuous Sachs, graph-only BGe raises AUROC by 0.121 over no-KG Top-K. In a controlled synthetic study, continued updating also reduces mean reliability estimation error and held-out report log loss compared with one-time estimation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.04559 — 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 2026-10-06T05:24:05.881Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).