Metraipolis › archive › dc_7ce4a5cb8f67

Accuracy Is Not Enough: A Cross-Architecture Audit of Demographic Bias in Deep Knowledge Tracing — arXiv:2609.20249

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.20249
License (as archived)
CC0-1.0
Archived
2026-10-10T04:08:57.188Z
SHA-256 of the archived text
f1ae318bbf4029df366a4b18122fcbdbae77e61c7621d566ef691b73754892b8
Stable URL of this page
https://metraipolis.com/documento/dc_7ce4a5cb8f67

How to cite

Accuracy Is Not Enough: A Cross-Architecture Audit of Demographic Bias in Deep Knowledge Tracing — arXiv:2609.20249 — arXiv (en), https://arxiv.org/abs/2609.20249 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7ce4a5cb8f67 (Metraipolis archive dc_7ce4a5cb8f67, sha256 f1ae318bbf4029df366a4b18122fcbdbae77e61c7621d566ef691b73754892b8)
BibTeX
@misc{dc7ce4a5cb8f67,
  title = {Accuracy Is Not Enough: A Cross-Architecture Audit of Demographic Bias in Deep Knowledge Tracing — arXiv:2609.20249},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.20249},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7ce4a5cb8f67; sha256 f1ae318bbf4029df366a4b18122fcbdbae77e61c7621d566ef691b73754892b8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20249. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Deep knowledge tracing (DKT) models implicitly decide which students an adaptive system believes have mastered a skill, yet almost all evidence on their demographic fairness comes from Bayesian knowledge tracing; the deep models that power modern systems have received no comparable cross-architecture audit. We close this gap: four architectures (DKT, DKVMN, SAKT, AKT) trained under three regimes (standard, reweighting, adversarial) on two public datasets with demographic metadata, Eedi (15.9M interactions) and OULAD (167k after preprocessing), evaluated with ABROCA, student-level bootstrap confidence intervals, and permutation tests addressing recent critiques of fairness-metric instability. Three findings emerge. (i) Bias is real but context-dependent: every architecture shows a significant socioeconomic ABROCA on Eedi (0.018-0.023, $p<0.005$), with per-group AUC lower for economically disadvantaged students, while gender bias is significant on OULAD for three of four architectures after multiplicity correction yet negligible on Eedi. (ii) The most accurate architecture is the most biased: AKT gains about 4 AUC points from item-level Rasch embeddings and shows the largest socioeconomic ABROCA, exceeding every other architecture under a paired bootstrap ($p\leq0.002$); ablating only the Rasch embeddings removes the accuracy gain and the excess bias together. (iii) Standard mitigation is unreliable: reweighting and adversarial debiasing leave ABROCA essentially unchanged in every configuration that preserves accuracy, even though the adversary is pinned at chance at full reversal strength and a weak-strength positive control rules out a dead probe.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.20249 — 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 2026-10-10T03:48:18.339Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).