Metraipolis › archive › dc_ba562e1680ec
A Framework for Task-Agnostic Counterfactual Fairness in Transformers through Activation Steering — arXiv:2611.00298
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.00298
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T10:45:49.447Z
- SHA-256 of the archived text
7700e8ff0975dd62d5c859e5560f8adc76098f88a15b759d492c39969f64056b- Stable URL of this page
https://metraipolis.com/documento/dc_ba562e1680ec
How to cite
A Framework for Task-Agnostic Counterfactual Fairness in Transformers through Activation Steering — arXiv:2611.00298 — arXiv (en), https://arxiv.org/abs/2611.00298 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ba562e1680ec (Metraipolis archive dc_ba562e1680ec, sha256 7700e8ff0975dd62d5c859e5560f8adc76098f88a15b759d492c39969f64056b)
BibTeX
@misc{dcba562e1680ec,
title = {A Framework for Task-Agnostic Counterfactual Fairness in Transformers through Activation Steering — arXiv:2611.00298},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.00298},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ba562e1680ec; sha256 7700e8ff0975dd62d5c859e5560f8adc76098f88a15b759d492c39969f64056b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00298. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Bias in transformer-based models is a growing concern as these architectures become more prevalent in automated decision-making systems, such as prediction models for hiring, lending, and clinical treatment recommendation. Mechanistic interpretability (MI) is a growing body of work that measures meaningful causal effects of activations on model output by identifying human-interpretable features encoded in model activations. When such features correspond to sensitive attributes, they provide a direct way to quantify the bias those attributes introduce and to causally intervene to mitigate it. However, counterfactual fairness (CF) methods have not been systematically adapted to this setting of intervening on interpretable mechanisms in transformers. We propose a framework, Mechanistic Interpretability for Counterfactual Fairness (MICF), that utilizes MI tools to reformulate CF constraints in terms of transformer activations, unifying former approaches into a single pipeline for fairness analysis in transformers. MICF does not require a prior causal model or specific transformer architecture. We evaluate MICF on real-world tabular data and show that CF can be improved across different prediction tasks. 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); Computers and Society (cs.CY). Pagina del lavoro: https://arxiv.org/abs/2611.00298 — 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 2026-11-03T10:39:50.391Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).