Metraipolis › archive › dc_8136556752f8

MUtE: A Dual Framework for Concept Erasure and Counterfactual Interventions — arXiv:2609.11253

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.11253
License (as archived)
CC0-1.0
Archived
2026-10-16T05:44:24.722Z
SHA-256 of the archived text
9c78311ef2d1376722e3eeabc92e6eaeaf5a85a6f192a2cf94d1e11bef5e793e
Stable URL of this page
https://metraipolis.com/documento/dc_8136556752f8

How to cite

MUtE: A Dual Framework for Concept Erasure and Counterfactual Interventions — arXiv:2609.11253 — arXiv (en), https://arxiv.org/abs/2609.11253 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8136556752f8 (Metraipolis archive dc_8136556752f8, sha256 9c78311ef2d1376722e3eeabc92e6eaeaf5a85a6f192a2cf94d1e11bef5e793e)
BibTeX
@misc{dc8136556752f8,
  title = {MUtE: A Dual Framework for Concept Erasure and Counterfactual Interventions — arXiv:2609.11253},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.11253},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8136556752f8; sha256 9c78311ef2d1376722e3eeabc92e6eaeaf5a85a6f192a2cf94d1e11bef5e793e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.11253. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Erasing concept-specific information from representations has been proven useful for mitigating bias or interpreting model decisions. The joint objective is to transform the original representations such that the target concept becomes unpredictable, while maximally preserving concept-unrelated information. In this work, we revisit the optimal bounds of concept erasure to derive a novel class of erasure functions that naturally induce a deterministic, dual counterfactual mapping. Bridging the gap between theoretical optimality and practical representation learning, we design an implementation that imposes a translational bias on counterfactual trajectories - a constraint that aligns with how many concepts geometrically manifest in modern language models. Our framework enables seamless navigation between concept erasure and counterfactual generation. We empirically demonstrate its efficacy in improving downstream algorithmic fairness and generating counterfactual texts.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-11 in cs.CL).
Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.11253 — 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 2026-10-16T05:23:57.043Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).