Attune, Don’t Prune: A Conformal Framework for Fact Preservation in News Content Attunement

Alice Evelyn Ashby, Mai Nguyen, Zhiyuan Luo, Khuong An Nguyen
Proceedings of the Fifteenth Symposium on Conformal and Probabilistic Prediction with Applications, PMLR 329:1084-1086, 2026.

Abstract

We propose a news content attunement system that rewrites articles according to a reader-defined graphic sensitivity setting, or abstains if rewriting is not possible. Conformal importance selection provides a finite-sample article-level recall guarantee for important source sentences, while conformal risk control calibrates a bounded aggregate attunement loss balancing preservation, residual graphic content, and utility. On a held-out dataset, we preserve 86% of essential facts whilst neutralising $\approx$97% of graphic content.

Cite this Paper

BibTeX
@InProceedings{pmlr-v329-ashby26a, title = {Attune, Don’t Prune: A Conformal Framework for Fact Preservation in News Content Attunement}, author = {Ashby, Alice Evelyn and Nguyen, Mai and Luo, Zhiyuan and Nguyen, Khuong An}, booktitle = {Proceedings of the Fifteenth Symposium on Conformal and Probabilistic Prediction with Applications}, pages = {1084--1086}, year = {2026}, editor = {Ahlberg, Ernst and Johansson, Ulf and Boström, Henrik and Carlevaro, Alberto and Hallberg Szabadváry, Johan and Carlsson, Lars}, volume = {329}, series = {Proceedings of Machine Learning Research}, month = {02--04 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v329/main/assets/ashby26a/ashby26a.pdf}, url = {https://proceedings.mlr.press/v329/ashby26a.html}, abstract = {We propose a news content attunement system that rewrites articles according to a reader-defined graphic sensitivity setting, or abstains if rewriting is not possible. Conformal importance selection provides a finite-sample article-level recall guarantee for important source sentences, while conformal risk control calibrates a bounded aggregate attunement loss balancing preservation, residual graphic content, and utility. On a held-out dataset, we preserve 86% of essential facts whilst neutralising $\approx$97% of graphic content.} }
Endnote
%0 Conference Paper %T Attune, Don’t Prune: A Conformal Framework for Fact Preservation in News Content Attunement %A Alice Evelyn Ashby %A Mai Nguyen %A Zhiyuan Luo %A Khuong An Nguyen %B Proceedings of the Fifteenth Symposium on Conformal and Probabilistic Prediction with Applications %C Proceedings of Machine Learning Research %D 2026 %E Ernst Ahlberg %E Ulf Johansson %E Henrik Boström %E Alberto Carlevaro %E Johan Hallberg Szabadváry %E Lars Carlsson %F pmlr-v329-ashby26a %I PMLR %P 1084--1086 %U https://proceedings.mlr.press/v329/ashby26a.html %V 329 %X We propose a news content attunement system that rewrites articles according to a reader-defined graphic sensitivity setting, or abstains if rewriting is not possible. Conformal importance selection provides a finite-sample article-level recall guarantee for important source sentences, while conformal risk control calibrates a bounded aggregate attunement loss balancing preservation, residual graphic content, and utility. On a held-out dataset, we preserve 86% of essential facts whilst neutralising $\approx$97% of graphic content.
APA
Ashby, A.E., Nguyen, M., Luo, Z. & Nguyen, K.A.. (2026). Attune, Don’t Prune: A Conformal Framework for Fact Preservation in News Content Attunement. Proceedings of the Fifteenth Symposium on Conformal and Probabilistic Prediction with Applications, in Proceedings of Machine Learning Research 329:1084-1086 Available from https://proceedings.mlr.press/v329/ashby26a.html.

Related Material