Toward a Geometric Perspective on Trustworthy AI: Diagnostic Signatures for High-Stakes NLU

Eduardo Sanchez-Karhunen, Jose F Quesada, Miguel A. Gutiérrez-Naranjo
Proceedings of Fifth European Conference on Algorithmic Fairness, PMLR 350:24-39, 2026.

Abstract

The deployment of Natural Language Understanding (NLU) systems in high-risk sectors such as clinical triage, emergency response, and financial eligibility, is now governed by the transparency and accountability mandates of the European AI Act. Despite these legal obligations, a significant "Accountability Gap" persists: regulators demand documentation of a system’s "general logic", while the deep learning models powering these systems remain structurally opaque. While existing Explainable AI (XAI) methods provide post-hoc, model-agnostic approximations, they often describe where a model attended rather than how it computed its decision, a distinction that is increasingly consequential in high-stakes settings. This paper proposes that the Geometric Framework for Intent Detection offers a complementary analytical lens for addressing these challenges. By interpreting semantic resolution as a dynamical trajectory through a hidden state manifold converging toward stable attractor basins, this framework makes visible certain structural properties of the model’s computation that are currently invisible to behavioral audits. We argue that this geometric perspective opens a set of promising research directions for each of the five pillars of Trustworthy AI: Transparency, Technical Robustness, Fairness, Human Oversight, and Societal Well-being. Specifically, it suggests candidate diagnostic signatures for failures such as algorithmic bias, adversarial vulnerability, out-ofscope instability, and multilingual degradation. We present this framework as a conceptual toolbox and a research roadmap intended to complement existing auditing practices and contribute toward the goal of verifiable, mechanistically grounded NLU.

Cite this Paper

BibTeX
@InProceedings{pmlr-v350-sanchez-karhunen26a, title = {Toward a Geometric Perspective on Trustworthy AI: Diagnostic Signatures for High-Stakes NLU}, author = {Sanchez-Karhunen, Eduardo and Quesada, Jose F and Guti\'errez-Naranjo, Miguel A.}, booktitle = {Proceedings of Fifth European Conference on Algorithmic Fairness}, pages = {24--39}, year = {2026}, editor = {De Bie, Tijl and Defrance, MaryBeth and Calders, Toon and Nguyen, Dennis}, volume = {350}, series = {Proceedings of Machine Learning Research}, month = {02--04 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v350/main/assets/sanchez-karhunen26a/sanchez-karhunen26a.pdf}, url = {https://proceedings.mlr.press/v350/sanchez-karhunen26a.html}, abstract = {The deployment of Natural Language Understanding (NLU) systems in high-risk sectors such as clinical triage, emergency response, and financial eligibility, is now governed by the transparency and accountability mandates of the European AI Act. Despite these legal obligations, a significant "Accountability Gap" persists: regulators demand documentation of a system’s "general logic", while the deep learning models powering these systems remain structurally opaque. While existing Explainable AI (XAI) methods provide post-hoc, model-agnostic approximations, they often describe where a model attended rather than how it computed its decision, a distinction that is increasingly consequential in high-stakes settings. This paper proposes that the Geometric Framework for Intent Detection offers a complementary analytical lens for addressing these challenges. By interpreting semantic resolution as a dynamical trajectory through a hidden state manifold converging toward stable attractor basins, this framework makes visible certain structural properties of the model’s computation that are currently invisible to behavioral audits. We argue that this geometric perspective opens a set of promising research directions for each of the five pillars of Trustworthy AI: Transparency, Technical Robustness, Fairness, Human Oversight, and Societal Well-being. Specifically, it suggests candidate diagnostic signatures for failures such as algorithmic bias, adversarial vulnerability, out-ofscope instability, and multilingual degradation. We present this framework as a conceptual toolbox and a research roadmap intended to complement existing auditing practices and contribute toward the goal of verifiable, mechanistically grounded NLU.} }
Endnote
%0 Conference Paper %T Toward a Geometric Perspective on Trustworthy AI: Diagnostic Signatures for High-Stakes NLU %A Eduardo Sanchez-Karhunen %A Jose F Quesada %A Miguel A. Gutiérrez-Naranjo %B Proceedings of Fifth European Conference on Algorithmic Fairness %C Proceedings of Machine Learning Research %D 2026 %E Tijl De Bie %E MaryBeth Defrance %E Toon Calders %E Dennis Nguyen %F pmlr-v350-sanchez-karhunen26a %I PMLR %P 24--39 %U https://proceedings.mlr.press/v350/sanchez-karhunen26a.html %V 350 %X The deployment of Natural Language Understanding (NLU) systems in high-risk sectors such as clinical triage, emergency response, and financial eligibility, is now governed by the transparency and accountability mandates of the European AI Act. Despite these legal obligations, a significant "Accountability Gap" persists: regulators demand documentation of a system’s "general logic", while the deep learning models powering these systems remain structurally opaque. While existing Explainable AI (XAI) methods provide post-hoc, model-agnostic approximations, they often describe where a model attended rather than how it computed its decision, a distinction that is increasingly consequential in high-stakes settings. This paper proposes that the Geometric Framework for Intent Detection offers a complementary analytical lens for addressing these challenges. By interpreting semantic resolution as a dynamical trajectory through a hidden state manifold converging toward stable attractor basins, this framework makes visible certain structural properties of the model’s computation that are currently invisible to behavioral audits. We argue that this geometric perspective opens a set of promising research directions for each of the five pillars of Trustworthy AI: Transparency, Technical Robustness, Fairness, Human Oversight, and Societal Well-being. Specifically, it suggests candidate diagnostic signatures for failures such as algorithmic bias, adversarial vulnerability, out-ofscope instability, and multilingual degradation. We present this framework as a conceptual toolbox and a research roadmap intended to complement existing auditing practices and contribute toward the goal of verifiable, mechanistically grounded NLU.
APA
Sanchez-Karhunen, E., Quesada, J.F. & Gutiérrez-Naranjo, M.A.. (2026). Toward a Geometric Perspective on Trustworthy AI: Diagnostic Signatures for High-Stakes NLU. Proceedings of Fifth European Conference on Algorithmic Fairness, in Proceedings of Machine Learning Research 350:24-39 Available from https://proceedings.mlr.press/v350/sanchez-karhunen26a.html.

Related Material