Metraipolis › archive › dc_0d4dd89bf4fe

The Latent Diagnostic Taxonomy: A Framework for Constructing Classifiers and Diagnosing Their Decisions, Applied to Prompt Injection Detection — arXiv:2608.26423

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.26423
License (as archived)
CC0-1.0
Archived
2026-10-17T14:25:05.372Z
SHA-256 of the archived text
dd3601f8cf7a4ac31a21c26c08b6ce3e8b8153ed14b77f063d6b426e42b15b5a
Stable URL of this page
https://metraipolis.com/documento/dc_0d4dd89bf4fe

How to cite

The Latent Diagnostic Taxonomy: A Framework for Constructing Classifiers and Diagnosing Their Decisions, Applied to Prompt Injection Detection — arXiv:2608.26423 — arXiv (en), https://arxiv.org/abs/2608.26423 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0d4dd89bf4fe (Metraipolis archive dc_0d4dd89bf4fe, sha256 dd3601f8cf7a4ac31a21c26c08b6ce3e8b8153ed14b77f063d6b426e42b15b5a)
BibTeX
@misc{dc0d4dd89bf4fe,
  title = {The Latent Diagnostic Taxonomy: A Framework for Constructing Classifiers and Diagnosing Their Decisions, Applied to Prompt Injection Detection — arXiv:2608.26423},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.26423},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0d4dd89bf4fe; sha256 dd3601f8cf7a4ac31a21c26c08b6ce3e8b8153ed14b77f063d6b426e42b15b5a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26423. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

This paper proposes a framework for constructing a classifier as a safeguard layer, and for developing a complementary diagnostic that identifies which of the classifier's confident decisions can be trusted. This framework, the Latent Diagnostic Taxonomy, consists of (i) constructing a dimensionality-optimized classifier, in which the embedding dimensionality is empirically selected via cross-validated performance rather than fixed a priori, (ii) locating a relatively small set of latent support vectors (~ 29% of total training examples) representing influential prompts for identifying tokens that alter the classifier's predicted labels, and (iii) utilizing such tokens and their associated attack magnitudes for constructing a diagnostic taxonomy. This diagnostic taxonomy provides an end-to-end guideline for flagging prompts that require different treatments: rely Safely on the classifier's decision; flag Heuristic Bias and Heuristic Override cases; route Insufficient Context cases for further human/safety review. Applying the framework to a classifier trained on a public prompt injection dataset, we find that a substantial fraction of its confident decisions (~ 77%) are not robust to removing a single token, and that this brittleness separates into two distinct failure patterns: a confidence calibration failure and a genuinely exploitable shortcut. For each zone of the taxonomy, we also recommend strategies for remediating diagnosed prompts. We illustrate the framework as a series of steps, demonstrating how each step operates.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2608.26423 — 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 2026-10-17T13:16:53.639Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).