Metraipolis › archive › dc_78b4fa845bac

Can the Confidence and Decisions of System 1 Models Be Trusted? An Explainability and Perturbation Audit of Jev-Style Decision Models — arXiv:2611.12521

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.12521
License (as archived)
CC0-1.0
Archived
2026-11-10T18:45:11.806Z
SHA-256 of the archived text
fd92c6ff6594630e20ff351ddfe4c3d36bd4c6dc0675bd63f65bd44e79b8e922
Stable URL of this page
https://metraipolis.com/documento/dc_78b4fa845bac

How to cite

Can the Confidence and Decisions of System 1 Models Be Trusted? An Explainability and Perturbation Audit of Jev-Style Decision Models — arXiv:2611.12521 — arXiv (en), https://arxiv.org/abs/2611.12521 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_78b4fa845bac (Metraipolis archive dc_78b4fa845bac, sha256 fd92c6ff6594630e20ff351ddfe4c3d36bd4c6dc0675bd63f65bd44e79b8e922)
BibTeX
@misc{dc78b4fa845bac,
  title = {Can the Confidence and Decisions of System 1 Models Be Trusted? An Explainability and Perturbation Audit of Jev-Style Decision Models — arXiv:2611.12521},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.12521},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_78b4fa845bac; sha256 fd92c6ff6594630e20ff351ddfe4c3d36bd4c6dc0675bd63f65bd44e79b8e922},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.12521. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

"System 1" decision models (pioneered by TypeSafe AI's Jev) replace autoregressive text generation with a single forward pass returning typed probability distributions over declared schemas (Choice, Score, Noul). While deployed for low-latency routing and automated triage, their statistical calibration and post-hoc explainability remain uncharacterized. We introduce an open, deterministic audit benchmark evaluating System 1 decision models across calibration under label-preserving perturbations, cross-question interference, and post-hoc attribution faithfulness. Evaluating five open-weight models (0.8B to 27B) alongside generative System 2 baselines -- both single-token logprob scoring and structured JSON/CoT decoding -- across NLI (e-SNLI), adversarial syntactic shortcuts (HANS), synthetic clinical triage (N=900), prompt injection defense (Tensor Trust, N=250), and an 86-case logic-gate trigger suite yields five findings: (1) provider confidence is an exact closed-form dispersion transform (1 - H(p)/ln K on sub-10B models; normalized margin on Clef-27B) rather than empirical accuracy; (2) bundling multiple questions in one request induces directional cross-question interference (11.7% to 22.4% binary flips on Tev1-0.8B, vs. 0.1% to 0.2% on Tev1-4B); (3) System 1 models with >= 4B parameters are faster and better calibrated under syntactic shifts (ECE_w = 0.030 to 0.087) than logprob (0.144) and JSON (0.378) baselines; (4) for prompt-injection defense (N=250), >= 4B models achieve 0.988 to 0.998 AUROC with up to 95.2% recall and 0.0% false alarms at 48 to 60 ms; and (5) single-pass inference makes 512-coalition XAI interactive (1.3 to 3.2 s/input), while KernelSHAP with a logit link (S1-SHAP) mitigates softmax saturation on disjunctive OR gates (89.3% to 100.0% gate recovery vs. 57.1% to 82.1% in prob space; p_adj = 0.004).

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2611.12521 — 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 2026-11-10T18:37:52.485Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).