跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.25566 · v1

Diagnosis is not a policy: a state-transition audit of caoability-conditioned test-time inference

首次提交 2026-11-19 14:09
版本更新 2026-11-19 14:09
雷达观测 2026-11-20 19:58
arXiv ↗PDF ↗

摘要

Test-time actions can satisfy retrieval, state-realization, program-consistency, or rendering objectives without correcting an answer. Test-time Error Response and Realizability Audit (TERRA) is an evaluation/auditing protocol, not a new inference algorithm. It uses a unified decomposition of intervention outcomes into applicability, target-state realization, endpoint translation, harm, and resource/inference cost. Paired comparisons fix the incoming state and evaluation conditions; cross-model analyses vary the receiving configuration. Executable financial reasoning experiments yield three findings: the same context expansion can harm one configuration and benefit another; sufficient evidence does not ensure correctness because configurations differ in evidence-to-answer translation; and diagnostic headroom---accuracy attainable with idealized diagnostic information over the measured baseline---plus local consistency does not ensure practical gains. Action value is V(a∣x,m)V(a\mid x,m), not diagnosis alone; diagnosis is not a deployable policy π(a∣x,m)π(a\mid x,m).

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-19 14:09 · 最新观测版本

Diagnosis is not a policy: a state-transition audit of caoability-conditioned test-time inference

Test-time actions can satisfy retrieval, state-realization, program-consistency, or rendering objectives without correcting an answer. Test-time Error Response and Realizability Audit (TERRA) is an evaluation/auditing protocol, not a new inference algorithm. It uses a unified decomposition of intervention outcomes into applicability, target-state realization, endpoint translation, harm, and resource/inference cost. Paired comparisons fix the incoming state and evaluation conditions; cross-model analyses vary the receiving configuration. Executable financial reasoning experiments yield three findings: the same context expansion can harm one configuration and benefit another; sufficient evidence does not ensure correctness because configurations differ in evidence-to-answer translation; and diagnostic headroom---accuracy attainable with idealized diagnostic information over the measured baseline---plus local consistency does not ensure practical gains. Action value is V(a∣x,m)V(a\mid x,m), not diagnosis alone; diagnosis is not a deployable policy π(a∣x,m)π(a\mid x,m).

查看此版本解读 →查看此版本原文 ↗