← 研踪 PaperTrailarXiv:2611.08229 · v1
Escalate or Close: How Fast-Tier Models Handle Held-Out Web Attack Families in Two-Tier Triage
首次提交 2026-11-06 00:13
版本更新 2026-11-06 00:13
雷达观测 2026-11-06 12:09
摘要
In two-tier triage a fast model decides most web requests and escalates the rest to a slow tier, so an attack the fast tier closes is never examined. We hold each of seven attack families out of training, recalibration and thresholds, splice the payloads into parameters of real browsing traffic, and compare zero-shot decision models, which return a probability for a typed question, with supervised detectors, a local language model, the OWASP Core Rule Set and an anomaly baseline. Detectors trained on the known families can give much of a held-out family a confident benign verdict: over 10 training runs, ModernBERT gives 37% to 60% of held-out path traversal a probability below 0.05 and escalates only 50% to 81% of it within a 5% benign budget, against nearly all of the same requests when traversal is known. Jev and Clef, two hosted decision models whose budget decisions do not depend on the held-out family, escalate at least 97.6% of every family within the same budget, but their undisclosed training data may contain these public families. Tests on recent exploits need controls for request shape: Jev escalates 37% of benign twins (exploit requests with the attack replaced by legitimate values), and detectors trained on older exploits learn what scanner requests look like. A search guided by Jev's probability gets 43% of held-out attacks past Jev within 300 queries, but Jev's answers vary: asked again, it escalates those requests in 60% of its answers.
作者备注：20 pages, 4 figures, 24 tables
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 · 2026-11-06 00:13 · 最新观测版本
Escalate or Close: How Fast-Tier Models Handle Held-Out Web Attack Families in Two-Tier Triage
In two-tier triage a fast model decides most web requests and escalates the rest to a slow tier, so an attack the fast tier closes is never examined. We hold each of seven attack families out of training, recalibration and thresholds, splice the payloads into parameters of real browsing traffic, and compare zero-shot decision models, which return a probability for a typed question, with supervised detectors, a local language model, the OWASP Core Rule Set and an anomaly baseline. Detectors trained on the known families can give much of a held-out family a confident benign verdict: over 10 training runs, ModernBERT gives 37% to 60% of held-out path traversal a probability below 0.05 and escalates only 50% to 81% of it within a 5% benign budget, against nearly all of the same requests when traversal is known. Jev and Clef, two hosted decision models whose budget decisions do not depend on the held-out family, escalate at least 97.6% of every family within the same budget, but their undisclosed training data may contain these public families. Tests on recent exploits need controls for request shape: Jev escalates 37% of benign twins (exploit requests with the attack replaced by legitimate values), and detectors trained on older exploits learn what scanner requests look like. A search guided by Jev's probability gets 43% of held-out attacks past Jev within 300 queries, but Jev's answers vary: asked again, it escalates those requests in 60% of its answers.
查看此版本解读 →查看此版本原文 ↗