Agents Can Use Base Models to Evade AI Detection
arXiv:2609.31876v1 Announce Type: new Abstract: We show that coding agents equipped with a base language model can successfully assemble responses from its samples to evade detection. Base models have been shown to evade commercial detectors, however, prior "humanization" techniques rely on using t…
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-09-29 04:00 · arXiv cs.CL
Agents Can Use Base Models to Evade AI Detection