# Beating the Style Detector: Three Hours of Agentic Research on the AI-Text Arms Race

arXiv: https://arxiv.org/abs/2605.02620 · HTML: https://mlsanity.com/paper/2605.02620/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use.
- Builds on 4 of its 12 references that are in the map.
- First posted on arXiv 4 May 2026.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 4 of 4)

- [A Gentle Introduction to Deep Learning in Medical Image Processing](https://mlsanity.com/paper/1810.05401/) (2018)
- [Known Operator Learning and Hybrid Machine Learning in Medical Imaging -- A Review of the Past, the Present, and the Future](https://mlsanity.com/paper/2108.04543/) (2021)
- [Precision Learning: Towards Use of Known Operators in Neural Networks](https://mlsanity.com/paper/1712.00374/) (2017)
- [Can You Make It Sound Like You? Post-Editing LLM-Generated Text for Personal Style](https://mlsanity.com/paper/2604.24444/) (2026)

## Built on by (most cited first, 1 of 1)

- [Just Ask for a Table: A Thirty-Token User Prompt Defeats Sponsored Recommendations in Twelve LLMs](https://mlsanity.com/paper/2605.12772/) (2026)

## Abstract (arXiv)

Reproducing an empirical NLP study used to take weeks. Given the released data and a modern agentic-research harness, we redo every experiment of a recent ACL,2026 study on personal-style post-editing of LLM drafts -- and add three new ones -- with the human investigator acting only as a reviewer-in-the-loop. We reproduce all seven preregistered hypotheses and recover the paper's headline correlation between perceived self-similarity and embedding-measured self-similarity to three decimal places (r=+0.244, p<10-8, n=648). Under a leakage-free held-out protocol, GPT-5.5 and Claude,Opus,4.7 close 71--75,% of the style gap to the same-author ceiling on 324 paired tasks, against 24,% for the human post-edit, and beat the human post-edit on ~80,% of tasks. We then frame the same data as an AI-text detection arms race. A leave-authors-out linear SVM on LUAR-MUD embeddings reaches AUC 0.93--1.00 across approaches; six diagnostics show that GPT-5.5 detection is mostly a length confound while Opus detection is a genuine stylistic signature. Given T=20 feedback iterations against the frozen detector, an Opus agent flips two of five held-out test mimics to the human half-space and shrinks every margin by an order of magnitude. With moderate effort against a known detector, a frontier LLM can already efficiently lower its own AI-detection probability. All code, 648 mimic drafts, trained detectors, diagnostics, and adversarial trajectories are released.