# The Impact of Editorial Intervention on Detecting Native Language Traces

arXiv: https://arxiv.org/abs/2605.10216 · HTML: https://mlsanity.com/paper/2605.10216/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 41 references that are in the map.
- First posted on arXiv 11 May 2026.

## Builds on (most cited first, 10 of 10)

- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift](https://mlsanity.com/paper/1906.02530/) (2019)
- [Authorship Attribution in the Era of LLMs: Problems, Methodologies, and Challenges](https://mlsanity.com/paper/2408.08946/) (2024)
- [Unveiling Attractor Cycles in Large Language Models: A Dynamical Systems View of Successive Paraphrasing](https://mlsanity.com/paper/2502.15208/) (2025)
- [Catch Me If You Can? Not Yet: LLMs Still Struggle to Imitate the Implicit Writing Styles of Everyday Authors](https://mlsanity.com/paper/2509.14543/) (2025)
- [Adapting LLMs for Minimal-edit Grammatical Error Correction](https://mlsanity.com/paper/2506.13148/) (2025)
- [Machine Translation Models are Zero-Shot Detectors of Translation Direction](https://mlsanity.com/paper/2401.06769/) (2024)
- [Introducing OmniGEC: A Silver Multilingual Dataset for Grammatical Error Correction](https://mlsanity.com/paper/2509.14504/) (2025)
- [Robust Native Language Identification through Agentic Decomposition](https://mlsanity.com/paper/2509.16666/) (2025)
- [Neurocomputational Mechanisms of Syntactic Transfer in Bilingual Sentence Production](https://mlsanity.com/paper/2601.18056/) (2026)

## Abstract (arXiv)

Native Language Identification (NLI) is the task of determining an author's native language (L1) from their non-native writing. With the advent of human-AI co-authorship, learner texts are routinely corrected and rewritten by large language models, fundamentally altering the linguistic features NLI approaches depend on. In this paper, we investigate the robustness of L1 traces across increasing degrees of editorial intervention. By processing 450 essays from the Write & Improve 2024 (W&I) corpus through varying levels of grammatical error correction and paraphrasing, we demonstrate that L1 attribution does not depend solely on surface-level errors. Instead, the detection models appear to leverage deeper L1-related features, including unidiomatic lexico-semantic choices and pragmatic transfer. We find that minimal edits preserve these structural traces and maintain high L1 attribution accuracy. In contrast, fluency edits and paraphrasing normalize these L1 features, leading to a sharp decline in performance.