The Impact of Editorial Intervention on Detecting Native Language Traces
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909031384219648 |
|---|---|
| author | Uluslu, Ahmet Yavuz Gales, Mark Knill, Kate Schneider, Gerold |
| author_facet | Uluslu, Ahmet Yavuz Gales, Mark Knill, Kate Schneider, Gerold |
| contents | Native Language Identification (NLI) is the task of determining an author's native language (L1) from their non-native writings. With the advent of human-AI co-authorship, non-native texts are routinely corrected and rewritten by large language models, fundamentally altering the linguistic features NLI models 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 corpus through varying levels of grammatical error correction (GEC) and paraphrasing, we demonstrate that L1 attribution does not entirely depend on surface-level errors. Instead, the detection models leverage deeper L1 features: unidiomatic lexico-semantic choices, pragmatic transfer, and the author's underlying cultural perspective. We find that minimal edits preserve these structural traces and maintain high profiling accuracy. In contrast, fluency edits and paraphrasing normalize these L1 features, leading to a severe degradation in performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_10216 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | The Impact of Editorial Intervention on Detecting Native Language Traces Uluslu, Ahmet Yavuz Gales, Mark Knill, Kate Schneider, Gerold Computation and Language Native Language Identification (NLI) is the task of determining an author's native language (L1) from their non-native writings. With the advent of human-AI co-authorship, non-native texts are routinely corrected and rewritten by large language models, fundamentally altering the linguistic features NLI models 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 corpus through varying levels of grammatical error correction (GEC) and paraphrasing, we demonstrate that L1 attribution does not entirely depend on surface-level errors. Instead, the detection models leverage deeper L1 features: unidiomatic lexico-semantic choices, pragmatic transfer, and the author's underlying cultural perspective. We find that minimal edits preserve these structural traces and maintain high profiling accuracy. In contrast, fluency edits and paraphrasing normalize these L1 features, leading to a severe degradation in performance. |
| title | The Impact of Editorial Intervention on Detecting Native Language Traces |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2605.10216 |