Metraipolis › archive › dc_86aae7d89c07
Not X, but When: Diagnosing a Mandarin LLM Writing Tic through Contextual Appropriateness — arXiv:2610.22249
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.22249
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T10:02:02.967Z
- SHA-256 of the archived text
3a8098adeb6e4324b4a887670d4e92ef2d5c5291fc295bc582c6d8bb6822c91c- Stable URL of this page
https://metraipolis.com/documento/dc_86aae7d89c07
How to cite
Not X, but When: Diagnosing a Mandarin LLM Writing Tic through Contextual Appropriateness — arXiv:2610.22249 — arXiv (en), https://arxiv.org/abs/2610.22249 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_86aae7d89c07 (Metraipolis archive dc_86aae7d89c07, sha256 3a8098adeb6e4324b4a887670d4e92ef2d5c5291fc295bc582c6d8bb6822c91c)
BibTeX
@misc{dc86aae7d89c07,
title = {Not X, but When: Diagnosing a Mandarin LLM Writing Tic through Contextual Appropriateness — arXiv:2610.22249},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.22249},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_86aae7d89c07; sha256 3a8098adeb6e4324b4a887670d4e92ef2d5c5291fc295bc582c6d8bb6822c91c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22249. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Recurring expressions in model-written text are often treated as signatures of machine authorship, yet most of them serve legitimate functions. Whether such an expression is a tic depends on whether the model knows when it is called for. This judgment of appropriateness, central to communicative competence, remains largely unexamined for language models. We study it through the Mandarin corrective construction bushi X, ershi Y ("not X, but Y"). Drawing on Gricean pragmatics and corrective negation, we introduce a decision tree that classifies uses as substantive, optional, or defective by testing whether the context makes X a live alternative and whether rejecting X is informative. Five commercial models use the construction 2.5 to 9.2 times as densely as a human reference. Of 307 annotated uses, only 1.30% are defective while 52.8% are optional, that is, correct but unneeded, so models fall short in appropriateness far more often than in correctness. Several reward models favor the construction even over quality-matched variants, an LLM judge favors it even where the context does not license it, and both supervised fine-tuning and preference optimization tend to raise its frequency in open-weight models. We hope this work encourages the development of models attentive to the appropriateness of their expressions.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.22249 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T08:44:47.704Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).