Metraipolis › archive › dc_04d2f8022cd6
A Layer-wise Study of How Style Rewriting Breaks the Calibration of Zero-Shot LLM Fake-News Detectors — arXiv:2611.11239
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.11239
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-10T18:43:14.218Z
- SHA-256 of the archived text
bfca3ead4b558e16898e12c9bb058ec2b0b2e076a6164f9ee9150a0b71d1c17b- Stable URL of this page
https://metraipolis.com/documento/dc_04d2f8022cd6
How to cite
A Layer-wise Study of How Style Rewriting Breaks the Calibration of Zero-Shot LLM Fake-News Detectors — arXiv:2611.11239 — arXiv (en), https://arxiv.org/abs/2611.11239 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_04d2f8022cd6 (Metraipolis archive dc_04d2f8022cd6, sha256 bfca3ead4b558e16898e12c9bb058ec2b0b2e076a6164f9ee9150a0b71d1c17b)
BibTeX
@misc{dc04d2f8022cd6,
title = {A Layer-wise Study of How Style Rewriting Breaks the Calibration of Zero-Shot LLM Fake-News Detectors — arXiv:2611.11239},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.11239},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_04d2f8022cd6; sha256 bfca3ead4b558e16898e12c9bb058ec2b0b2e076a6164f9ee9150a0b71d1c17b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.11239. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) are increasingly asked to judge whether a news article is true, and their stated confidence is often taken at face value. We test what happens to that confidence when an article's writing style, but not its facts, is manipulated, as in recent \emph{style attacks} on fake-news detectors. Tracing how the prediction and its confidence form layer by layer inside the network, we study three model families (LLaMA-3-8B, Qwen2.5-7B, Mistral-7B) on three fake-news benchmarks, and find the same three patterns in every model--dataset pairing. First, on the original articles the models are competent and largely honest judges: they settle on an answer, briefly become overconfident, and then correct their confidence in the final layers. Second, style-rewritten articles break this self-correction: the models become far less accurate, under the strongest attack, worse than random guessing, yet remain just as confident, so they are \emph{confidently wrong} precisely when a warning is most needed. Third, nearly all of the damage comes from rewriting into sensational or emotionally charged styles, while sober, neutral rewrites leave the models' honesty largely intact. The practical implication is that an LLM's confidence about veracity cannot be trusted as a safety signal when an article's style may have been manipulated.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.11239 — 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 2026-11-10T18:37:37.632Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).