按年被提到
20262026 · 2 次2026
大模型与语言模型›大模型文本、写作与版权·arXiv:2609.15369 · cs.CL

SlopShape: Identifying AI-Generated Commercial Web Content

Jochen Madler · 2026
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2026-09-22 · 最近 2026-10-14
2不同的人
2活跃月份
2讨论串
0被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Word-level detectors identify unedited AI-generated text almost perfectly, but the literature documents their brittleness under rewording, and a word-level score neither characterizes a text nor identifies which AI model wrote it. We ask whether AI-generated text can be identified one level deeper, from structural signatures: how information is presented, in what order, with what evidence, and in what voice. We replicate StoryScope (Russell et al., 2026), which showed such patterns for AI-generated fiction, on commercial content: 2,250 pre-ChatGPT human blog posts from 268 company domains against 11,250 AI mirrors from five frontier models. A 203-feature instrument, applied by an LLM and validated in a human gold-annotation session (human-human kappa 0.939, human-model 0.951), detects AI posts from its 176 structural features alone at 97.0 macro-F1 on held-out companies, nearly unchanged (96.1) when every AI post is reworded by its own model. The signal characterizes and attributes: AI posts share a tidy, self-announcing shape, 68.6% are attributed to the correct source against a 16.7% chance rate, and human posts occupy rare structural configurations. All effects replicate StoryScope's, consistent in direction and at least as large in magnitude. We release pipeline, instrument, prompts, code, and aggregate artifacts.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论corysama · 2026-10

“Supposedly, SlopShape https://arxiv.org/abs/2609.15369 can ID not just human vs bot but which bot via structural patterns in the writing. Next we need a "Verbose useless content" detector and we'll be set up to GAN the whole internet into https://xkcd.com/810/ "Mission Accomplished".”

Has AI tells原评论 ↗
被提交为 story · 1 次