按年被提到
20252026 · 2 次2026
AI 安全、伦理与社会影响›模型窃取、记忆与水印·arXiv:2512.17075 · cs.CL

Perturb Your Data: Paraphrase-Guided Training Data Watermarking

Pranav Shetty, Mirazul Haque, Petr Babkin, Zhiqiang Ma, Xiaomo Liu, Manuela Veloso · 2025
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2026-01-20 · 最近 2026-11-02
2不同的人
2活跃月份
2讨论串
1被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Training data detection is critical for enforcing copyright and data licensing, as Large Language Models (LLM) are trained on massive text corpora scraped from the internet. We present SPECTRA, a watermarking approach that makes training data reliably detectable even when it comprises less than 0.001% of the training corpus. SPECTRA works by paraphrasing text using an LLM and assigning a score based on how likely each paraphrase is, according to a separate scoring model. A paraphrase is chosen so that its score closely matches that of the original text, to avoid introducing any distribution shifts. To test whether a suspect model has been trained on the watermarked data, we compare its token probabilities against those of the scoring model. We demonstrate that SPECTRA achieves a consistent p-value gap of over nine orders of magnitude when detecting data used for training versus data not used for training, which is greater than all baselines tested. SPECTRA equips data owners with a scalable, deploy-before-release watermark that survives even large-scale LLM training.

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throwaway219450 · 2026-11

“In a self-supervised setting you don't need ground truth data, as long as you have a proxy task that you can train on which forces the model to learn something useful. Like cutting out bits of images and telling the model to reconstruct the gaps (i.e. MAEs). There are also techniques to watermark model weights to confirm what training data were used: htt…”

Astra can read and write weights to improve Qwen原评论 ↗
被提交为 story · 1 次