按年被提到
20232023 · 2 次2026
AI 安全、伦理与社会影响›模型窃取、记忆与水印·arXiv:2302.10149 · cs.CR

Poisoning Web-Scale Training Datasets is Practical

Nicholas Carlini, Matthew Jagielski, Christopher A. Choquette-Choo, Daniel Paleka, Will Pearce, Hyrum Anderson, Andreas Terzis, Kurt Thomas, Florian Tramèr · 2023
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 3 人首次 2023-02-21 · 最近 2025-03-21
3不同的人
2活跃月份
3讨论串
166被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Deep learning models are often trained on distributed, web-scale datasets crawled from the internet. In this paper, we introduce two new dataset poisoning attacks that intentionally introduce malicious examples to a model's performance. Our attacks are immediately practical and could, today, poison 10 popular datasets. Our first attack, split-view poisoning, exploits the mutable nature of internet content to ensure a dataset annotator's initial view of the dataset differs from the view downloaded by subsequent clients. By exploiting specific invalid trust assumptions, we show how we could have poisoned 0.01% of the LAION-400M or COYO-700M datasets for just $60 USD. Our second attack, frontrunning poisoning, targets web-scale datasets that periodically snapshot crowd-sourced content -- such as Wikipedia -- where an attacker only needs a time-limited window to inject malicious examples. In light of both attacks, we notify the maintainers of each affected dataset and recommended several low-overhead defenses.

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Terr_ · 2025-03

“> LLMs are already trained on a lot of dubious sources (for example Quora and Reddit) There's a very big difference between crawling documents where people happen to say something wrong (e.g. "Tortoises are amphibians") versus the impact of widespread poisoning [0] or indirect prompt injection. [1] [0] https://arxiv.org/html/2302.10149v2 [1] https://ce…”

ChatGPT hit with privacy complaint over defamatory hallucinations原评论 ↗
被提交为 story · 2 次