新闻
Auditing black-box clinical prediction models with a large language model through queries alone: a blinded test with planted flaws
arXiv cs.AI · 发布于 · 阅读约3分钟
30秒读懂
- 发生了什么
- Researchers tested whether GPT-6 Astra could audit black-box clinical prediction models by querying them, finding planted flaws in 96 percent of cases.
- 为何重要
- Matters for engineers building clinical decision support systems who need to verify model behavior without access to internal weights or architecture.
- 注意
- LLM reports of problems require verification against actual model behavior; discovered relationships held only 11 percent of the time on new data, not 33 percent.
- llm
- language model
- edge
这条新闻背后的模式
每个模式都讲清楚技术如何运作、何时值得投入,以及在哪里会失效。
The Agent Architect
每周一个模式、一个权衡、一个生产事故案例。为构建智能体系统的人准备的每周简报。
每周一封邮件,一键退订。您的地址仅用于发送简报。