创建

登录 ReadmeX

或

了解我们如何处理你的信息: 隐私政策

ReadmeX

陪你读懂 AI 动态

今天,想先了解什么?

选一份动态总结,或直接聊聊你的问题。

社区和关注动态总结只属于你自己。

资讯

工程师展示基于 Jev 的 system-one 应用

AI 总结

在 AI Engineer Code Summit 上,Glean、UiPath 和 Workday 的工程师展示了使用 Jev 和其他 system-one 模型构建的应用与原型。据 Latent Space 报道,这类非生成式模型返回结构化选择、评分或概率,通常比推理 LLM 更快、更便宜。展示案例包括 Glean 将 Jev 与 LLM 和其他分类器结合使用、UiPath 通过多轮“锦标赛”测试文档排序,以及 Workday 原型化一个在用户输入时自动填写休假申请单的界面。TypeSafe AI CEO Diogo Almeida 主张将 system-one 模型直接集成到代码中,并称 Jev 是“可靠智能”,但报道未独立验证这一厂商说法。

深度见解

这些展示呈现了一种不同的 AI 集成方式:工程师正将结构化的非生成式模型与 LLM 配合使用,以完成文档排序和界面操作等具体应用决策。

Today at the AI Engineer Code Summit, the main theme was system one models — non-generative models that return structured choices, scores, or probabilities. They’re also typically much faster and cheaper than reasoning LLMs (a.k.a. system two models). 

And who better to give an inspirational rallying cry for system one models than Diogo Almeida, co-founder and CEO of TypeSafe AI. Almeida was lead creator of Jev, the product that kicked off this trend less than two months ago.

Jev was an instant revelation for AI engineers when it was released in September. In a podcast with Latent Space on launch (which quickly became our most popular episode of the year), Almeida explained the core problem with making LLMs the heart of software: reliability. 

In his keynote at AIE Code today, he took it a step further by claiming that Jev is “reliable intelligence” that should be used for “just automating everything.” 

#Jev#TypeSafe AI#Glean#UiPath#Workday#AI Engineer Code Summit

原文约 4 分钟读完阅读原文
群聊

还没有评论,来说说你的看法。