aifollow.news
Back Rohan Paul
Rohan Paul· @rohanpaul_ai · X· · Original publication time AI score54

OpenAI releases LifeVal to test AI on everyday errandsMachine translation

AI introduction

LifeVal covers tasks in caregiving, health, education, finance, shopping and household administration. OpenAI reports a score of 80.6% for GPT-6 Astra on everyday tasks, compared with 49.3% for GPT-5.2 less than nine months earlier. These are benchmark scores, not a measure of real-world task completion.

Original post · AI translation

@rohanpaul_ai
Translation · refer to the original

OpenAI 发布了 LifeVal,这是一项测试 AI 模型能否替人完成日常事务的基准测试。

GPT-6 Astra 在日常任务上的得分为 80.6%,高于不到 9 个月前 GPT-5.2 的 49.3%。

LifeVal 的任务涵盖照护、健康、教育、财务、购物和家庭事务管理,从预约车辆检验到在预算内订购清真食品。

QuoteOpenAI@OpenAI
Translation
要让 AGI 惠及全人类,它就应该帮助人们处理对他们最重要的日常事务。因此,我们开发了 LifeVal 这一新的基准测试,其设计参考了数千名 ChatGPT 用户的反馈,了解他们最希望个人助理在哪些方面提供帮助——从寻找儿童照护服务,到获取政府服务,再到管理个人财务。 过去 1 年里,我们一直使用 LifeVal 来衡量并改进我们的模型处理这些日常任务的能力。 https://t.co/GlUjixzEM0
Original

For AGI to benefit all of humanity, it should help people with the everyday tasks that matter most to them. So we built LifeVal, a new benchmark shaped by feedback from thousands of ChatGPT users on where they most want help from a personal assistant—from finding childcare to accessing government services to managing personal finances. Over the past year, we’ve used LifeVal to measure and improve how well our models handle these everyday tasks. https://t.co/GlUjixzEM0

View this post on X ↗

Source: Rohan Paul · x.com

Research
Found an error? Send a correction