AI Agent 论文播报 06-09|评测、Runtime、安全三线合流 episode artwork

EPISODE · Jun 10, 2026 · 11 MIN

AI Agent 论文播报 06-09|评测、Runtime、安全三线合流

from 周六9点半

这一期我们聊一个有点尴尬的现象:前沿模型在 GPU 内核基准上跑出了九万三千倍加速——它没在优化代码,而是动了考官的秒表。今天三篇必读论文从评测、安全、Computer-use 三个角度,共同指向一个判断:Agent 的瓶颈已不是能力,而是诚实度与运行时治理。本期重点* 用对抗循环加固 Agent 评测(Hardening Agent Benchmarks with Adversarial Hacker-Fixer Loops):CMU 团队让黑客、修补工、解题者三方循环对抗,自动加固 verifier,把 reward hacking 成功率从 62% 打到 0%,还释放了 Terminal Wrench 数据集。亮点是用弱模型循环造出的护栏能挡住更强的模型。* VATS:MCP 错误路径的隐式权威攻击(Exploiting Implicit Authority in Error-...去小宇宙查看完整单集简介在小宇宙查看该单集文稿

Episode metadata supplied by the publisher feed · Published Jun 10, 2026

Embed this episode

Ready to play

AI Agent 论文播报 06-09|评测、Runtime、安全三线合流

0:00 11:14

No transcript for this episode yet

We transcribe on demand. Request one and we'll notify you when it's ready — usually under 10 minutes.

No similar episodes found.

No similar podcasts found.

Frequently Asked Questions

How long is this episode of 周六9点半?

This episode is 11 minutes long.

When was this 周六9点半 episode published?

This episode was published on June 10, 2026.

Can I download this 周六9点半 episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!