Skip to content

“我看到了 Scaling Law 的信号” | 对谈清华叉院助理教授徐梦迪:具身智能、世界模型、真正的泛化

September 21, 2026 · 1 h 21 min

Show notes

➤ 本周「十字路口」的嘉宾是清华叉院助理教授徐梦迪。从清华车辆工程出发,途经 Johns Hopkins、CMU,再到 Stanford 吴佳俊、李飞飞组做博士后,2024 年初她选择回国,加入清华交叉信息研究院。她的核心命题是机器人的 in-context learning (ICL):让机器人到一个新环境,通过一两次交互当场学会新任务,而且越学越快。
本期节目还有一个及时的注脚:就在我们录完节目后的第二周,Generalist 发布 GEN-1.5 ——只给机器人看一段 3-12 秒的动作示范作为 "physical prompt",不做微调、不更新参数,它就能当场尝试完成新任务(10 项任务平均成功率约 59%)。ICL 随即成为全球具身智能领域最受关注的话题之一,而这正是梦迪持续押注的方向。
除了 ICL,我们还聊到:
* 预训练的天花板
* 真正的泛化
* Scaling Law...去小宇宙查看完整单集简介
在小宇宙查看该单集文稿

All episodes of 十字路口Crossing

Settings

Display

Language

Decides which stations you see first on the home page.

History

Recently played stations and episodes and your latest searches are only stored in this browser.

Notifications

A short notice when a podcast you follow has a new episode. Turn off single podcasts with the bell on their page.

Account

Save favorites – in the app and on the website.

Keyboard shortcuts

/
Open search
Space
Play or pause
← →
In the player: back 15 s, forward 30 s (episodes) · in a row: next tile
↑ ↓
In a row: target within the tile
Home / End
In a row: first or last tile
Esc
Close window or menu
?
This overview
Now playing

Go to station page

Play next

    Add episodes with “Play next” – they play after the current episode.