/Tech2d ago

Stanford's Yoonho Lee argues that text optimization methods like prompting are undervalued compared to gradient-based weight training

The methods offer superior sample efficiency in low-data settings.

53373342449.1K
Original postalex zhang#905
Yoonho Lee@yoonholeee

http://x.com/i/article/2064017981982859264

9:51 AM · Jun 8, 2026 · 57.5K Views
Sentiment
Sentiment building, check back later.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS43.7KBOOKMARKS394LIKES313RETWEETS31REPLIES4
Omar Khattab@lateinteraction

as usual from @yoonholeee, this is extremely well-written and a great way to organize the major arguments in this space

Yoonho Lee@yoonholeee

http://x.com/i/article/2064017981982859264

2dViews 43.7KLikes 313Bookmarks 394
Ahmad Beirami@abeirami

Highly recommended post by @yoonholeee discussing rich new research directions!

Harness engineering will come to an end. An era of harness learning is in front of us, with massive room for empirical and theoretical research on data, architectures, and algorithms.

Yoonho Lee@yoonholeee

http://x.com/i/article/2064017981982859264

1dViews 5.4KLikes 24Bookmarks 30
Ahmad Beirami@abeirami

Excellent piece by @yoonholeee!

Harness should be learned, not engineered, and you'll find rich research directions on harness optimization in Yoonhoo's post.

Yoonho Lee@yoonholeee

http://x.com/i/article/2064017981982859264

1dViews 86Likes 0Bookmarks 2
Ahmad Beirami@abeirami

Excellent piece by @yoonholeee!

Harness should be learned, not engineered. You'll find rich research directions on harness optimization in Yoonhoo's post.

Yoonho Lee@yoonholeee

http://x.com/i/article/2064017981982859264

1dViews 151Likes 1Bookmarks 1
Ahmad Beirami@abeirami

Highly recommended post by @yoonholeee discussing rich new research directions!

Harness engineering will come to an end. An era of harness learning is in front of us, with massive room for empirical and theoretical research on data, architectures, and algorithms.

Yoonho Lee@yoonholeee

http://x.com/i/article/2064017981982859264

1dViews 9Likes 0Bookmarks 1
oso@osoleve

@yoonholeee @a1zhang I started making Tools For This, glitchlings has tools for corrupting text as well as a starter toolkit for comparative sequence analysis (and lots of integrations), all FL/OSS 🫡

https://github.com/osoleve/glitchlings

2dViews 91Likes 2
Rugbist@rugbist_

@lateinteraction @yoonholeee rare to see someone break it down without taking sides tbh

what stood out to u the most from it?

2dViews 159

@yoonholeee ^^ I'm building alot of this. Some differences. Improvement has to be gated, versioned for compliance. But there's ways to have best of both worlds. https://chatgpt.com/share/6a2731ea-d2d8-8330-a501-6f8001b7bb43

2dViews 8