dsa @dsa · 16 Jul 2026
We found a general recipe for Solves: scale pretraining, then hill-climb with minimal in-house data. For the first time, one fine-tuning example can teach a new behavior that generalizes. Below: 4 folding strategies, learned from one example each and tested on held-out setups. https://t.co/dW6WZNDVKo
170 305Views
323Likes
28Reposts
3Replies
22Quotes
0Bookmarks
Is that a lot?
84.4×vs this author's median2 017 views is typical
83Percentile for this authorof 6 recent posts
58.8×vs 10K–100K median2 896 views is typical
16.5× audienceReachviews ÷ followers
0.22%Engagement rateof viewers reacted