tweetindex

Phillip Isola @phillip_isola · 07 Jul 2026

Two common q's about RandOpt: 1. What if you repeat the process? 2. Can you avoid K-pass inference? We tried: guess --> ensemble --> distill --> repeat It works, and results in a final 1-pass inference model with good preformance. Code here: https://t.co/inXmt4sC3Z
14 337Views
98Likes
7Reposts
0Replies
0Quotes
52Bookmarks

Is that a lot?

1.02×vs this author's median14 021 views is typical
54Percentile for this authorof 11 recent posts
12.0×vs 10K–100K median1 195 views is typical
66.22%Reachviews ÷ followers
0.73%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →