Sergey Levine @svlevine · 07 Aug 2026
Action chunking is a mysteriously effective method. Modern large-scale imitation learning basically doesn't work without it. But why does it actually help? In our new paper, we try to break down the reasons. As the saying goes, what happened next might surprise you...
81 631Views
744Likes
70Reposts
16Replies
0Quotes
678Bookmarks
Is that a lot?
1.69×vs this author's median48 357 views is typical
64Percentile for this authorof 11 recent posts
14.9×vs 100K–1M median5 495 views is typical
59.94%Reachviews ÷ followers
1.02%Engagement rateof viewers reacted