tweetindex

Alex Zhavoronkov, PhD (aka Aleksandrs Zavoronkovs) @biogerontology · 03 Sep 2026

To explain a black-box medical image classifier, most counterfactual methods train a second black box (GAN/diffusion). This King's College London paper drops the generator, building counterfactuals from the classifier's causal evidence. 85% on brain MRI. https://t.co/oHljtEL6nQ https://t.co/9oTQVSHHND
1 177Views
18Likes
5Reposts
0Replies
0Quotes
2Bookmarks

Is that a lot?

1.77×vs this author's median664 views is typical
54Percentile for this authorof 11 recent posts
0.46×vs 10K–100K median2 578 views is typical
2.99%Reachviews ÷ followers
1.95%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →