Kundaje thread (5/9): short-context models ill-suited for motif selectivity questions

anshulkundaje · x · 2026-11-19

Part 5 of Anshul Kundaje's thread: short-context models like BPNet are not trained for or strong at selectivity questions; genome-wide long-context models have much better discriminative performance. See thread-start item.

Related event: Stanford's Kundaje fires nine-tweet rebuttal over TF cooperativity study(10 posts)→

Original post →

More from Research

Research channel →