Mixture of Sparse Attention: Content-Based Learnable Sparse Attention via Expert-Choice Routing · HackerPapers discussionshttps://hackerpaper.kaipfister.de/p/2505.00315/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMTMixture of Sparse Attention: Content-Based Learnable Sparse Attention via Expert-Choice Routing — mentioned by remexrehttps://news.ycombinator.com/item?id=46882268urn:hackerpapers:mention:2505.00315:46882268Wed, 04 Feb 2026 06:36:45 GMT<p>how does this compare to MoSA (arXiv:2505.00315)? do you require that there&#x27;s a single contiguous window? and do you literally predict on position, or with a computed feature?</p><p><a href="https://hackerpaper.kaipfister.de/p/2505.00315/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LG