ZHA: Zyphra Hierarchical Attention
Comments
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Summary
Zyphra's ZHA (Zyphra Hierarchical Attention) is a sparse attention method that builds a tree of compressed summaries over the key-value cache and routes each query coarse-to-fine through it, selecting only relevant tokens.