The paper derives a characterization of minimax risk for KV cache compression and maps it to practical design principles and an algorithm tested on LongBench.
Revisiting frank-wolfe: Projection-free sparse convex optimization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
The risk of KV cache compression
The paper derives a characterization of minimax risk for KV cache compression and maps it to practical design principles and an algorithm tested on LongBench.