RippleKV allocates KV cache budgets across transformer layers by measuring how much perturbing each layer's value cache shifts the model's output distribution, and it reports the best average LongBench score among the compared compression methods at matched budgets.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
RippleKV: Cross-Layer KV Cache Allocation via Perturbation Propagation
RippleKV allocates KV cache budgets across transformer layers by measuring how much perturbing each layer's value cache shifts the model's output distribution, and it reports the best average LongBench score among the compared compression methods at matched budgets.