2026
ManifoldKV: Training-Free KV Cache Compression via Euclidean Outlier Detection
ICML 2026poster
Long-context inference is constrained by KV-cache memory, which grows linearly with sequence length; KV-cache compression therefore hinges on reliably selecting which past tokens to retain. Most geometry-based eviction methods score keys by cosine similarity to a global centroid, but cosine is scale…