ManifoldKV: Training-Free KV Cache Compression via Euclidean Outlier Detection

TL;DR

ManifoldKV introduces a training-free, Euclidean distance-based method for KV-cache compression that improves robustness and performance in long-context inference by effectively selecting salient tokens.

Contribution

It proposes a novel Euclidean distance-based scoring method for KV-cache compression, outperforming cosine-based methods and requiring minimal implementation effort.

Findings

Achieves 95.7% accuracy at 4K-16K contexts with 20% compression.

Reduces directional collisions in multi-key retrieval, improving accuracy.

Restores accuracy to 84.3% at 25% compression with WindowedManifoldKV.

Abstract

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-invariant and can discard magnitude cues that distinguish semantically salient tokens. We propose ManifoldKV, a training-free scorer that ranks tokens by Euclidean distance to the key centroid, capturing both angular and radial deviations. On the RULER benchmark, ManifoldKV achieves 95.7% accuracy at 4K-16K contexts with 20% compression; matching the best geometric baseline while improving robustness in two regimes where cosine scoring fails. First, on multi-key retrieval, ManifoldKV reduces directional collisions, achieving 92.4% vs KeyDiff's 77.0% (+15.4 points) on 3-key NIAH at 50% compression. Second, to address dilution and performance collapse of global centroids at 64K context, we introduce WindowedManifoldKV, which restores accuracy to 84.3% at 25% compression, a 49-point recovery over global L2 and +3.2 points over KeyDiff. The method requires only 3 lines of code and works across 4 architectures without tuning.