DepthWeave-KV achieves 8.3x KV cache memory reduction with near-full-cache task quality by factorizing key-value states across transformer layers using shared bases and token-adaptive residuals.
The modern mathematics of deep learning
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4representative citing papers
DMK extended to rectangular cuboids with arbitrary periodicity via localized octree evaluations on cubical tilings and Fourier-space root-level summation with truncated kernels for reduced periodicity.
Frequency-guided inter-layer KV sharing with logit-aware head routing nearly matches full-cache long-context accuracy at about 3.9× lower peak KV memory.
A frozen video diffusion backbone augmented with low-rank temporal adapters and a recursive prompt bank outperforms prior long-video generation methods on six benchmarks while tuning only 3.8% of parameters.
citing papers explorer
-
DepthWeave-KV: Token-Adaptive Cross-Layer Residual Factorization for Long-Context KV Cache Compression
DepthWeave-KV achieves 8.3x KV cache memory reduction with near-full-cache task quality by factorizing key-value states across transformer layers using shared bases and token-adaptive residuals.
-
Fast summation on rectangular cuboids with arbitrary periodicity in the DMK framework
DMK extended to rectangular cuboids with arbitrary periodicity via localized octree evaluations on cubical tilings and Fourier-space root-level summation with truncated kernels for reduced periodicity.
-
FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference
Frequency-guided inter-layer KV sharing with logit-aware head routing nearly matches full-cache long-context accuracy at about 3.9× lower peak KV memory.
-
Prompt-Adapter Context Routing for Parameter-Efficient Multi-Shot Long Video Extrapolation
A frozen video diffusion backbone augmented with low-rank temporal adapters and a recursive prompt bank outperforms prior long-video generation methods on six benchmarks while tuning only 3.8% of parameters.