FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference
Long-context LLM inference is increasingly limited by the memory and bandwidth cost of KV caches, yet aggressive compression can remove the layer-specific evidence needed for retrieval and multi-step reasoning. We introduce FreqDepthKV, an inference-time cache compression method that factorizes adjacent-layer KV states into shared low-frequency depth components and sparse high-frequency residuals. A lightweight online probe assigns attention heads to shared-depth, residual-depth, or exact cache modes according to their contribution to reconstruction-sensitive attention logits, allowing the com
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 61%pythongiant/KVBoost →
- PossiblePossibly related (embedding) · 60%xcena-dev/maru →
- PossiblePossibly related (embedding) · 59%luziyao1995/vllm →
- PossiblePossibly related (embedding) · 57%Breakthrough in long-context efficiency announced →
- PossiblePossibly related (embedding) · 26%vllm-project/vllm →
“Possibly related via embedding similarity 0.56 (not asserted). Timestamp check: artifact slightly before paper (-5d).”
- PossiblePossibly related (embedding) · 26%LMCache/LMCache →
“Possibly related via embedding similarity 0.56 (not asserted). Timestamp check: artifact slightly before paper (-5d).”
- LinkedLinked via arxiv author · 85%Anna Cordoba →
“FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference”
- LinkedLinked via arxiv author · 85%Adam Puente Tercero →
“FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference”