← KV Cache Optimization for Long-Context LLM Inference
Page History
Every revision of this knowledge page since it was created on June 16, 2026. Each entry shows what kind of change was made, how the page grew or shrank, and the source articles that drove it — so the paper's memory is inspectable, not taken on faith.
Updated from new reportingOctober 21, 2026 (+2,906 chars)
Updated from new reportingOctober 16, 2026 (+2,445 chars)
Updated from new reportingOctober 15, 2026 (+3,427 chars)
- BELT : BEhavior-Aligned Low-Rank KV Cache Compression with Learned Top-k Rank Allocation
- CacheLink: Functional KV Transport for Zero-Prefill Cross-Model Context Handoff
- E8 Lattice Key Quantization for Low-Bit KV-Cache Compression
- KV4-RL: Unlocking Faster Asynchronous Rollouts for RLVR with INT4 KV Cache Quantization
- ChampionKV: Head-Wise Adaptive Budget Allocation for KV Cache Compression
- LatentRemedy: Efficient Remedy For Non-Prefix KV Cache Reuse in Multi-Agent Systems
- …and 1 more source articles
Updated from new reportingOctober 14, 2026 (+1,824 chars)
Updated from new reportingOctober 9, 2026 (+1,937 chars)
Updated from new reportingOctober 9, 2026 (+3,559 chars)
Updated from new reportingSeptember 25, 2026 (+3,159 chars)
Updated from new reportingSeptember 23, 2026 (+1,889 chars)
Updated from new reportingSeptember 22, 2026 (+1,840 chars)
Updated from new reportingSeptember 14, 2026 (+2,014 chars)
Updated from new reportingSeptember 14, 2026 (+2,827 chars)
- Decoupled Attention Fusion: Accelerating RAG with Efficient KV Cache Reuse
- DynaCalKV: Key-Value Cache Compression via Head Grouping and Adaptive Rank Allocation
- Stage-Replay Divergence Follows the KV Cache: Fixed-Prefix Precision Controls and Bidirectional Cache Transplantation
- MemForest: An Efficient Agent Memory System with Hierarchical Temporal Indexing
- HijackKV: New Threat in Position-Independent KV Cache Reuse
- RAP: KV-Cache Compression via RoPE-Aligned Pruning
Updated from new reportingSeptember 13, 2026 (+1,506 chars)
Updated from new reportingSeptember 12, 2026 (+1,277 chars)
Page createdJune 16, 2026 (+1,962 chars)
- Attenuation of Inflammatory Cytokines by Selective Vagal Motor Stimulation via Silicone Spiral Nerve Cuff
- Attenuation of Inflammatory Cytokines by Selective Vagal Motor Stimulation via Silicone Spiral Nerve Cuff
- UniversalRAG: Retrieval-Augmented Generation over Corpora of Diverse Modalities and Granularities
- Revisiting Outage for Edge Inference Systems
- ANSR-DT: A Neuro-Symbolic Framework for Adaptive and Explainable Digital Twins
- ANSR-DT: A Neuro-Symbolic Framework for Adaptive and Explainable Digital Twins
- …and 4 more source articles