KV Cache Compression
Methods for compressing the key-value cache in transformer inference to reduce memory while preserving reasoning quality.
4 papers
Where this stands
The written synthesis of this thread is for subscribers. Subscribe.
How this thread developed
Oct 2026 · The Hong Kong University of Science and Technology (Guangzhou), Guangdong OPPO Mobile Telecommunications Corp., Ltd.
Low-rank KV cache compression preserves all tokens during long reasoning
Proposes iS-KV, an online low-rank KV-cache compression method that preserves recent tokens at full precision and folds older tokens into bounded-rank representations.
Oct 2026 · Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences
Unified retention and compensation for KV cache eviction via coverage calibration
Derives an exact factorization of KV cache eviction error into evicted attention mass and directional gap, and proposes CORE for unified retention and compensation.
1 further paper
Oct 2026 · Seoul National University, Korea University
Test-time method prevents drift in long video generation by keeping caching in-distribution
Identifies KV-provenance drift and proposes in-distribution caching to maintain quality in long video generation.
3 of 4 papers shown