arXiv:2610.06927v1 Announce Type: cross Abstract: The key-value (KV) cache of autoregressive transformers grows linearly with context length and dominates memory at long context. Most training-free r...
arXiv:2606. 08382v1 Announce Type: cross Abstract: Low-rank projection has emerged as a promising approach for compressing the KV cache by exploiting hidden-dimension redundancy.
arXiv:2607. 24331v1 Announce Type: new Abstract: As the inference phase of Large Language Models (LLMs) requires handling long context windows, the Key-Value (KV) cache initially appears to address this challenge but eventually becomes a significant bottleneck as the context window continues to grow.
arXiv:2609.24298v1 Announce Type: new Abstract: What limits KV-cache compression at extreme bit-rates? We argue that it is not the choice of compression scheme, but how its budget is allocated across...
arXiv:2610.17966v1 Announce Type: cross Abstract: Long-context inference makes the key-value (KV) cache a major memory bottleneck for large language models. We propose BELT, a low-rank KV cache compr...
arXiv:2610.02235v1 Announce Type: cross Abstract: Long-context decoding is increasingly constrained by key--value (KV) cache memory and bandwidth. Existing fixed-budget compression methods typically...
RAQ-MLA introduces a role-aware quantization scheme for Multi-Head Latent Attention (MLA) caches used in large language models. It analyzes the distinct functions of the cached latent and positional key tensors, applying tailored invertible transforms to each based on their impact on attention output. The method achieves near-lossless performance while reducing the cache size to about 2.5 bits per element, effectively increasing theoretical KV-cache capacity by 6.5×.
The paper introduces CORE, a method that jointly calibrates coverage and redistributes evicted mass for key–value (KV) cache in long‑context decoding. CORE combines query utility and log‑determinant coverage into a lightweight indexer, using a single calibrated distribution to both retain complementary KV states and parameterize latent‑memory writes. Experiments on three backbones show CORE outperforms the RULER baseline by up to 3.78 points at 90% compression and demonstrates strong decoding efficiency on LongBench and repeated‑eviction tests.
The paper introduces JoLT, a training‑free compressor that jointly allocates rank and precision for key‑value (KV) cache compression in long‑context language models. JoLT treats grouped prefill caches as fourth‑order tensors, applies partial Tucker decomposition along token and feature modes, and uses a rotated low‑bit quantizer for residuals, all governed by a single Lagrangian dual under a global byte constraint. Across five models from four architecture families, JoLT achieves 2–3× compression with less than 0.2% perplexity loss, and near‑lossless retrieval accuracy on LLaMA‑3.1‑8B at 64K context up to 3× compression.
arXiv:2608.23843v1 Announce Type: new Abstract: Long-context inference in large language models (LLMs) is increasingly limited by the memory required for the key-value (KV) cache. KV cache compressio...
The paper investigates KV‑cache eviction strategies for sparse‑attention models, showing that selecting the largest attention weights is nearly optimal—closing only a median 2–5 % of the gap to full attention. It further demonstrates that differences in performance between eviction methods largely stem from memory usage, with the new training‑free ContourKV allocator outperforming state‑of‑the‑art methods in most pairwise comparisons while matching their byte‑efficiency.
arXiv:2609.13285v1 Announce Type: cross Abstract: The KV cache is a primary bottleneck for Transformer decoding: its memory footprint and cache-read traffic grow with sequence length. Grouped-query a...