← Back to all news
arXiv Machine Learning By Siddharth Bhandari, Lucas Gretta, Krishna Balasubramanian, Shiva Kasiviswanathan

Read What Matters: Query-Adaptive Quantization for KV Caches

Read the original on arXiv Machine Learning →

The paper introduces ReadKV, a query‑adaptive quantization method for KV caches that stores keys and values in progressive codes, allowing each decoding query to allocate different precision prefixes for keys and values. By optimizing a calibrated distortion objective under fixed budgets, ReadKV achieves near‑optimal allocation and demonstrates that query‑dependent access can outperform any query‑independent reader at the same read budget. Experiments on six base models show that reading only four bits on average from an eight‑bit cache increases perplexity by at most 0.66% while reducing logical reads and retained capacity compared to a 16‑bit cache, and it achieves lower latency than existing codecs on an NVIDIA A10G.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
5d ago

Dual-QK: Sharp Queries and Flat Keys for Prunable 2-bit KV Caches

Dual-QK introduces paired non‑orthogonal transforms for queries and keys to enable sharp queries and flat keys, improving 2‑bit quantization and dynamic channel pruning of KV caches. The method balances key scales for INT2 quantization while concentrating query energy for pruning, and supports channel‑0 protection and bucket‑relative RoPE for long‑context accuracy. Experiments on four models across five generative benchmarks show that Dual‑QK outperforms OSCAR at 40% query‑channel sparsity, achieving 6.8× KV‑cache compression, an 8.3× reduction in KV read volume, and up to 3.75× decoding throughput over unpruned BF16.

By Sunjoo Whang, Jungjun Oh, Minsung Kim, Dongho Seo, Jisu Shin, Gregory Kielian, Hoi-Jun Yoo, Sangjin Kim
efficiencybenchmarks
More like this →
arXiv AI
Jul 8

FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference

arXiv:2607. 06519v1 Announce Type: new Abstract: 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.

By Anna C\'ordoba, Adam Puente Tercero, Nerea Angulo Hijo, Mar Linares Tercero, Julia Barrientos, Ainhoa Miranda, Jes\'us Olivera
llmsnlpefficiencybenchmarks
More like this →