# DCKV: Learned Transform Coding of the Prompt KV Cache in Hybrid Language Models

Preprint · Egor Maximov, Aleksei Merkelov, Vasiliy Alekseev, Ivan Krivorotov, Valentin Beliaev, Nikita Arsenin, Dmitry Bocharov, Aleksei Goncharov

- Archive id: arxiv:2610.36950
- Published: 2026-10-27
- arXiv: https://arxiv.org/abs/2610.36950
- Citations: 0
- Source URL: https://arxiv.org/abs/2610.36950
- PDF: https://arxiv.org/pdf/2610.36950
- Provenance: arxiv

## Abstract

Hybrid language models keep a key-value (KV) cache only in their few full-attention layers, but this cache is still the part of a prompt's state that grows with the prompt and that must be stored or moved to reuse it. We present DCKV, a differentiable codec for the KV cache and the first KV-cache codec that is trained end to end. Its transform, its quantizer and its entropy models, which are conditioned on the prompt tokens and on already decoded neighbouring positions, are optimized together against the frozen language model, both for the code length that the entropy models assign and for the change of the model's output. On Qwen3.5-2B, DCKV stores a compressed token in more than 55 times fewer bytes than bf16 at a loss of at most 1.8 points on three long-context benchmarks, and in more than 100 times fewer bytes at a loss of at most 3.6 points.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

Topics: [Efficient Training & Inference](https://space-kitty.com/topics/efficient-ai), [LLM Architectures & Pretraining](https://space-kitty.com/topics/llm-architectures)

Cited in dossiers:
- [Efficient Training & Inference: State of the Field (October 2026)](https://space-kitty.com/dossiers/efficient-ai)
- [LLM Architectures & Pretraining: State of the Field (October 2026)](https://space-kitty.com/dossiers/llm-architectures)

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.