# High-accuracy Low-Bit KV-Cache Quantization via Local Distribution Restoration

arXiv: https://arxiv.org/abs/2607.16248 · HTML: https://mlsanity.com/paper/2607.16248/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in KV cache quantization, one of 12 directions in Efficient LLM inference and serving.
- Builds on 21 of its 25 references that are in the map.
- First posted on arXiv 27 Jun 2026.

## Placed in

- Efficient LLM inference and serving / [KV cache quantization](https://mlsanity.com/inference/kv-cache-quantization/)

## Builds on (most cited first, 21 of 21)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [Compressive Transformers for Long-Range Sequence Modelling](https://mlsanity.com/paper/1911.05507/) (2019)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization](https://mlsanity.com/paper/2401.18079/) (2024)
- [FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU](https://mlsanity.com/paper/2303.06865/) (2023)
- [ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference](https://mlsanity.com/paper/2410.21465/) (2024)
- [TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate](https://mlsanity.com/paper/2504.19874/) (2025)
- [ThinK: Thinner Key Cache by Query-Driven Pruning](https://mlsanity.com/paper/2407.21018/) (2024)
- [QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache](https://mlsanity.com/paper/2502.10424/) (2025)
- [LeanQuant: Accurate and Scalable Large Language Model Quantization with Loss-error-aware Grid](https://mlsanity.com/paper/2407.10032/) (2024)
- [PM-KVQ: Progressive Mixed-precision KV Cache Quantization for Long-CoT LLMs](https://mlsanity.com/paper/2505.18610/) (2025)
- [AsymKV: Enabling 1-Bit Quantization of KV Cache with Layer-Wise Asymmetric Quantization Configurations](https://mlsanity.com/paper/2410.13212/) (2024)
- [DartQuant: Efficient Rotational Distribution Calibration for LLM Quantization](https://mlsanity.com/paper/2511.04063/) (2025)

## Abstract (arXiv)

Long-context large language model inference relies on the KV cache to avoid redundant attention computation, but incurs high memory and bandwidth overheads. Low-bit KV-cache quantization reduces this cost, yet it severely degrade quality; particularly, one-bit quantization reduces accuracy from 84.2% to 47.8% on Llama-3.1-8B under RULER. Rather than common beliefs that absolute error of logits, we find that the root cause is structured local misranking, where the distribution of logits in top-K region is drifted. We thereby propose local distribution restoration, a new technique that detects steps with high local distribution risk from quantized-logit features and restores only the selected top-K candidate distribution before token selection. We implement DGAP to achieve local distribution restoration, with efficient risk detcetors and correctors. Expeirments show that on Llama-3.1-8B, DGAP recovers K1V1 RULER accuracy from 47.8% to 83.2% and reduces distribution drift from 0.38 to 0.14; across Llama, Mistral, and Qwen models, it preserves the persistent low-bit KV-cache footprint with modest decode overhead.