# DAMP: Decay-Aware Mixed-Precision Recurrent-State Quantization

arXiv: https://arxiv.org/abs/2608.27513 · HTML: https://mlsanity.com/paper/2608.27513/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in KV cache quantization, one of 12 directions in Efficient LLM inference and serving.
- Builds on 31 of its 40 references that are in the map.
- First posted on arXiv 27 Aug 2026.

## Placed in

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

## Builds on (most cited first, 30 of 31)

- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark](https://mlsanity.com/paper/2406.01574/) (2024)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [RULER: What's the Real Context Size of Your Long-Context Language Models?](https://mlsanity.com/paper/2404.06654/) (2024)
- [AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration](https://mlsanity.com/paper/2306.00978/) (2023)
- [Kimi k1.5: Scaling Reinforcement Learning with LLMs](https://mlsanity.com/paper/2501.12599/) (2025)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [Measuring Coding Challenge Competence With APPS](https://mlsanity.com/paper/2105.09938/) (2021)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs](https://mlsanity.com/paper/2404.00456/) (2024)
- [WildChat: 1M ChatGPT Interaction Logs in the Wild](https://mlsanity.com/paper/2405.01470/) (2024)
- [KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization](https://mlsanity.com/paper/2401.18079/) (2024)
- [Gated Delta Networks: Improving Mamba2 with Delta Rule](https://mlsanity.com/paper/2412.06464/) (2024)
- [SpinQuant: LLM quantization with learned rotations](https://mlsanity.com/paper/2405.16406/) (2024)
- [Kimi Linear: An Expressive, Efficient Attention Architecture](https://mlsanity.com/paper/2510.26692/) (2025)
- [Beyond Benchmarks: MathArena as an Evaluation Platform for Mathematics with LLMs](https://mlsanity.com/paper/2605.00674/) (2026)
- [TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate](https://mlsanity.com/paper/2504.19874/) (2025)
- [OstQuant: Refining Large Language Model Quantization with Orthogonal and Scaling Transformations for Better Distribution Fitting](https://mlsanity.com/paper/2501.13987/) (2025)
- [Pretraining Large Language Models with NVFP4](https://mlsanity.com/paper/2509.25149/) (2025)
- [Towards Robust Mathematical Reasoning](https://mlsanity.com/paper/2511.01846/) (2025)
- [ResQ: Mixed-Precision Quantization of Large Language Models with Low-Rank Residuals](https://mlsanity.com/paper/2412.14363/) (2024)
- [MambaQuant: Quantizing the Mamba Family with Variance Aligned Rotation Methods](https://mlsanity.com/paper/2501.13484/) (2025)
- [Quamba: A Post-Training Quantization Recipe for Selective State Space Models](https://mlsanity.com/paper/2410.13229/) (2024)
- [Quamba2: A Robust and Scalable Post-training Quantization Framework for Selective State Space Models](https://mlsanity.com/paper/2503.22879/) (2025)
- [MixKVQ: Query-Aware Mixed-Precision KV Cache Quantization for Long-Context Reasoning](https://mlsanity.com/paper/2512.19206/) (2025)
- [On State Reduction in Linear Attention](https://mlsanity.com/paper/2602.04852/) (2026)

## Built on by (most cited first, 6 of 6)

- [Low-Bit Recurrent States in Hybrid Language Models](https://mlsanity.com/paper/2609.30950/) (2026)
- [SketchSSM: Write to the Full State, Read from a Compact Sketch](https://mlsanity.com/paper/2609.33051/) (2026)
- [STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization](https://mlsanity.com/paper/2609.38169/) (2026)
- [Low-Discrepancy Dither for Quantized Recurrent State Caches](https://mlsanity.com/paper/2609.39185/) (2026)
- [When Dynamic Scaling Fails: A Scale-State Feedback Governs Recurrent-State Quantization in Linear-Attention and SSM LLMs](https://mlsanity.com/paper/2610.15139/) (2026)
- [Preserving Updates in Low-Bit Recurrent State Quantization](https://mlsanity.com/paper/2610.15709/) (2026)

## Abstract (arXiv)

Complex reasoning and agentic applications increasingly rely on long-context inference, where growing KV caches increase both memory usage and decoding overhead. Hybrid models reduce these costs by combining Softmax Attention with Gated DeltaNet (GDN) or Kimi Delta Attention (KDA), which maintain fixed-size recurrent states. These states are commonly stored in FP32 and consume substantial GPU memory, while their updates are limited by memory bandwidth. Quantization can reduce both storage footprint and memory traffic, but we find that uniform INT8 and FP8 degrade complex reasoning accuracy, while INT4 and NVFP4 collapse it to near zero. To our knowledge, this is the first study of post-training recurrent-state quantization for GDN and KDA. Our analysis reveals that outliers in GDN and KDA states are concentrated in particular key channels and value dimensions. Learned decay influences how much quantization error is retained. We find that largely the same GDN heads and KDA key channels exhibit slow decay across tasks. Based on these insights, we propose DAMP, which jointly considers quantization error and decay-based error retention to select high-risk key channels offline. Under a fixed storage budget, it retains these channels in FP16 and stores the remainder in INT8. We evaluate DAMP on Qwen3.6-35B, Kimi-Linear-48B and Kimi-K3 across six reasoning and code generation benchmarks. At 9.9 bits per state value, DAMP maintains average accuracy close to FP32. In SGLang, DAMP reduces recurrent-state storage by 69.1%, accelerates the recurrent-state update kernel by up to 2.59x , and lowers full-model time per output token by up to 19.0%.