Image de couverture de livre

Kitty: Accurate and Efficient 2-bit KV Cache Quantization with Dynamic Channel-wise Precision Boost

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Xia, Haojun, Wu, Xiaoxia, Li, Jisen, Wu, Robert, Wang, Junxiong, Wang, Jue, Li, Chenxi, Singhal, Aman, Shah, Alay Dilipbhai, Ariyak, Alpay, Zhuang, Donglin, Zhou, Zhongzhu, Athiwaratkun, Ben, Zheng, Zhen, Song, Shuaiwen Leon
Format:	Preprint
Publié:	2025
Sujets:	Machine Learning Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

OSCAR: Offline Spectral Covariance-Aware Rotation for 2-bit KV Cache Quantization
par: Zhou, Zhongzhu, et autres
Publié: (2026)

SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving
par: Jia, Jinda, et autres
Publié: (2026)

CARE: Covariance-Aware and Rank-Enhanced Decomposition for Enabling Multi-Head Latent Attention
par: Zhou, Zhongzhu, et autres
Publié: (2026)

Understanding and Steering the Cognitive Behaviors of Reasoning Models at Test-Time
par: Zhang, Zhenyu, et autres
Publié: (2025)

Imitate Optimal Policy: Prevail and Induce Action Collapse in Policy Gradient
par: Zhou, Zhongzhu, et autres
Publié: (2025)

Beat the long tail: Distribution-Aware Speculative Decoding for RL Training
par: Shao, Zelei, et autres
Publié: (2025)

When RL Meets Adaptive Speculative Training: A Unified Training-Serving System
par: Wang, Junxiong, et autres
Publié: (2026)

Introspective Diffusion Language Models
par: Yu, Yifan, et autres
Publié: (2026)

$V_1$: Unifying Generation and Self-Verification for Parallel Reasoners
par: Singh, Harman, et autres
Publié: (2026)

LayerKV: Optimizing Large Language Model Serving with Layer-wise KV Cache Management
par: Xiong, Yi, et autres
Publié: (2024)

Ladder-residual: parallelism-aware architecture for accelerating large model inference with communication overlapping
par: Zhang, Muru, et autres
Publié: (2025)

Accurate KV Cache Quantization with Outlier Tokens Tracing
par: Su, Yi, et autres
Publié: (2025)

FP6-LLM: Efficiently Serving Large Language Models Through FP6-Centric Algorithm-System Co-Design
par: Xia, Haojun, et autres
Publié: (2024)

ReST-KV: Robust KV Cache Eviction with Layer-wise Output Reconstruction and Spatial-Temporal Smoothing
par: An, Yongqi, et autres
Publié: (2026)

ZipCache: Accurate and Efficient KV Cache Quantization with Salient Token Identification
par: He, Yefei, et autres
Publié: (2024)

LAVa: Layer-wise KV Cache Eviction with Dynamic Budget Allocation
par: Shen, Yiqun, et autres
Publié: (2025)

Unlocking Data-free Low-bit Quantization with Matrix Decomposition for KV Cache Compression
par: Liu, Peiyu, et autres
Publié: (2024)

EMS: Adaptive Evict-then-Merge Strategy for Head-wise KV Cache Compression Based on Global-Local Importance
par: Li, Yingxin, et autres
Publié: (2024)

RotateKV: Accurate and Robust 2-Bit KV Cache Quantization for LLMs via Outlier-Aware Adaptive Rotations
par: Su, Zunhai, et autres
Publié: (2025)

FastCache: Optimizing Multimodal LLM Serving through Lightweight KV-Cache Compression Framework
par: Zhu, Jianian, et autres
Publié: (2025)

MadaKV: Adaptive Modality-Perception KV Cache Eviction for Efficient Multimodal Long-Context Inference
par: Li, Kunxi, et autres
Publié: (2025)

Squeeze Evolve: Unified Multi-Model Orchestration for Verifier-Free Evolution
par: Maheswaran, Monishwaran, et autres
Publié: (2026)

KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache
par: Liu, Zirui, et autres
Publié: (2024)

Data Diversification Methods In Alignment Enhance Math Performance In LLMs
par: Dokmeci, Berkan, et autres
Publié: (2025)

Think Deep, Think Fast: Investigating Efficiency of Verifier-free Inference-time-scaling Methods
par: Wang, Junlin, et autres
Publié: (2025)

Improving Model Alignment Through Collective Intelligence of Open-Source LLMS
par: Wang, Junlin, et autres
Publié: (2025)

AMS-KV: Adaptive KV Caching in Multi-Scale Visual Autoregressive Transformers
par: Xu, Boxun, et autres
Publié: (2025)

LouisKV: Efficient KV Cache Retrieval for Long Input-Output Sequences
par: Wu, Wenbo, et autres
Publié: (2025)

ContiguousKV: Accelerating LLM Prefill with Granularity-Aligned KV Cache Management
par: Zou, Jing, et autres
Publié: (2026)

Lossless KV Cache Compression to 2%
par: Yang, Zhen, et autres
Publié: (2024)

CorDA: Context-Oriented Decomposition Adaptation of Large Language Models for Task-Aware Parameter-Efficient Fine-tuning
par: Yang, Yibo, et autres
Publié: (2024)

LookaheadKV: Fast and Accurate KV Cache Eviction by Glimpsing into the Future without Generation
par: Ahn, Jinwoo, et autres
Publié: (2026)

Kitti_to_hdmapping_seq00
par: Chudziński
Publié: (2026)

CommonKV: Compressing KV Cache with Cross-layer Parameter Sharing
par: Wang, Yixuan, et autres
Publié: (2025)

PiKV: KV Cache Management System for Mixture of Experts
par: Liu, Dong, et autres
Publié: (2025)

CacheClip: Accelerating RAG with Effective KV Cache Reuse
par: Yang, Bin, et autres
Publié: (2025)

SqueezeAttention: 2D Management of KV-Cache in LLM Inference via Layer-wise Optimal Budget
par: Wang, Zihao, et autres
Publié: (2024)

MergeQuant: Accurate 4-bit Static Quantization of Large Language Models by Channel-wise Calibration
par: Wang, Jinguang, et autres
Publié: (2025)

KVmix: Gradient-Based Layer Importance-Aware Mixed-Precision Quantization for KV Cache
par: Li, Fei, et autres
Publié: (2025)

Accelerating LLM Inference Throughput via Asynchronous KV Cache Prefetching
par: Dong, Yanhao, et autres
Publié: (2025)