Papers › Unlocking Data-free Low-bit Quantization with Matrix Decomposition for KV Cache Compression

Unlocking Data-free Low-bit Quantization with Matrix Decomposition for KV Cache Compression

21 May 2024arXiv:2405.12591archive 2025-07-28

Peiyu Liu, Ze-Feng Gao, Wayne Xin Zhao, Yipeng Ma, Tao Wang, Ji-Rong Wen

Key-value~(KV) caching is an important technique to accelerate the inference of large language models~(LLMs), but incurs significant memory overhead. To compress the size of KV cache, existing methods often compromise precision or require extra data for calibration, limiting their practicality in LLM deployment. In this paper, we introduce \textbf{DecoQuant}, a novel data-free low-bit quantization technique based on tensor decomposition methods, to effectively compress KV cache. Our core idea is to adjust the outlier distribution of the original matrix by performing tensor decomposition, so that the quantization difficulties are migrated from the matrix to decomposed local tensors. Specially, we find that outliers mainly concentrate on small local tensors, while large tensors tend to have a narrower value range. Based on this finding, we propose to apply low-bit quantization to the large tensor, while maintaining high-precision representation for the small tensor. Furthermore, we utilize the proposed quantization method to compress the KV cache of LLMs to accelerate the inference and develop an efficient dequantization kernel tailored specifically for DecoQuant. Through extensive experiments, DecoQuant demonstrates remarkable efficiency gains, showcasing up to a ∼75\% reduction in memory footprint while maintaining comparable generation quality.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2405.12591")

Code

Syntology Ran 6 of 10 code samples harvested from 1 repository linked to this paper; 4 have no recorded run. Of those that ran: 3 ran · our draft was wrong; 3 ran with no contract checked.

By repository: official repository: 10 samples from 1 repository, 6 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

lpyhdzx/DecoQuant_code officialmentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

10 samples harvested; 6 ran; 0 honoured the contract we drafted; 4 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

3ran · our draft was wrong
3ran
4unverified

Licence: 10 of the 10 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from lpyhdzx/DecoQuant_code. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

get_last_checkpoint_or_last_model lpyhdzx/DecoQuant_code/icl/utils.py official repository ran no licence file found · pointer only · 55c898dd0bb4836e · report
get_parameter_number lpyhdzx/DecoQuant_code/icl/evaluate_icl_ori.py official repository ran · our draft was wrong no licence file found · pointer only · 383d339873e8f63f · report
parse_checkpoint_step lpyhdzx/DecoQuant_code/icl/utils.py official repository ran fingerprinted no licence file found · pointer only · 231f5cf156c8b486 · report
quantize_activation_per_token_absmax lpyhdzx/DecoQuant_code/icl/models/fake_quant.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · c1cb961b4f5cb75c · report
quantize_weight_per_channel_absmax lpyhdzx/DecoQuant_code/icl/models/fake_quant.py official repository ran fingerprinted no licence file found · pointer only · 39396670c3373c8e · report
quantize_weight_per_tensor_absmax lpyhdzx/DecoQuant_code/icl/models/fake_quant.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · dd3493503cb95888 · report
get_act_scales lpyhdzx/DecoQuant_code/language_model/smoothquant/calibration.py official repository unverified no licence file found · pointer only · 3fed3338f2ffb08f · report
get_dataset lpyhdzx/DecoQuant_code/icl/icl_dataset_loading.py official repository unverified no licence file found · pointer only · 062aa2adf5c6b989 · report
get_model_tokenizer_device_isac lpyhdzx/DecoQuant_code/icl/evaluate_icl_ori.py official repository unverified no licence file found · pointer only · 46b0ba2dfd30560a · report
get_static_decoder_layer_scales lpyhdzx/DecoQuant_code/language_model/smoothquant/calibration.py official repository unverified no licence file found · pointer only · a5a75a01c343fb3b · report

Tasks

QuantizationTensor Decomposition

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections