Papers › D²Quant: Accurate Low-bit Post-Training Weight Quantization for LLMs

D²Quant: Accurate Low-bit Post-Training Weight Quantization for LLMs

30 Jan 2026arXiv:2602.02546added by Syntology

Xianglong Yan, ChengZhu Bao, Zhiteng Li, Tianao Zhang, Shaoqiu Zhang, Ruobing Xie, Samm Sun, Yulun Zhang

Title, abstract, authors and date from arXiv's metadata (CC0); this paper is not in the Papers with Code archive (frozen 2025-07-28).

Large language models (LLMs) deliver strong performance, but their high compute and memory costs make deployment difficult in resource-constrained scenarios. Weight-only post-training quantization (PTQ) is appealing, as it reduces memory usage and enables practical speedup without low-bit operators or specialized hardware. However, accuracy often degrades significantly in weight-only PTQ at sub-4-bit precision, and our analysis identifies two main causes: (1) down-projection matrices are a well-known quantization bottleneck, but maintaining their fidelity often requires extra bit-width; (2) weight quantization induces activation deviations, but effective correction strategies remain underexplored. To address these issues, we propose D²Quant, a novel weight-only PTQ framework that improves quantization from both the weight and activation perspectives. On the weight side, we design a Dual-Scale Quantizer (DSQ) tailored to down-projection matrices, with an absorbable scaling factor that significantly improves accuracy without increasing the bit budget. On the activation side, we propose Deviation-Aware Correction (DAC), which incorporates a mean-shift correction within LayerNorm to mitigate quantization-induced activation distribution shifts. Extensive experiments across multiple LLM families and evaluation metrics show that D²Quant delivers superior performance for weight-only PTQ at sub-4-bit precision. The code and models will be available at https://github.com/XIANGLONGYAN/D2Quant.

PaperPDFCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 16 code samples harvested from 1 repository linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 5 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: found in paper text by Syntology: 16 samples from 1 repository, 11 ran (6 with no instrument failure; 5 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 11 samples that ran has no recorded date.

XIANGLONGYAN/D2Quant found in paper text by SyntologySyntology: 16 samples, 11 ran (5 instrument-failed) · 5 unverified report

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

16 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 5 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
3ran · our draft was wrong
2ran; the contract check could not run
5ran
5unverified

Licence: 0 of the 16 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 XIANGLONGYAN/D2Quant. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

XIANGLONGYAN/D2Quant found in paper text by Syntology 16 samples, 11 ran (5 instrument-failed) · 5 unverified
round_ste in 18 papers
census: 3 implementations from 6 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
d2quant/quantization/quantizer.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 675da8641c602b03 · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded, date not recorded
apply_mean_shift d2quant/quantization/dac.py ran Apache-2.0 (permissive) · c2745ed51515c941 · report fixture: signature rms:callable/0/?, y_student_list:list/1/Tensor, y_teacher_list:list/1/Tensor, clip:float/0/float, lmbd:float/0/float · returned Tensor · level recorded 2026-09-26
asym_quant in 5 papers d2quant/quantization/quantizer.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 5daeed014eb71da9 · report fixture: signature x:Tensor/2/float, scale:Tensor/1/float, zero:Tensor/1/int, maxq:int/0/int · returned tuple · level recorded 2026-09-22
compose_final_perm d2quant/quantization/gar.py ran Apache-2.0 (permissive) · 5a7bfb94df03681c · report fixture: signature local_perms:list/1/Tensor, global_perm:Tensor/1/int, group_size:int/0/int · returned Tensor · level recorded 2026-09-26
compute_global_perm d2quant/quantization/gar.py ran · our draft was wrong Apache-2.0 (permissive) · 4bcf3cf0ed281cca · report fixture: signature diag_h:Tensor/1/float, group_size:int/0/int · returned Tensor · level recorded 2026-09-26
compute_local_perms d2quant/quantization/gar.py ran Apache-2.0 (permissive) · d16524faf29750ef · report fixture: signature diag_h:Tensor/1/float, group_size:int/0/int · returned list · level recorded 2026-09-26
fit_dual_scale_2bit d2quant/quantization/dsq_2bit.py ran Apache-2.0 (permissive) · 1b4ea44880a00a67 · report fixture: signature x:Tensor/2/float, iter:int/0/int, eps:float/0/float, do_normalize:bool/0/bool, ridge:float/0/float · returned tuple · level recorded 2026-09-26
fit_dual_scale_3bit d2quant/quantization/dsq_3bit.py ran Apache-2.0 (permissive) · 2e8023af78a306fa · report fixture: signature x:Tensor/2/float, iter:int/0/int, eps:float/0/float, do_normalize:bool/0/bool, ridge:float/0/float · returned tuple · level recorded 2026-09-26
get_minq_maxq d2quant/quantization/quantizer.py ran · honoured contract Apache-2.0 (permissive) · 56fe9b504cdc9689 · report fixture: signature bits:int/0/int, sym:bool/0/bool · returned tuple · level recorded 2026-09-26
get_rope_function_name in 8 papers d2quant/model_utils.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · 8005fc23ccbe61fb · report fixture: signature model:callable/0/? · returned str · level recorded 2026-09-24
model_type_extractor in 6 papers d2quant/model_utils.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · 5dcced271ec7bc12 · report fixture: signature model:callable/0/? · returned type · level recorded 2026-09-26
apply_deviation_aware_correction d2quant/quantization/dac.py unverified Apache-2.0 (permissive) · 11ea86283bc189ec · report
cleanup_memory in 8 papers d2quant/config.py unverified Apache-2.0 (permissive) · 58973fc4e2b31774 · report
get_layers in 20 papers d2quant/model_utils.py unverified Apache-2.0 (permissive) · 5cfb768c5400311b · report
log_snr_from_lists d2quant/quantization/diagnostics.py unverified Apache-2.0 (permissive) · 54417cda50d297b8 · report
ppl_eval d2quant/evaluation.py unverified Apache-2.0 (permissive) · c63931568d9a0771 · report

Results from the paper

The Papers with Code archive ends with its 2025-07-28 snapshot. This paper's arXiv identifier, 2602.02546, was issued in February 2026, after that date, so the archive has no leaderboard rows for it.

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