Papers › Decoupled Kullback-Leibler Divergence Loss

Decoupled Kullback-Leibler Divergence Loss

23 May 2023arXiv:2305.13948archive 2025-07-28

Jiequan Cui, Zhuotao Tian, Zhisheng Zhong, Xiaojuan Qi, Bei Yu, Hanwang Zhang

In this paper, we delve deeper into the Kullback-Leibler (KL) Divergence loss and mathematically prove that it is equivalent to the Decoupled Kullback-Leibler (DKL) Divergence loss that consists of 1) a weighted Mean Square Error (wMSE) loss and 2) a Cross-Entropy loss incorporating soft labels. Thanks to the decomposed formulation of DKL loss, we have identified two areas for improvement. Firstly, we address the limitation of KL/DKL in scenarios like knowledge distillation by breaking its asymmetric optimization property. This modification ensures that the 𝐰MSE component is always effective during training, providing extra constructive cues. Secondly, we introduce class-wise global information into KL/DKL to mitigate bias from individual samples. With these two enhancements, we derive the Improved Kullback-Leibler (IKL) Divergence loss and evaluate its effectiveness by conducting experiments on CIFAR-10/100 and ImageNet datasets, focusing on adversarial training, and knowledge distillation tasks. The proposed approach achieves new state-of-the-art adversarial robustness on the public leaderboard -- RobustBench and competitive performance on knowledge distillation, demonstrating the substantial practical merits. Our code is available at https://github.com/jiequancui/DKL.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.

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="2305.13948")

Code

Syntology Ran 7 of 11 code samples harvested from 4 repositories linked to this paper; 4 have no recorded run. Of those that ran: 6 ran · our draft was wrong; 1 ran with no contract checked.

By repository: official repository: 4 samples from 1 repository, 4 ran; community (archive-listed): 7 samples from 3 repositories, 3 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

jiequancui/DKL officialmentioned in papermentioned on GitHubpytorch report
FPNAS/LBGAT mentioned on GitHubpytorchMIT report
dvlab-research/LBGAT mentioned on GitHubpytorch report
jiequancui/LBGAT mentioned on GitHubpytorchMIT 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

11 samples harvested; 7 ran; 0 honoured the contract we drafted; 4 have no recorded run. Read from Syntology's graph 2026-09-25; that is when this build read the record, not when the samples ran.

6ran · our draft was wrong
1ran
4unverified

Licence: 4 of the 11 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 4 repositories linked to this paper, official or community; each sample names its own and says which. “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.

IKL_KD jiequancui/DKL/DKLv1/KD-dkl/mdistiller/distillers/IKL_KD.py official repository ran no licence file found · pointer only · b32e10adc9bf6201 · report
cross_entropy jiequancui/dkl/DKLv1/Adv-training-dkl/train_dkl_cifar10.py official repository ran · our draft was wrong no licence file found · pointer only · 59f6b4ce1de254e2 · report
dkl_loss jiequancui/dkl/DKLv1/Adv-training-dkl/train_dkl_cifar10.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · eb1d0d5bb0f0c627 · report
dkl_loss jiequancui/dkl/DKLv1/KD-dkl/mdistiller/distillers/IKL_KD.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · 6ae3a3d2a8a1acc4 · report
l2_norm FPNAS/LBGAT/lbgat.py community (archive-listed) ran · our draft was wrong fingerprinted MIT (permissive) · bbffec6aadbbfd95 · report
lbgat_loss dvlab-research/LBGAT/lbgat.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 342a07cb13a5e0bc · report
squared_l2_norm FPNAS/LBGAT/lbgat.py community (archive-listed) ran · our draft was wrong fingerprinted MIT (permissive) · 724c88e3f81d2123 · report
ResNet152_cifar jiequancui/LBGAT/models/resnetcifar.py community (archive-listed) unverified MIT (permissive) · 87fe041ab6db516b · report
ResNet18_cifar jiequancui/LBGAT/models/resnetcifar.py community (archive-listed) unverified MIT (permissive) · 60d3d2bb8b6c30e2 · report
ResNet34_cifar jiequancui/LBGAT/models/resnetcifar.py community (archive-listed) unverified MIT (permissive) · 4386c6d158344286 · report
lbgat_loss jiequancui/LBGAT/lbgat.py community (archive-listed) unverified MIT (permissive) · 2533491b22247ed9 · report

Tasks

Adversarial DefenseAdversarial RobustnessKnowledge Distillation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

DKLKnowledge Distillation

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