Papers › Benchmarking Neural Network Robustness to Common Corruptions and Perturbations

Benchmarking Neural Network Robustness to Common Corruptions and Perturbations

28 Mar 2019ICLR 2019 5arXiv:1903.12261archive 2025-07-28

Dan Hendrycks, Thomas Dietterich

In this paper we establish rigorous benchmarks for image classifier robustness. Our first benchmark, ImageNet-C, standardizes and expands the corruption robustness topic, while showing which classifiers are preferable in safety-critical applications. Then we propose a new dataset called ImageNet-P which enables researchers to benchmark a classifier's robustness to common perturbations. Unlike recent robustness research, this benchmark evaluates performance on common corruptions and perturbations not worst-case adversarial perturbations. We find that there are negligible changes in relative corruption robustness from AlexNet classifiers to ResNet classifiers. Afterward we discover ways to enhance corruption and perturbation robustness. We even find that a bypassed adversarial defense provides substantial common perturbation robustness. Together our benchmarks may aid future work toward networks that robustly generalize.

PaperPDFConference PDFCodeCode 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="1903.12261")

Code

Syntology Ran 2 of 3 code samples harvested from 1 repository linked to this paper; 1 has no recorded run. Of those that ran: 2 ran · our draft was wrong.

By repository: official repository: 3 samples from 1 repository, 2 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

14 repositories listed; official and paper-mentioned ones first.

hendrycks/robustness officialmentioned in papermentioned on GitHubpytorch report
EPFL-VILAB/XDEnsembles mentioned on GitHubpytorch report
MKYucel/zero_shot_corruption_benchmarks mentioned on GitHubpytorch report
YutingLi0606/SURE mentioned on GitHubpytorch report
allenai/robustnav mentioned on GitHubpytorchNOASSERTION report
automl/nes mentioned on GitHubpytorch report
deeplearning-wisc/hypo mentioned on GitHubpytorch report
feedbackward/addro mentioned on GitHubpytorch report
mr-eggplant/eata mentioned on GitHubpytorchNOASSERTION report
mr-eggplant/sar mentioned on GitHubpytorch report
yaodongyu/projnorm mentioned on GitHubpytorch report
yueatsprograms/ttt_cifar_release mentioned on GitHubpytorch report
yueatsprograms/ttt_imagenet_release mentioned on GitHubpytorch 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

3 samples harvested; 2 ran; 0 honoured the contract we drafted; 1 has no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · our draft was wrong
1unverified

Licence: 0 of the 3 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 hendrycks/robustness. “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.

clipped_zoom hendrycks/robustness/ImageNet-C/imagenet_c/imagenet_c/corruptions.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 5f0ee8450d2eac09 · report
clipped_zoom hendrycks/robustness/ImageNet-P/create_p/make_imagenet_p.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 466ca3cdf4524d82 · report
plasma_fractal hendrycks/robustness/ImageNet-C/imagenet_c/imagenet_c/corruptions.py official repository unverified Apache-2.0 (permissive) · 3fe212e00cbe6fb3 · report

Tasks

Adversarial DefenseBenchmarkingDomain Generalization

Datasets

Introduced by this paper, per the archive.

CIFAR-10CImageNet-CImageNet-PTiny-ImageNet-C

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Domain Generalization ImageNet-C ResNet-50 mean Corruption Error (mCE) 76.7 #41 of 47 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

1x1 ConvolutionAverage PoolingBatch NormalizationBottleneck Residual BlockConvolutionDense ConnectionsDropoutGlobal Average PoolingGrouped ConvolutionKaiming InitializationLocal Response NormalizationMax PoolingReLUResidual BlockResidual ConnectionSoftmax

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