Papers › Towards Evaluating the Robustness of Neural Networks

Towards Evaluating the Robustness of Neural Networks

16 Aug 2016arXiv:1608.04644archive 2025-07-28

Nicholas Carlini, David Wagner

Neural networks provide state-of-the-art results for most machine learning tasks. Unfortunately, neural networks are vulnerable to adversarial examples: given an input x and any target classification t, it is possible to find a new input x′ that is similar to x but classified as t. This makes it difficult to apply neural networks in security-critical areas. Defensive distillation is a recently proposed approach that can take an arbitrary neural network, and increase its robustness, reducing the success rate of current attacks' ability to find adversarial examples from 95% to 0.5%. In this paper, we demonstrate that defensive distillation does not significantly increase the robustness of neural networks by introducing three new attack algorithms that are successful on both distilled and undistilled neural networks with 100% probability. Our attacks are tailored to three distance metrics used previously in the literature, and when compared to previous adversarial example generation algorithms, our attacks are often much more effective (and never worse). Furthermore, we propose using high-confidence adversarial examples in a simple transferability test we show can also be used to break defensive distillation. We hope our attacks will be used as a benchmark in future defense attempts to create neural networks that resist adversarial examples.

PaperPDFCodeCode 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 13 code samples harvested from 6 repositories linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 13 samples from 6 repositories, 12 ran (10 with no instrument failure; 2 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.

No sample is linked to this paper from 4 of its repositories that Syntology harvested for other papers:

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.

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

B-Yassine/Carlini-and-Wagner_InceptionV3_Imagenet mentioned on GitHubtfSyntology: not harvested report
Bhushan-Jagtap-2013/Adversarial_Attack_on_RNN mentioned on GitHubtfSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified report
Carco-git/CW_Attack_on_MNIST mentioned on GitHubpytorchSyntology: not harvested report
KevinHuang841006/AI_Security_training mentioned on GitHubtfSyntology: not harvested report
LeMinhThong/blackbox-attack mentioned on GitHubpytorchSyntology: not harvested report
MadryLab/cifar10_challenge mentioned on GitHubtfMITSyntology: not harvested report
OwenSec/DeepDetector mentioned on GitHubtfSyntology: not harvested report
PerryXDeng/adversarial_mnist_attacks mentioned on GitHubMITSyntology: 6 samples, 6 ran (0 instrument-failed) · 0 unverified report
SifatMd/Research-Papers mentioned on GitHubSyntology: not harvested report
abahram77/mnistChallenge mentioned on GitHubtfSyntology: not harvested report
abahram77/mnist_challenge mentioned on GitHubtfSyntology: not harvested report
bethgelab/cifar10_challenge mentioned on GitHubtfSyntology: not harvested report
dacostaHugo/Adversarial_attacks mentioned on GitHubSyntology: not harvested report
hope-yao/robust_attention_cifar mentioned on GitHubtfSyntology: not harvested report
inspire-group/advml-traffic-sign mentioned on GitHubMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to DARTS: Deceiving Autonomous Cars with Toxic Signs (arXiv:1802.06430; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 8 unverified); nothing checks that they implement this paper's method. report
jeromerony/adversarial-library mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Trust Region Based Adversarial Attack on Neural Networks (arXiv:1812.06371; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 7 where Syntology's instrument failed) · 2 unverified), Minimally distorted Adversarial Examples with a Fast Adaptive Boundary Attack (arXiv:1907.02044; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 4 where Syntology's instrument failed) · 2 unverified), Towards Large yet Imperceptible Adversarial Image Perturbations with Perceptual Color Distance (arXiv:1911.02466; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 1 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified) and 5 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
jfc43/pixel-discretization mentioned on GitHubtfApache-2.0Syntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
jfc43/robust-attribution-regularization mentioned on GitHubtfApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Robust Attribution Regularization (arXiv:1905.09957; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
khieu/cifar10_challenge mentioned on GitHubtfSyntology: not harvested report
kkew3/pytorch-cw2 mentioned on GitHubpytorchMITSyntology: 3 samples, 2 ran (0 instrument-failed) · 1 unverified report
matanbt/attack-tabular mentioned on GitHubSyntology: not harvested report
taranO/defending-adversarial-attacks-by-RD mentioned on GitHubtfSyntology: not harvested report
tataiani/enhanced_regularizers_attributional_robustness mentioned on GitHubtfSyntology: not harvested report
ukky17/invert_MV_pytorch mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Understanding Deep Image Representations by Inverting Them (arXiv:1412.0035; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report
yevgm/deepadv3d mentioned on GitHubpytorchSyntology: not harvested report
zetongqi/attacking_neural_networks mentioned on GitHubtfSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
carlini/nn_robust_attacks tfSyntology: 1 sample, 1 ran (1 instrument-failed) · 0 unverified 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

13 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
2ran · our draft was wrong
9ran
1unverified

Licence: 1 of the 13 samples is 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 6 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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.

Bhushan-Jagtap-2013/Adversarial_Attack_on_RNN community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
generate_data in 66 papers test_attack.py ran · our draft was wrong BSD-2-Clause (permissive) · ae7a589f65750d71 · report fixture: signature data:FakeData/0/?, samples:int/0/int, targeted:bool/0/bool, start:int/0/int, inception:bool/0/bool · returned tuple · level recorded 2026-09-22
PerryXDeng/adversarial_mnist_attacks community (archive-listed) 6 samples, 6 ran (0 instrument-failed) · 0 unverified
activation in 30 papers neural_network.py ran MIT (permissive) · cc16b0d5bc15bb67 · report fixture: signature prev:ndarray/2/float, weights:ndarray/2/float, bias:ndarray/1/float · returned ndarray · level recorded 2026-09-26
cross_entropy in 92 papers
census: 6 implementations from 8 papers, 5 distinct outputs across 2 buckets, one shared input per bucket
neural_network.py ran fingerprinted MIT (permissive) · 39df9d693fbdc4de · report fixture: signature h:ndarray/2/float, y:ndarray/2/float · returned float32 · level recorded 2026-09-26
denormalize in 64 papers
census: 9 implementations from 11 papers, 8 distinct outputs across 4 buckets, one shared input per bucket
data_preparation.py ran fingerprinted MIT (permissive) · 6c0b9095663ceff6 · report fixture: signature x:ndarray/3/float · returned ndarray · level recorded 2026-09-26
normalize in 638 papers
census: 70 implementations from 82 papers, 51 distinct outputs across 11 buckets, one shared input per bucket
data_preparation.py ran fingerprinted MIT (permissive) · f33c321f1f8ee09c · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded 2026-09-26
sigmoid in 216 papers
census: 36 implementations from 56 papers, 7 distinct outputs across 3 buckets, one shared input per bucket
neural_network.py ran fingerprinted MIT (permissive) · cf589faab9919534 · report fixture: signature z:ndarray/1/float · returned ndarray · level recorded 2026-09-26
vectorized_label data_preparation.py ran fingerprinted MIT (permissive) · aca0af8d07cfa276 · report fixture: signature label:int/0/int · returned ndarray · level recorded 2026-09-26
jfc43/pixel-discretization community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
preprocess in 284 papers
census: 11 implementations from 13 papers, 10 distinct outputs across 5 buckets, one shared input per bucket
data-specific-discretization/ImageNet/util.py ran Apache-2.0 (permissive) · 544bb80adbb9ade1 · report fixture: signature images0:ndarray/3/int, codes:ndarray/2/float · returned ndarray · level recorded 2026-09-26
kkew3/pytorch-cw2 community (archive-listed) 3 samples, 2 ran (0 instrument-failed) · 1 unverified
get_cuda_state runutils.py ran MIT (permissive) · ab527cad8796ffd2 · report fixture: signature obj:callable/0/? · returned bool · level recorded 2026-09-26
predict in 126 papers runutils.py ran MIT (permissive) · 6189cef972a4615e · report fixture: signature net:callable/0/?, inputs:Tensor/2/float · returned Tensor · level recorded 2026-09-26
make_cuda_consistent runutils.py unverified MIT (permissive) · 9a04b87a21ee7dbd · report
zetongqi/attacking_neural_networks community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
load_image in 273 papers CW_l2attack.py ran · honoured contract no licence file found · pointer only · 0cb12596aa3c0436 · report fixture: signature image_location:str/0/str · returned ndarray · level recorded 2026-09-22
carlini/nn_robust_attacks community (archive-listed) 1 sample, 1 ran (1 instrument-failed) · 0 unverified
generate_data in 66 papers test_attack.py ran · our draft was wrong BSD-2-Clause (permissive) · 33c3e5273c448151 · report fixture: signature data:MockData/0/?, samples:int/0/int, targeted:bool/0/bool, start:int/0/int, inception:bool/0/bool · returned tuple · level recorded 2026-09-22

Tasks

Adversarial Attack

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