Papers › Rethinking the Backward Propagation for Adversarial Transferability

Rethinking the Backward Propagation for Adversarial Transferability

22 Jun 2023NeurIPS 2023 11arXiv:2306.12685archive 2025-07-28

Xiaosen Wang, Kangheng Tong, Kun He

Transfer-based attacks generate adversarial examples on the surrogate model, which can mislead other black-box models without access, making it promising to attack real-world applications. Recently, several works have been proposed to boost adversarial transferability, in which the surrogate model is usually overlooked. In this work, we identify that non-linear layers (e.g., ReLU, max-pooling, etc.) truncate the gradient during backward propagation, making the gradient w.r.t. input image imprecise to the loss function. We hypothesize and empirically validate that such truncation undermines the transferability of adversarial examples. Based on these findings, we propose a novel method called Backward Propagation Attack (BPA) to increase the relevance between the gradient w.r.t. input image and loss function so as to generate adversarial examples with higher transferability. Specifically, BPA adopts a non-monotonic function as the derivative of ReLU and incorporates softmax with temperature to smooth the derivative of max-pooling, thereby mitigating the information loss during the backward propagation of gradients. Empirical results on the ImageNet dataset demonstrate that not only does our method substantially boost the adversarial transferability, but it is also general to existing transfer-based attacks. Code is available at https://github.com/Trustworthy-AI-Group/RPA.

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="2306.12685")

Code

Syntology Ran 3 of 15 code samples harvested from 2 repositories linked to this paper; 12 have no recorded run. Of those that ran: 2 ran · our draft was wrong; 1 ran · fixture could not drive it.

By repository: official repository: 14 samples from 1 repository, 3 ran; community (archive-listed): 1 sample from 1 repository, 0 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

trustworthy-ai-group/rpa officialmentioned in papermentioned on GitHubpytorchMIT report
Trustworthy-AI-Group/BPA 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

15 samples harvested; 3 ran; 0 honoured the contract we drafted; 12 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.

2ran · our draft was wrong
1ran · fixture could not drive it
12unverified

Licence: 0 of the 15 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 2 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.

conv1x1 Trustworthy-AI-Group/RPA/models/ghost_resnet.py official repository ran · our draft was wrong MIT (permissive) · d9def42110729a85 · report
conv3x3 Trustworthy-AI-Group/RPA/models/ghost_resnet.py official repository ran · our draft was wrong MIT (permissive) · 160bb14bd76201b4 · report
normalize Trustworthy-AI-Group/RPA/attack_eval_imagenet.py official repository ran · fixture could not drive it MIT (permissive) · a3268937736bc478 · report
dct Trustworthy-AI-Group/RPA/dct.py official repository unverified MIT (permissive) · 29cfe60b8aa41fee · report
dct1 Trustworthy-AI-Group/RPA/dct.py official repository unverified MIT (permissive) · aeed5d1d9b114dfb · report
idct1 Trustworthy-AI-Group/RPA/dct.py official repository unverified MIT (permissive) · b0140352504bce90 · report
load_weights Trustworthy-AI-Group/RPA/models/adv_inception_v3.py official repository unverified MIT (permissive) · 3db00532c993361c · report
make_layers Trustworthy-AI-Group/RPA/models/ghost_vgg.py official repository unverified MIT (permissive) · 9c8a208b6e0b0e40 · report
pnasnet5large Trustworthy-AI-Group/RPA/models/pnasnet.py official repository unverified MIT (permissive) · 2bc94b2177d6547b · report
resnet18 Trustworthy-AI-Group/RPA/models/ghost_resnet.py official repository unverified MIT (permissive) · e6c0acd71f6122eb · report
resnet18 Trustworthy-AI-Group/RPA/models/resnet.py official repository unverified MIT (permissive) · 426bc0e3032d5716 · report
senet154 Trustworthy-AI-Group/RPA/models/senet.py official repository unverified MIT (permissive) · cc8e994f8f117773 · report
trans_incep Trustworthy-AI-Group/RPA/attack_eval_imagenet.py official repository unverified MIT (permissive) · c530d5f24594c37a · report
trans_pnas Trustworthy-AI-Group/RPA/attack_eval_imagenet.py official repository unverified MIT (permissive) · ef20835b8d4b66b1 · report
linbp_backw_resnet50 Trustworthy-AI-Group/BPA/utils.py community (archive-listed) unverified MIT (permissive) · 561e559369d81957 · report

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Softmax

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