SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.05255 · 2022

Narcissus: A Practical Clean-Label Backdoor Attack with Limited Information

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 2 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2204.05255.json · badge: /api/ran/2204.05255.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
reds-lab/narcissus community (classifier label)
classifier basis not recorded on the link
2 ran · 1 no recorded run
FunctionStatusWhere it lives
drop_connect Ran
ran; our contract draft was wrong, not the code
reds-lab/narcissus/models/efficientnet.py
pointer only (licence: MIT); the stored copy get_code serves, from EnumaCaliber/BRIDGE, is pointer only (licence: NONE) · get_code("4304a326c593f8db")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from EnumaCaliber/BRIDGE, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

swish Ran
ran; our contract draft was wrong, not the code
reds-lab/narcissus/models/efficientnet.py
pointer only (licence: MIT); the stored copy get_code serves, from EnumaCaliber/BRIDGE, is pointer only (licence: NONE) · get_code("8737c82de631cffc")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from EnumaCaliber/BRIDGE, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

narcissus_gen Not yet run
unverified (harvested, no recorded run)
reds-lab/narcissus/narcissus_function.py
code served (permissive licence) · get_code("4eff6a7b69bf9fe8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 2 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
swish 32 other papers
5 dated earlier, 27 later
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
2002.11318 (2020-02)
Can we have it all? On the Trade-off between Spatial and Adversarial Robustness of Neural Networks
2106.01440 (2021-06)
Memory Wrap: a Data-Efficient and Interpretable Extension to Image Classification Models
and 29 more not named here
drop_connect 29 other papers
7 dated earlier, 22 later
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
1905.11946 (2019-05)
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
2002.09564 (2020-02)
Towards Robust and Reproducible Active Learning Using Neural Networks
and 26 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Backdoor attacks insert malicious data into a training set so that, during inference time, it misclassifies inputs that have been patched with a backdoor trigger as the malware specified label. For backdoor attacks to bypass human inspection, it is essential that the injected data appear to be correctly labeled. The attacks with such property are often referred to as "clean-label attacks." Existing clean-label backdoor attacks require knowledge of the entire training set to be effective. Obtaining such knowledge is difficult or impossible because training data are often gathered from multiple sources (e.g., face images from different users). It remains a question whether backdoor attacks still present a real threat. This paper provides an affirmative answer to this question by designing an algorithm to mount clean-label backdoor attacks based only on the knowledge of representative examples from the target class. With poisoning equal to or less than 0.5% of the target-class data and 0.05% of the training set, we can train a model to classify test examples from arbitrary classes into the target class when the examples are patched with a backdoor trigger. Our attack works well across datasets and models, even when the trigger presents in the physical world. We explore the space of defenses and find that, surprisingly, our attack can evade the latest state-of-the-art defenses in their vanilla form, or after a simple twist, we can adapt to the downstream defenses. We study the cause of the intriguing effectiveness and find that because the trigger synthesized by our attack contains features as persistent as the original semantic features of the target class, any attempt to remove such triggers would inevitably hurt the model accuracy first.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2204.05255")
get_code_for_paper("2204.05255")
have("2204.05255")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2204.05255.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2204.05255.svg)](https://syntology.ai/paper/2204.05255)

Connect an agent — have() is free.