SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2009.02276 · ICLR · 2021

Witches' Brew: Industrial Scale Data Poisoning via Gradient Matching

Liam Fowl, Tom Goldstein, W Huang, Jonas Geiping, Gavin Taylor, Wojciech Czaja, Michael Moeller

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 1 has 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/2009.02276.json · badge: /api/ran/2009.02276.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
JonasGeiping/poisoning-gradient-matching marked official in Papers with Code
Papers with Code archive
2 ran · 1 no recorded run
zjfheart/poison-adv-training community
Papers with Code archive
2 ran · 0 no recorded run
FunctionStatusWhere it lives
_Witch Ran
ran on a synthesized input · constructed an object; no result was computed
JonasGeiping/poisoning-gradient-matching/forest/witchcoven/witch_matching.py
pointer only (licence: LGPL-2.1) · get_code("21400790f199e260")
How it ran

The body of this sample is withheld: LGPL-2.1, as GitHub detects it on this paper's repository, JonasGeiping/poisoning-gradient-matching, is not in the permissive set Syntology serves inline. 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.

_passenger_loss Ran
ran, honoured the contract we drafted
zjfheart/poison-adv-training/tar_tools/losses.py
pointer only (licence: NONE) · get_code("06d4e925899711d2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zjfheart/poison-adv-training 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.

cw_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
JonasGeiping/poisoning-gradient-matching/forest/witchcoven/witch_matching.py
pointer only (licence: LGPL-2.1) · get_code("6a5967011bf8de7a")
How it ran

The body of this sample is withheld: LGPL-2.1, as GitHub detects it on this paper's repository, JonasGeiping/poisoning-gradient-matching, is not in the permissive set Syntology serves inline. 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.

similarity_loss Ran
ran; our contract draft was wrong, not the code
zjfheart/poison-adv-training/tar_tools/losses.py
pointer only (licence: NONE) · get_code("aa126611580474c9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zjfheart/poison-adv-training 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.

WitchGradientMatching Not yet run
unverified (harvested, no recorded run)
JonasGeiping/poisoning-gradient-matching/forest/witchcoven/witch_matching.py
pointer only (licence: LGPL-2.1) · get_code("a798eb5b7cd3ebaf")
How it ran

The body of this sample is withheld: LGPL-2.1, as GitHub detects it on this paper's repository, JonasGeiping/poisoning-gradient-matching, is not in the permissive set Syntology serves inline. 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. No invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 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, 1 later) and 3 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
cw_loss
brick of this paper's assembly: helper
2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
2004.00225 (2020-04)
MetaPoison: Practical General-purpose Clean-label Data Poisoning
2411.13731 (2024-11)
∆-Influence: Unlearning Poisons via Influence Functions
_Witch
brick of this paper's assembly: helper
1 other paper
1 dated earlier · 1 only as an assembly brick
2004.00225 (2020-04)
MetaPoison: Practical General-purpose Clean-label Data Poisoning

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

Data Poisoning attacks modify training data to maliciously control a model trained on such data. In this work, we focus on targeted poisoning attacks which cause a reclassification of an unmodified test image and as such breach model integrity. We consider a particularly malicious poisoning attack that is both "from scratch" and "clean label", meaning we analyze an attack that successfully works against new, randomly initialized models, and is nearly imperceptible to humans, all while perturbing only a small fraction of the training data. Previous poisoning attacks against deep neural networks in this setting have been limited in scope and success, working only in simplified settings or being prohibitively expensive for large datasets. The central mechanism of the new attack is matching the gradient direction of malicious examples. We analyze why this works, supplement with practical considerations. and show its threat to real-world practitioners, finding that it is the first poisoning method to cause targeted misclassification in modern deep networks trained from scratch on a full-sized, poisoned ImageNet dataset. Finally we demonstrate the limitations of existing defensive strategies against such an attack, concluding that data poisoning is a credible threat, even for large-scale deep learning systems.

For agents

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

get_harvested_code_for_paper("2009.02276")
get_code_for_paper("2009.02276")
have("2009.02276")

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

curl https://syntology.ai/api/ran/2009.02276.json

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

[![Syntology run record](https://syntology.ai/api/ran/2009.02276.svg)](https://syntology.ai/paper/2009.02276)

Connect an agent — have() is free.