SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.01460 · ICML · 2024

Purify Unlearnable Examples via Rate-Constrained Variational Autoencoders

Yufei Wang, Alex Kot, Shijian Lu, Yap-Peng Tan, Yi Yu, Wenhan Yang, Song Xia, Em Lsp

arXiv · PDF · Open in the Atlas

Code that ran

11 of 13 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; 0 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 10 failures of our instrument; 2 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/2405.01460.json · badge: /api/ran/2405.01460.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
yuyi-sd/D-VAE canonical (classifier label)
classifier basis: census
10 ran · 2 no recorded run
yuyi-sd/d-vae canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
BDR Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/dataset.py
code served (permissive licence) · get_code("d67c0086a92ec8d7")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# BDR expects a tensor x that supports:
# - multiplication by scale (int)
# - .int() conversion
# - division by scale
# - .float() conversion
# Create a small float tensor
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
DenseNet121 Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/models/DenseNet.py
pointer only (licence: MIT); the stored copy get_code serves, from LIONS-EPFL/SAMPa, is pointer only (licence: NONE) · get_code("dfb395bae3d07d76")
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 LIONS-EPFL/SAMPa, 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.

DenseNet169 Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/models/DenseNet.py
pointer only (licence: MIT); the stored copy get_code serves, from LIONS-EPFL/SAMPa, is pointer only (licence: NONE) · get_code("f295fe3c209ecf78")
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 LIONS-EPFL/SAMPa, 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.

DenseNet201 Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/models/DenseNet.py
pointer only (licence: MIT); the stored copy get_code serves, from LIONS-EPFL/SAMPa, is pointer only (licence: NONE) · get_code("6383aee1c940e7a5")
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 LIONS-EPFL/SAMPa, 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.

JPEGcompression Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/dataset.py
code served (permissive licence) · get_code("85f37398aa70266a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned JpegImageFile. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

from PIL import Image
import io

# Create a small test image
img = Image.new('RGB', (10, 10), color='red')

ARGS = {
    'image': img
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type JpegImageFile — a JpegImageFile output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a JpegImageFile output the recorded ndim 3 and dtype family int describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
ResNet18 Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/models/ResNet.py
code served (permissive licence) · get_code("68634298232db423")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ResNet. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'num_classes': 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
ResNet34 Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/models/ResNet.py
code served (permissive licence) · get_code("b2e6f658713469be")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ResNet. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "num_classes": 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
ResNet50 Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/models/ResNet.py
code served (permissive licence) · get_code("97c797446a5cfee3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ResNet. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'num_classes': 10,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
ResNetDVAE Ran
ran on a synthesized input · constructed an object; no result was computed
yuyi-sd/d-vae/models/ResNet.py
code served (permissive licence) · get_code("7a6f115523f576f3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, ResNetDVAE: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

class BasicBlock(nn.Module):
    expansion = 1
    
    def __init__(self, in_planes, planes, stride=1):
        super(BasicBlock, self).__init__()
        self.conv1 = nn.Conv2d(in_planes, planes, kernel_size=3, stride=stride, padding=1, bias=False)
        self.bn1 = nn.BatchNorm2d(planes)
        self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=1, padding=1, bias=False)
        self.bn2 = nn.BatchNorm2d(planes)
        self.shortcut = nn.Sequential()
        if stride != 1 or in_planes != self.expansion * planes:
            self.shortcut = nn.Sequential(
                nn.Conv2d(in_planes, self.expansion * planes, kernel_size=1, stride=stride, bias=False),
                nn.BatchNorm2d(self.expansion * planes)
            )
    
    def forward(self, x):
        out = torch.relu(self.bn1(self.conv1(x)))
        out = self.bn2(self.conv2(out))
        out += self.shortcut(x)
        out = torch.relu(out)
        return out

ARGS = {
    "block": BasicBlock,
    "num_blocks": [2, 2, 2, 2],
    "num_classes": 10,
    "stem_end_block": 5,
    "latent_dim": 128,
    "imagenet": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResNetDVAE — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a ResNetDVAE output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
conv1x1 Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/models/resnet_official.py
pointer only (licence: MIT); the stored copy get_code serves, from medaitech/scrd4ad, is pointer only (licence: NONE) · get_code("2a80220dabcb742a")
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 medaitech/scrd4ad, 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.

conv3x3 Ran
ran; our contract draft was wrong, not the code
yuyi-sd/D-VAE/models/resnet_official.py
pointer only (licence: MIT); the stored copy get_code serves, from medaitech/scrd4ad, is pointer only (licence: NONE) · get_code("600ff2c45e0de056")
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 medaitech/scrd4ad, 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.

patch_noise_extend_to_img Not yet run
unverified (harvested, no recorded run)
yuyi-sd/D-VAE/dataset.py
code served (permissive licence) · get_code("aefa2123c98d35c6")
How it ran

no invocation recorded.

resnet18 Not yet run
unverified (harvested, no recorded run)
yuyi-sd/D-VAE/models/resnet_official.py
code served (permissive licence) · get_code("0e2be202574017d8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 10 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 9 the same month, 10 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
conv1x1 112 other papers
78 dated earlier, 3 the same month, 27 later, 4 same year, order not determinable · 6 only as an assembly brick
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
1806.01531 (2018-06)
Deep Mixture of Experts via Shallow Embedding
1905.12794 (2019-05)
Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback
and 109 more not named here
conv3x3 111 other papers
78 dated earlier, 3 the same month, 26 later, 4 same year, order not determinable · 6 only as an assembly brick
1512.03385 (2015-12)
Deep Residual Learning for Image Recognition
1806.01531 (2018-06)
Deep Mixture of Experts via Shallow Embedding
1905.12794 (2019-05)
Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback
and 108 more not named here
ResNet18 27 other papers
16 dated earlier, 1 the same month, 10 later
1605.07146 (2016-05)
Wide Residual Networks
1901.09847 (2019-01)
Error Feedback Fixes SignSGD and other Gradient Compression Schemes
1910.09652 (2019-10)
Kernelized Wasserstein Natural Gradient
and 24 more not named here
ResNet34 24 other papers
18 dated earlier, 1 the same month, 5 later
1605.07146 (2016-05)
Wide Residual Networks
1811.12814 (2018-11)
Graph-Based Global Reasoning Networks
2002.12047 (2020-02)
FMix: Enhancing Mixed Sample Data Augmentation
and 21 more not named here
ResNet50 14 other papers
8 dated earlier, 1 the same month, 5 later
1605.07146 (2016-05)
Wide Residual Networks
1811.12814 (2018-11)
Graph-Based Global Reasoning Networks
2006.13554 (2020-06)
Normalized Loss Functions for Deep Learning with Noisy Labels
and 11 more not named here
DenseNet121 11 other papers
8 dated earlier, 1 the same month, 2 later
2006.15815 (2020-06)
Adaptive Inertia: Disentangling the Effects of Adaptive Learning Rate and Momentum
2105.13937 (2021-05)
Polygonal Unadjusted Langevin Algorithms: Creating stable and efficient adaptive algorithms for neural networks
2108.09976 (2021-08)
Revealing the Distributional Vulnerability of Discriminators by Implicit Generators
and 8 more not named here
DenseNet169 9 other papers
7 dated earlier, 1 the same month, 1 later
2006.15815 (2020-06)
Adaptive Inertia: Disentangling the Effects of Adaptive Learning Rate and Momentum
2108.09976 (2021-08)
Revealing the Distributional Vulnerability of Discriminators by Implicit Generators
2201.13329 (2022-01)
Can Adversarial Training Be Manipulated By Non-Robust Features?
and 6 more not named here
DenseNet201 9 other papers
7 dated earlier, 1 the same month, 1 later
2006.15815 (2020-06)
Adaptive Inertia: Disentangling the Effects of Adaptive Learning Rate and Momentum
2108.09976 (2021-08)
Revealing the Distributional Vulnerability of Discriminators by Implicit Generators
2201.13329 (2022-01)
Can Adversarial Training Be Manipulated By Non-Robust Features?
and 6 more not named here
patch_noise_extend_to_img 2 other papers
1 the same month, 1 later
2405.03316 (2024-05)
Provably Unlearnable Data Examples
2603.04731 (2026-03)
When Priors Backfire: On the Vulnerability of Unlearnable Examples to Pretraining
resnet18 2 other papers
2 later
2512.10275 (2025-12)
Sample-wise Adaptive Weighting for Transfer Consistency in Adversarial Distillation
2605.21999 (2026-05)
Toward Understanding Adversarial Distillation: Why Robust Teachers Fail

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

Unlearnable examples (UEs) seek to maximize testing error by making subtle modifications to training examples that are correctly labeled. Defenses against these poisoning attacks can be categorized based on whether specific interventions are adopted during training. The first approach is training-time defense, such as adversarial training, which can mitigate poisoning effects but is computationally intensive. The other approach is pre-training purification, e.g., image short squeezing, which consists of several simple compressions but often encounters challenges in dealing with various UEs. Our work provides a novel disentanglement mechanism to build an efficient pre-training purification method. Firstly, we uncover rate-constrained variational autoencoders (VAEs), demonstrating a clear tendency to suppress the perturbations in UEs. We subsequently conduct a theoretical analysis for this phenomenon. Building upon these insights, we introduce a disentangle variational autoencoder (D-VAE), capable of disentangling the perturbations with learnable class-wise embeddings. Based on this network, a two-stage purification approach is naturally developed. The first stage focuses on roughly eliminating perturbations, while the second stage produces refined, poison-free results, ensuring effectiveness and robustness across various scenarios. Extensive experiments demonstrate the remarkable performance of our method across CIFAR-10, CIFAR-100, and a 100-class ImageNet-subset. Code is available at https: //github.com/yuyi-sd/D-VAE.

For agents

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

get_harvested_code_for_paper("2405.01460")
get_code_for_paper("2405.01460")
have("2405.01460")

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

curl https://syntology.ai/api/ran/2405.01460.json

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

[![Syntology run record](https://syntology.ai/api/ran/2405.01460.svg)](https://syntology.ai/paper/2405.01460)

Connect an agent — have() is free.