SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.12758 · 2023

Defending Against Backdoor Attacks by Layer-wise Feature Analysis

arXiv · PDF · Open in the Atlas

Code that ran

5 of 9 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 4 have no contract verdict; 1 failure of our instrument; 4 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/2302.12758.json · badge: /api/ran/2302.12758.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
najeebjebreel/dbalfa canonical (classifier label)
classifier basis not recorded on the link
5 ran · 4 no recorded run
FunctionStatusWhere it lives
ResNet Ran
ran on a synthesized input
najeebjebreel/dbalfa/my_models/resnet.py
code served (permissive licence) · get_code("aa8e4685d62e34d7")
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": 18,
    "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
double_conv Ran
ran; our contract draft was wrong, not the code
najeebjebreel/dbalfa/core/attacks/LIRA.py
code served (permissive licence) · get_code("6197a19b88fb1aed")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned Sequential. 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)

# The function double_conv takes in_channels and out_channels as parameters
# and returns a Sequential module with Conv2d, BatchNorm2d, and ReLU layers

in_channels = 3
out_channels = 16

ARGS = {
    'in_channels': in_channels,
    'out_channels': out_channels
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
get_secret_acc Ran
ran on a synthesized input
najeebjebreel/dbalfa/core/attacks/ISSBA.py
code served (permissive licence) · get_code("e3052317d27ff60f")
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 0, dtype family float, shape () 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

torch.manual_seed(0)

# Create small tensors for secret_true and secret_pred
# secret_true: ground truth labels (shape: batch_size x num_bits)
# secret_pred: model predictions before sigmoid (shape: batch_size x num_bits)
batch_size = 2
num_bits = 4

secret_true = torch.randint(0, 2, (batch_size, num_bits)).float()
secret_pred = torch.randn(batch_size, num_bits)

ARGS = {
    "secret_true": secret_true,
    "secret_pred": secret_pred,
}
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 0, dtype family float, shape () — shape () 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
l2_regularizer Ran
ran on a synthesized input
najeebjebreel/dbalfa/my_models/curves.py
code served (permissive licence) · get_code("cb72b7e1024c31b0")
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 function. 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)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    @property
    def l2(self):
        # Compute L2 norm of all parameters
        l2_norm = 0.0
        for param in self.parameters():
            l2_norm += torch.sum(param ** 2)
        return l2_norm

model = SimpleModel()
weight_decay = 0.01

ARGS = {
    "weight_decay": weight_decay,
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
my_imread Ran
ran on a synthesized input
najeebjebreel/dbalfa/core/attacks/LabelConsistent.py
code served (permissive licence) · get_code("3c2902242add3c64")
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 ndarray, ndim 3, dtype family int, shape (4, 3, 3) 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 cv2
import numpy as np
import tempfile
import os

# Create a temporary image file for testing
temp_dir = tempfile.mkdtemp()
temp_image_path = os.path.join(temp_dir, "test_image.png")

# Create a small test image and save it
test_image = np.zeros((4, 3, 3), dtype=np.uint8)
test_image[0:2, 0:2, 0] = 255  # Red channel
test_image[1:3, 1:3, 1] = 128  # Green channel
test_image[2:4, 0:3, 2] = 64   # Blue channel
cv2.imwrite(temp_image_path, test_image)

ARGS = {
    "file_path": temp_image_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (4, 3, 3) — shape (4, 3, 3) 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
get_features Not yet run
unverified (harvested, no recorded run)
najeebjebreel/dbalfa/utils.py
code served (permissive licence) · get_code("c469ddae1a6c4775")
How it ran

no invocation recorded.

get_inference_result Not yet run
unverified (harvested, no recorded run)
najeebjebreel/dbalfa/core/attacks/Blind.py
code served (permissive licence) · get_code("fc7187cc1bac10cc")
How it ran

no invocation recorded.

th Not yet run
unverified (harvested, no recorded run)
najeebjebreel/dbalfa/core/attacks/Blind.py
code served (permissive licence) · get_code("b1cfa48eb3407994")
How it ran

no invocation recorded.

thp Not yet run
unverified (harvested, no recorded run)
najeebjebreel/dbalfa/core/attacks/Blind.py
code served (permissive licence) · get_code("3dd252e1513af683")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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 7 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
double_conv 3 other papers
2 dated earlier, 1 later
1901.11390 (2019-01)
MONet: Unsupervised Scene Decomposition and Representation
2112.10098 (2021-12)
Initiative Defense against Facial Manipulation
2312.00833 (2023-12)
Lasagna: Layered Score Distillation for Disentangled Object Relighting
l2_regularizer 2 other papers
2 dated earlier
2005.00060 (2020-05)
Bridging Mode Connectivity in Loss Landscapes and Adversarial Robustness
2107.11228 (2021-07)
Taxonomizing local versus global structure in neural network loss landscapes

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

Training deep neural networks (DNNs) usually requires massive training data and computational resources. Users who cannot afford this may prefer to outsource training to a third party or resort to publicly available pre-trained models. Unfortunately, doing so facilitates a new training-time attack (i.e., backdoor attack) against DNNs. This attack aims to induce misclassification of input samples containing adversary-specified trigger patterns. In this paper, we first conduct a layer-wise feature analysis of poisoned and benign samples from the target class. We find out that the feature difference between benign and poisoned samples tends to be maximum at a critical layer, which is not always the one typically used in existing defenses, namely the layer before fully-connected layers. We also demonstrate how to locate this critical layer based on the behaviors of benign samples. We then propose a simple yet effective method to filter poisoned samples by analyzing the feature differences between suspicious and benign samples at the critical layer. We conduct extensive experiments on two benchmark datasets, which confirm the effectiveness of our defense.

For agents

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

get_harvested_code_for_paper("2302.12758")
get_code_for_paper("2302.12758")
have("2302.12758")

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

curl https://syntology.ai/api/ran/2302.12758.json

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

[![Syntology run record](https://syntology.ai/api/ran/2302.12758.svg)](https://syntology.ai/paper/2302.12758)

Connect an agent — have() is free.