SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.08972 · 2024

Revealing the Dark Secrets of Extremely Large Kernel ConvNets on Robustness

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 4 failures of our instrument; 0 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/2407.08972.json · badge: /api/ran/2407.08972.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
lauch1ng/lkrobust canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
FunctionStatusWhere it lives
clean_accuracy Ran
ran; our contract draft was wrong, not the code
lauch1ng/lkrobust/adversarial_attack/utils.py
code served (permissive licence) · get_code("9d57e8d027d73f7d")
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 float, 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
import torch.nn as nn
import math

torch.manual_seed(0)

# Create a simple model that takes input and produces class logits
class SimpleClassifier(nn.Module):
    def __init__(self, input_dim: int, num_classes: int):
        super().__init__()
        self.linear = nn.Linear(input_dim, num_classes)
    
    def forward(self, x: torch.Tensor) -> torch.Tensor:
        return self.linear(x)

# Create model with input_dim=4, num_classes=3
model = SimpleClassifier(input_dim=4, num_classes=3)
model.eval()

# Create small tensors: batch_size=2, input_dim=4
x = torch.randn(2, 4)
# Create labels: 2 samples with classes in range [0, 2]
y = torch.tensor([0, 2])

# Define batch_size
batch_size = 100

# Device can be None (will default to x.device)
device = None

ARGS = {
    "model": model,
    "x": x,
    "y": y,
    "batch_size": batch_size,
    "device": device,
}
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 float, 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
cosine_scheduler Ran
ran, honoured the contract we drafted
lauch1ng/lkrobust/adversarial_attack/convnext_utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from boyazeng/understand_bias, is pointer only (licence: NONE) · get_code("b700fb0a3496b255")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from boyazeng/understand_bias, 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.

get_bn Ran
ran; our contract draft was wrong, not the code
lauch1ng/lkrobust/adversarial_attack/replknet.py
code served (permissive licence) · get_code("b4613da968a83fc8")
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 BatchNorm2d. 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 get_bn takes one parameter: channels
# It returns a BatchNorm2d or SyncBatchNorm based on use_sync_bn flag
# channels should be an integer representing the number of channels

ARGS = {
    "channels": 64
}
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 BatchNorm2d — a BatchNorm2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a BatchNorm2d 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
get_imagenet_data Ran
ran; our contract draft was wrong, not the code
lauch1ng/lkrobust/adversarial_attack/utils.py
code served (permissive licence) · get_code("811ac4191033004e")
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 DataLoader. 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
from torchvision import datasets, transforms
import os
import tempfile
from PIL import Image

torch.manual_seed(0)

# Create a minimal mock ImageFolder dataset to avoid file I/O
class MockImageFolder(torch.utils.data.Dataset):
    def __init__(self, root, transform=None):
        self.transform = transform
        self.samples = []
        # Create 2 dummy samples
        for i in range(2):
            self.samples.append((f"dummy_{i}.jpg", 0))
    
    def __len__(self):
        return len(self.samples)
    
    def __getitem__(self, idx):
        # Return a dummy RGB image (3 channels, 256x256)
        img = Image.new('RGB', (256, 256), color=(73, 109, 137))
        if self.transform:
            img = self.transform(img)
        return img, 0

# Monkey-patch datasets.ImageFolder to use our mock
original_imagefolder = datasets.ImageFolder
datasets.ImageFolder = MockImageFolder

ARGS = {
    'mean': [0.5, 0.5, 0.5],
    'std': [0.5, 0.5, 0.5],
    'bs': 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataLoader — a DataLoader output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataLoader 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
get_pred Ran
ran; our contract draft was wrong, not the code
lauch1ng/lkrobust/adversarial_attack/utils.py
code served (permissive licence) · get_code("317cf3d87d623d14")
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 1, dtype family int, shape (2,) 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)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()
images = torch.randn(2, 4)
device = "cpu"

ARGS = {
    "model": model,
    "images": images,
    "device": device
}
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 1, dtype family int, shape (2,) — shape (2,) 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

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, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 2 later) and 2 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
cosine_scheduler 26 other papers
22 dated earlier, 3 later, 1 same year, order not determinable
1905.02244 (2019-05)
Searching for MobileNetV3
136810051 (2022 (year only))
AMixer: Adaptive Weight Mixing for Self-Attention Free Vision Transformers
136860514 (2022 (year only))
SLIP: Self-supervision meets Language-Image Pre-training
and 23 more not named here
get_bn 4 other papers
3 dated earlier, 1 later
2203.06717 (2022-03)
Scaling Up Your Kernels to 31x31: Revisiting Large Kernel Design in CNNs
2304.02330 (2023-04)
SMPConv: Self-moving Point Representations for Continuous Convolution
2305.19412 (2023-05)
Are Large Kernels Better Teachers than Transformers for ConvNets?
and 1 more not named here
clean_accuracy 1 other paper
1 dated earlier
2310.12793 (2023-10)
OODRobustBench: a Benchmark and Large-Scale Analysis of Adversarial Robustness under Distribution Shift

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

Robustness is a vital aspect to consider when deploying deep learning models into the wild. Numerous studies have been dedicated to the study of the robustness of vision transformers (ViTs), which have dominated as the mainstream backbone choice for vision tasks since the dawn of 2020s. Recently, some large kernel convnets make a comeback with impressive performance and efficiency. However, it still remains unclear whether large kernel networks are robust and the attribution of their robustness. In this paper, we first conduct a comprehensive evaluation of large kernel convnets' robustness and their differences from typical small kernel counterparts and ViTs on six diverse robustness benchmark datasets. Then to analyze the underlying factors behind their strong robustness, we design experiments from both quantitative and qualitative perspectives to reveal large kernel convnets' intriguing properties that are completely different from typical convnets. Our experiments demonstrate for the first time that pure CNNs can achieve exceptional robustness comparable or even superior to that of ViTs. Our analysis on occlusion invariance, kernel attention patterns and frequency characteristics provide novel insights into the source of robustness.

For agents

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

get_harvested_code_for_paper("2407.08972")
get_code_for_paper("2407.08972")
have("2407.08972")

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

curl https://syntology.ai/api/ran/2407.08972.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.08972.svg)](https://syntology.ai/paper/2407.08972)

Connect an agent — have() is free.