SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.16802 · ECCV · 2024

Distribution-Aware Robust Learning from Long-Tailed Data with Noisy Labels

Jun Choi, Jae Baik, In Yoon, Kun Kim

arXiv · PDF · Open in the Atlas

Code that ran

14 of 17 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; 3 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 11 failures of our instrument; 3 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/2407.16802.json · badge: /api/ran/2407.16802.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
JaesoonBaik1213/DaSC canonical (classifier label)
classifier basis: census
14 ran · 3 no recorded run
FunctionStatusWhere it lives
Preact_ResNet18 Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/models/PreResNet.py
code served (permissive licence) · get_code("edb2f37e016d955a")
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)

class Args:
    def __init__(self):
        self.hidden_dim = 8
        self.dataset = 'cifar10'
        self.num_classes = 10

ARGS = {
    'num_classes': 10,
    'args': Args()
}
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
ResNet18 Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/models/PreResNet.py
code served (permissive licence) · get_code("621298c338df1815")
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)

class ArgsStandIn:
    def __init__(self):
        self.hidden_dim = 8
        self.dataset = 'cifar10'
        self.num_classes = 10

ARGS = {
    'num_classes': 10,
    'args': ArgsStandIn()
}
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
conv3x3 Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/models/PreResNet.py
pointer only (licence: MIT); the stored copy get_code serves, from amerch/CIFAR100-Training, is pointer only (licence: NONE) · get_code("583f9780bdd00a45")
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 amerch/CIFAR100-Training, 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_imbalance_ratios Ran
ran, honoured the contract we drafted
JaesoonBaik1213/DaSC/datas/dataloader_red_mini_imagenet.py
code served (permissive licence) · get_code("d6b493f24b80f8d3")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np

torch.manual_seed(0)

ARGS = {
    "imb_factor": 0.01,
    "cls_num": 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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_imbalance_ratios_1 Ran
ran, honoured the contract we drafted
JaesoonBaik1213/DaSC/datas/data_utils_N.py
code served (permissive licence) · get_code("23afa08e1cbdacce")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np

# Analyze the function:
# - imb_factor: a numeric value (likely float) used in exponentiation
# - cls_num: an integer representing the number of classes
# The function computes imbalance_ratios by iterating through cls_num
# and applying the formula: ratio = imb_factor ** (cls_idx / (cls_num - 1.0))
# Returns a list of ratios

ARGS = {
    'imb_factor': 0.01,
    'cls_num': 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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_img_num_per_cls Ran
ran, honoured the contract we drafted
JaesoonBaik1213/DaSC/datas/dataloader_red_mini_imagenet.py
code served (permissive licence) · get_code("b419771b07d6b313")
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 list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np

# Setup for get_img_num_per_cls function
# Reading the function body:
# - img_num: used as img_max (int)
# - cls_num: used in range() and get_imbalance_ratios (int)
# - imb_factor: passed to get_imbalance_ratios, can be None or float
# - num_meta: not used in the function body

torch.manual_seed(0)

ARGS = {
    "img_num": 5000,
    "cls_num": 10,
    "imb_factor": 0.01,
    "num_meta": None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
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
sample_dataset Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/datas/dataloader_red_mini_imagenet.py
code served (permissive licence) · get_code("04303b0f4abc9b27")
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 list, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Create sample data
# train_data: list of indices (0 to 9)
train_data = list(range(10))

# train_label: list of class labels for each data point
# 5 samples of class 0, 5 samples of class 1
train_label = [0, 0, 0, 0, 0, 1, 1, 1, 1, 1]

# img_num_list: number of images per class
# For class 0: keep/delete 2, for class 1: keep/delete 3
img_num_list = [2, 3]

# num_classes: total number of classes
num_classes = 2

# kind: either 'delete' or other (determines which samples to remove)
kind = 'delete'

ARGS = {
    'train_data': train_data,
    'train_label': train_label,
    'img_num_list': img_num_list,
    'num_classes': num_classes,
    'kind': kind
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family int — a list output has no array shape of its own; none is stated
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
sample_dataset_1 Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/datas/data_utils.py
code served (permissive licence) · get_code("a399ccf88d7166c6")
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 tuple. 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 numpy as np

# Create small dummy data
np.random.seed(0)

# Parameters for the function
num_classes = 3
num_samples = 12

# Create training data (12 samples, 4 features)
train_data = np.random.rand(num_samples, 4).astype(np.float32)

# Create training labels (3 classes, with 4 samples per class)
train_label = np.array([0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2])

# Create noise labels (same shape as train_label)
noise_label = np.array([0, 1, 0, 0, 1, 1, 2, 1, 2, 2, 1, 2])

# img_num_list specifies how many samples to keep/delete per class
# For 3 classes, we specify how many to delete (or keep) from each
img_num_list = [1, 2, 1]  # Delete 1 from class 0, 2 from class 1, 1 from class 2

# kind parameter specifies whether to delete or keep the specified indices
kind = 'delete'

ARGS = {
    'train_data': train_data,
    'train_label': train_label,
    'noise_label': noise_label,
    'img_num_list': img_num_list,
    'num_classes': num_classes,
    'kind': kind
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
sample_dataset_1 Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/datas/data_utils_N.py
code served (permissive licence) · get_code("082b83cd93beed20")
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 tuple. 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 copy
import numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Create small sample data
num_samples = 10
num_classes = 3
num_features = 4

# Create training data as numpy array (num_samples, num_features)
train_data = np.random.randn(num_samples, num_features).astype(np.float32)

# Create training labels - distribute samples across classes
train_label = np.array([0, 0, 0, 1, 1, 1, 2, 2, 2, 0], dtype=np.int64)

# Create img_num_list specifying how many samples per class to keep/delete
# For 3 classes, specify count for each
img_num_list = np.array([2, 2, 2], dtype=np.int64)

# Number of classes
num_classes = 3

# Kind parameter - either 'delete' or 'keep'
kind = 'delete'

# Optional human_label parameter (None or custom labels)
human_label = None

ARGS = {
    'train_data': train_data,
    'train_label': train_label,
    'img_num_list': img_num_list,
    'num_classes': num_classes,
    'kind': kind,
    'human_label': human_label
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
set_device Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/settings/configs.py
code served (permissive licence) · get_code("1edaa58918fd8481")
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 device. 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

class Args:
    def __init__(self, no_cuda=False):
        self.no_cuda = no_cuda

ARGS = {
    'args': Args(no_cuda=True)
}
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 device — a device output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a device 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
set_tensor Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/utils/util.py
code served (permissive licence) · get_code("56df0904ca7bc6a3")
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

torch.manual_seed(0)

# Create a small tensor (batch=2, dim=4)
tensor_var = torch.randn(2, 4)

# Define ARGS dict with all parameters
ARGS = {
    "tensor_var": tensor_var,
    "boolen": True,
    "device": "cpu"
}
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
to_var Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/models/resnet.py
code served (permissive licence) · get_code("0d26b5abf61353f9")
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. 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, 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)

# The function takes a tensor x and an optional requires_grad boolean
# x should be a torch.Tensor
# requires_grad is a boolean (default True)

x = torch.randn(2, 4)

ARGS = {
    "x": x,
    "requires_grad": True,
}
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
unpickle Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/datas/data_utils.py
code served (permissive licence) · get_code("5543609031c0e900")
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 dict. 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 pickle
import tempfile
import os

# Create a temporary pickle file for testing
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_pickle.pkl")

# Create sample data to pickle
sample_data = {
    'key1': 'value1',
    'key2': [1, 2, 3],
    'key3': {'nested': 'dict'}
}

# Write the sample data to a pickle file
with open(temp_file, 'wb') as f:
    pickle.dump(sample_data, f)

ARGS = {
    'file': temp_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
unpickle Ran
ran; our contract draft was wrong, not the code
JaesoonBaik1213/DaSC/datas/data_utils_N.py
code served (permissive licence) · get_code("ef49e82a7403eee1")
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 dict. 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 _pickle as cPickle
import tempfile
import os

# Create a temporary pickle file with sample data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.pkl")

# Create some sample data and pickle it
sample_data = {
    'key1': 'value1',
    'key2': [1, 2, 3],
    'key3': {'nested': 'dict'}
}

with open(temp_file, 'wb') as f:
    cPickle.dump(sample_data, f)

ARGS = {
    'file': temp_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
accuracy Not yet run
unverified (harvested, no recorded run)
JaesoonBaik1213/DaSC/models/metric.py
code served (permissive licence) · get_code("f0c9a29156911331")
How it ran

no invocation recorded.

balanced_softmax_loss Not yet run
unverified (harvested, no recorded run)
JaesoonBaik1213/DaSC/experiment/train_cifar_ssl.py
code served (permissive licence) · get_code("fd714bbda15c3c95")
How it ran

no invocation recorded.

balanced_softmax_loss_semi Not yet run
unverified (harvested, no recorded run)
JaesoonBaik1213/DaSC/experiment/train_cifar_ssl.py
code served (permissive licence) · get_code("15a4b28e19a85a2a")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 2 the same month, 2 later) and 14 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
conv3x3 686 other papers
603 dated earlier, 14 the same month, 52 later, 17 same year, order not determinable · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
accuracy 71 other papers
68 dated earlier, 3 later · 2 only as an assembly brick
1505.05424 (2015-05)
Weight Uncertainty in Neural Networks
1705.07204 (2017-05)
Ensemble Adversarial Training: Attacks and Defenses
1802.10399 (2018-02)
Compressing Neural Networks using the Variational Information Bottleneck
and 68 more not named here
unpickle 7 other papers
6 dated earlier, 1 the same month
2105.14083 (2021-05)
Rethinking Noisy Label Models: Labeler-Dependent Noise with Adversarial Awareness
2203.14542 (2022-03)
UNICON: Combating Label Noise Through Uniform Selection and Contrastive Learning
2207.14476 (2022-07)
Centrality and Consistency: Two-Stage Clean Samples Identification for Learning with Instance-Dependent Noisy Labels
and 4 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

Deep neural networks have demonstrated remarkable advancements in various fields using large, well-annotated datasets. However, real-world data often exhibit long-tailed distributions and label noise, significantly degrading generalization performance. Recent studies addressing these issues have focused on noisy sample selection methods that estimate the centroid of each class based on high-confidence samples within each target class. The performance of these methods is limited because they use only the training samples within each class for class centroid estimation, making the quality of centroids susceptible to longtailed distributions and noisy labels. In this study, we present a robust training framework called Distribution-aware Sample Selection and Contrastive Learning (DaSC). Specifically, DaSC introduces a Distributionaware Class Centroid Estimation (DaCC) to generate enhanced class centroids. DaCC performs weighted averaging of the features from all samples, with weights determined based on model predictions. Additionally, we propose a confidence-aware contrastive learning strategy to obtain balanced and robust representations. The training samples are categorized into high-confidence and low-confidence samples. Our method then applies Semi-supervised Balanced Contrastive Loss (SBCL) using high-confidence samples, leveraging reliable label information to mitigate class bias. For the low-confidence samples, our method computes Mixup-enhanced Instance Discrimination Loss (MIDL) to improve their representations in a self-supervised manner. Our experimental results on CIFAR and real-world noisy-label datasets demonstrate the superior performance of the proposed DaSC compared to previous approaches. Our code is available at https://github.com/JaesoonBaik1213/DaSC.

For agents

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

get_harvested_code_for_paper("2407.16802")
get_code_for_paper("2407.16802")
have("2407.16802")

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

curl https://syntology.ai/api/ran/2407.16802.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.16802.svg)](https://syntology.ai/paper/2407.16802)

Connect an agent — have() is free.