SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.13378 · ECCV · 2022

Identifying Hard Noise in Long-Tailed Sample Distribution

Xian-Sheng Hua, Hanwang Zhang, Kaihua Tang, Xuanyu Yi, Joo-Hwee Lim

arXiv · PDF · Open in the Atlas

Code that ran

9 of 11 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 7 have no contract verdict; 2 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/2207.13378.json · badge: /api/ran/2207.13378.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
yxymessi/H2E-Framework canonical (classifier label)
classifier basis not recorded on the link
9 ran · 2 no recorded run
FunctionStatusWhere it lives
call_bn Ran
ran; our contract draft was wrong, not the code
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/model.py
pointer only (licence: MIT); the stored copy get_code serves, from yeachan-kr/pytorch-coteaching, is pointer only (licence: NONE) · get_code("7937f88152d63f61")
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 yeachan-kr/pytorch-coteaching, 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.

count_dataset Ran
ran on a synthesized input
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/loss.py
code served (permissive licence) · get_code("6c1b73e88ea70693")
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 1, dtype family int, shape (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 torch
import numpy as np
from torch.utils.data import TensorDataset, DataLoader

torch.manual_seed(0)

# Create a small dataset with targets
x = torch.randn(6, 4)
targets = [0, 1, 0, 2, 1, 2]
dataset = TensorDataset(x, torch.tensor(targets))

# Create a data loader
train_loader = DataLoader(dataset, batch_size=2)

# Verify the loader has the dataset.targets attribute needed by the function
train_loader.dataset.targets = targets

ARGS = {
    "train_loader": train_loader
}
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 ndarray, ndim 1, dtype family int, shape (3,) — shape (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
create_classifer Ran
ran on a synthesized input
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/model.py
code served (permissive licence) · get_code("e6571d10bfbf14f3")
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 ClassifierLWS. 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 = {
    "feat_dim": 512,
    "num_classes": 100,
}
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 ClassifierLWS — a ClassifierLWS output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ClassifierLWS 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
create_model Ran
ran on a synthesized input
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/model.py
code served (permissive licence) · get_code("5cbc579cb1e2d98a")
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
import torchvision.models as models

torch.manual_seed(0)

ARGS = {
    'm_type': 'resnet101'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — 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
loss_coteaching Ran
ran on a synthesized input
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/loss.py
code served (permissive licence) · get_code("098713e67b736d08")
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, 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 torch.nn.functional as F
import numpy as np

torch.manual_seed(0)

# Set up parameters for loss_coteaching function
batch_size = 2
num_classes = 4

# y_1: logits from first model (batch_size, num_classes)
y_1 = torch.randn(batch_size, num_classes)

# y_2: logits from second model (batch_size, num_classes)
y_2 = torch.randn(batch_size, num_classes)

# t: target labels (batch_size,)
t = torch.tensor([0, 2], dtype=torch.long)

# forget_rate: fraction of samples to forget (between 0 and 1)
forget_rate = 0.5

# ind: indices parameter (not used in function body, but required)
ind = torch.tensor([0, 1], dtype=torch.long)

ARGS = {
    "y_1": y_1,
    "y_2": y_2,
    "t": t,
    "forget_rate": forget_rate,
    "ind": ind
}
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 tuple, ndim 1, dtype family float — a tuple 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
mixup_criterion Ran
ran on a synthesized input
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/utils.py
code served (permissive licence) · get_code("79edc1957c1e3747")
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)

# y_a and y_b are target tensors for mixup
# They should be class indices for classification
y_a = torch.tensor([0, 1], dtype=torch.long)
y_b = torch.tensor([1, 0], dtype=torch.long)

# lam is the mixup coefficient (between 0 and 1)
lam = 0.7

ARGS = {
    "y_a": y_a,
    "y_b": y_b,
    "lam": lam,
}
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
mixup_data Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from boone891214/sanity-check-LTH, is pointer only (licence: NONE) · get_code("fd81207449c239dd")
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 boone891214/sanity-check-LTH, 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.

mixup_data_weight Ran
ran on a synthesized input
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/utils.py
code served (permissive licence) · get_code("aad6a1787762c8ee")
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. 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

torch.manual_seed(0)

# Construct arguments for mixup_data_weight
batch_size = 2
num_channels = 3
height = 4
width = 5

x = torch.randn(batch_size, num_channels, height, width)
y = torch.tensor([0, 1])
score_list = [0.6, 0.4]

ARGS = {
    'x': x,
    'y': y,
    'score_list': score_list
}
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 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
source_import Ran
ran on a synthesized input
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/stage1.py
code served (permissive licence) · get_code("07cca9631fe4f26a")
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 module. 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 importlib
import tempfile
import os

# Create a temporary Python module file to import
temp_dir = tempfile.mkdtemp()
module_file = os.path.join(temp_dir, "test_module.py")

with open(module_file, "w") as f:
    f.write("""
def hello_world():
    return "Hello from imported module"

class TestClass:
    def __init__(self):
        self.value = 42
""")

ARGS = {
    "file_path": module_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 module — a module output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a module 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
compute_adjustment Not yet run
unverified (harvested, no recorded run)
yxymessi/H2E-Framework/eccv_github/noise_longtail/code/stage1.py
code served (permissive licence) · get_code("2d468638b40f3381")
How it ran

no invocation recorded.

read_data Not yet run
unverified (harvested, no recorded run)
yxymessi/H2E-Framework/eccv_github/noise_longtail/construct_data/get_dataset.py
code served (permissive licence) · get_code("5007bbe5663c0042")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 3 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
source_import 4 other papers
3 dated earlier, 1 later
2004.03706 (2020-04)
Long-Tailed Recognition Using Class-Balanced Experts
2007.11978 (2020-07)
The Devil is in Classification: A Simple Framework for Long-tail Object Detection and Instance Segmentation
2111.13998 (2021-11)
Targeted Supervised Contrastive Learning for Long-Tailed Recognition
and 1 more not named here
call_bn 3 other papers
2 dated earlier, 1 later
1804.06872 (2018-04)
Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels
2106.10891 (2021-06)
Open-set Label Noise Can Improve Robustness Against Inherent Label Noise
2311.01196 (2023-11)
Combating Bilateral Edge Noise for Robust Link Prediction
mixup_data 3 other papers
2 dated earlier, 1 later
2107.00166 (2021-07)
Sanity Checks for Lottery Tickets: Does Your Winning Ticket Really Win the Jackpot?
2110.14032 (2021-10)
MEST: Accurate and Fast Memory-Economic Sparse Training Framework on the Edge
2209.09476 (2022-09)
SparCL: Sparse Continual Learning on the Edge
mixup_criterion 1 other paper
1 dated earlier
2002.06815 (2020-02)
Class-Imbalanced Semi-Supervised Learning

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

Conventional de-noising methods rely on the assumption that all samples are independent and identically distributed, so the resultant classifier, though disturbed by noise, can still easily identify the noises as the outliers of training distribution. However, the assumption is unrealistic in large-scale data that is inevitably long-tailed. Such imbalanced training data makes a classifier less discriminative for the tail classes, whose previously "easy" noises are now turned into "hard" ones-they are almost as outliers as the clean tail samples. We introduce this new challenge as Noisy Long-Tailed Classification (NLT). Not surprisingly, we find that most de-noising methods fail to identify the hard noises, resulting in significant performance drop on the three proposed NLT benchmarks: ImageNet-NLT, Animal10-NLT, and Food101-NLT. To this end, we design an iterative noisy learning framework called Hard-to-Easy (H2E). Our bootstrapping philosophy is to first learn a classifier as noise identifier invariant to the class and context distributional changes, reducing "hard" noises to "easy" ones, whose removal further improves the invariance. Experimental results show that our H2E outperforms state-of-the-art de-noising methods and their ablations on long-tailed settings while maintaining a stable performance on the conventional balanced settings. Datasets and codes are available at https://github.com/yxymessi/H2E-Framework.

For agents

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

get_harvested_code_for_paper("2207.13378")
get_code_for_paper("2207.13378")
have("2207.13378")

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

curl https://syntology.ai/api/ran/2207.13378.json

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

[![Syntology run record](https://syntology.ai/api/ran/2207.13378.svg)](https://syntology.ai/paper/2207.13378)

Connect an agent — have() is free.