SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.13282 · 2022

On the Eigenvalues of Global Covariance Pooling for Fine-grained Visual Recognition

arXiv · PDF · Open in the Atlas

Code that ran

12 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; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 9 failures of our instrument; 5 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/2205.13282.json · badge: /api/ran/2205.13282.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
KingJamesSong/DifferentiableSVD canonical (classifier label)
classifier basis not recorded on the link
12 ran · 5 no recorded run
FunctionStatusWhere it lives
alexnet Ran
ran; our contract draft was wrong, not the code
KingJamesSong/DifferentiableSVD/src/network/alexnet.py
code served (permissive licence) · get_code("2f5db20279adcab2")
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 AlexNet. 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 = {
    'pretrained': 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 AlexNet — a AlexNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a AlexNet 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
center_crop_with_flip Ran
ran on a synthesized input
KingJamesSong/DifferentiableSVD/imagepreprocess.py
code served (permissive licence) · get_code("e1bba953865399c4")
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 4, 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
from PIL import Image
import numpy as np

torch.manual_seed(0)

# Create a small PIL Image for testing
# The function expects a PIL Image or tensor that works with F.center_crop, F.vflip, F.hflip
img_array = np.random.randint(0, 256, (64, 48, 3), dtype=np.uint8)
img = Image.fromarray(img_array)

# Define the crop size
size = (32, 24)

ARGS = {
    "img": img,
    "size": size,
    "vertical_flip": False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 4, dtype family int — 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
conv1x1 Ran
ran; our contract draft was wrong, not the code
KingJamesSong/DifferentiableSVD/src/network/resnet.py
code served (permissive licence) · get_code("d9def42110729a85")
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) and returned Conv2d. 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 conv1x1 takes in_planes, out_planes, and stride as parameters
# It returns an nn.Conv2d layer configured for 1x1 convolution
# We need to provide valid integer arguments for these parameters

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1
}
Setup block drafted by
a language model, Claude Haiku 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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
KingJamesSong/DifferentiableSVD/src/network/resnet.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from lixiaotong97/DSU, is pointer only (licence: NONE) · get_code("fac5364e2f53c6db")
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 lixiaotong97/DSU, 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.

decode_params Ran
ran on a synthesized input
KingJamesSong/DifferentiableSVD/functions.py
code served (permissive licence) · get_code("e9ae7b09751b4912")
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

torch.manual_seed(0)

def decode_params(input_params):
    params = input_params[0]
    out_params = []
    _start = 0
    _end = 0
    for i in range(len(params)):
        if params[i] == ',':
            out_params.append(float(params[_start:_end]))
            _start = _end + 1
        _end += 1
    out_params.append(float(params[_start:_end]))
    return out_params

ARGS = {
    "input_params": ["1.5,2.3,3.7,4.2"]
}
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 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
densenet169 Ran
ran; our contract draft was wrong, not the code
KingJamesSong/DifferentiableSVD/src/network/densenet.py
code served (permissive licence) · get_code("4983bf3da7e3bc95")
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 DenseNet. 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 = {
    'pretrained': 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 DenseNet — a DenseNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DenseNet 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
densenet201 Ran
ran; our contract draft was wrong, not the code
KingJamesSong/DifferentiableSVD/src/network/densenet.py
code served (permissive licence) · get_code("c26ebe510580f5fa")
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 DenseNet. 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 = {
    'pretrained': 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 DenseNet — a DenseNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DenseNet 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
make_layers Ran
ran; our contract draft was wrong, not the code
KingJamesSong/DifferentiableSVD/src/network/mpncovvgg.py
code served (permissive licence) · get_code("ac62432dc5134b0d")
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 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 takes two parameters: cfg and batch_norm
# cfg should be a list/sequence of either integers (channel counts) or 'M' (maxpool marker)
# batch_norm is a boolean flag

# Using one of the predefined configurations from the module-level cfg dict
cfg_list = [64, 'M', 128, 'M', 256, 256, 'M', 512, 512, 'M', 512, 512, 'M']

ARGS = {
    'cfg': cfg_list,
    'batch_norm': 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 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
preprocess_strategy Ran
ran on a synthesized input
KingJamesSong/DifferentiableSVD/imagepreprocess.py
code served (permissive licence) · get_code("7d5e18eb2768dd94")
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 numbers
from torchvision import transforms
from torchvision.transforms import functional as F

torch.manual_seed(0)

# Test with a simple dataset string
ARGS = {
    "dataset": "CUB"
}
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 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
resnet18 Ran
ran; our contract draft was wrong, not the code
KingJamesSong/DifferentiableSVD/src/network/resnet.py
code served (permissive licence) · get_code("bc9961dacb0c6e52")
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 = {
    'pretrained': 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 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
vgg11 Ran
ran; our contract draft was wrong, not the code
KingJamesSong/DifferentiableSVD/src/network/vgg.py
code served (permissive licence) · get_code("9ec7dbb47bd1938b")
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 VGG. 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 vgg11 function takes the following parameters:
# - pretrained: bool (default False)
# - **kwargs: additional keyword arguments passed to VGG constructor
#   The VGG constructor accepts: num_classes, init_weights

ARGS = {
    'pretrained': False,
    'num_classes': 1000,
    'init_weights': 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 VGG — a VGG output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a VGG 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
vgg11_bn Ran
ran; our contract draft was wrong, not the code
KingJamesSong/DifferentiableSVD/src/network/vgg.py
code served (permissive licence) · get_code("57ac34428a02311d")
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 VGG. 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 = {
    'pretrained': 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 VGG — a VGG output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a VGG 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
densenet121 Not yet run
unverified (harvested, no recorded run)
KingJamesSong/DifferentiableSVD/src/network/densenet.py
code served (permissive licence) · get_code("10d70cd9de88accd")
How it ran

no invocation recorded.

get_basemodel Not yet run
unverified (harvested, no recorded run)
KingJamesSong/DifferentiableSVD/src/network/base.py
code served (permissive licence) · get_code("b3aff4f65217439f")
How it ran

no invocation recorded.

inception_v3 Not yet run
unverified (harvested, no recorded run)
KingJamesSong/DifferentiableSVD/src/network/inception.py
code served (permissive licence) · get_code("99f547d08bd82760")
How it ran

no invocation recorded.

mpncovresnet101 Not yet run
unverified (harvested, no recorded run)
KingJamesSong/DifferentiableSVD/src/network/mpncovresnet.py
code served (permissive licence) · get_code("821722d0a708c2b9")
How it ran

no invocation recorded.

mpncovresnet50 Not yet run
unverified (harvested, no recorded run)
KingJamesSong/DifferentiableSVD/src/network/mpncovresnet.py
code served (permissive licence) · get_code("da56958c53b9740c")
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, 13 are also present in other papers whose code we hold (13 with at least one paper dated earlier, 3 the same month, 11 later) and 4 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
389 dated earlier, 4 the same month, 282 later, 11 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
conv1x1 431 other papers
189 dated earlier, 2 the same month, 228 later, 11 same year, order not determinable, 1 undated · 38 only as an assembly brick
1409.1556 (2014-09)
Very Deep Convolutional Networks for Large-Scale Image Recognition
1506.01186 (2015-06)
Cyclical Learning Rates for Training Neural Networks
1507.05717 (2015-07)
An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
and 428 more not named here
resnet18 89 other papers
49 dated earlier, 38 later, 2 same year, order not determinable
1705.09759 (2017-05)
CASENet: Deep Category-Aware Semantic Edge Detection
1801.07698 (2018-01)
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
1803.00839 (2018-03)
Pose-Robust Face Recognition via Deep Residual Equivariant Mapping
and 86 more not named here
make_layers 59 other papers
30 dated earlier, 1 the same month, 27 later, 1 same year, order not determinable
1411.4038 (2014-11)
Fully Convolutional Networks for Semantic Segmentation
1511.00561 (2015-11)
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
1511.05950 (2015-11)
Staleness-aware Async-SGD for Distributed Deep Learning
and 56 more not named here
vgg11 10 other papers
4 dated earlier, 5 later, 1 same year, order not determinable
1610.02391 (2016-10)
Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
1904.03441 (2019-04)
Iterative Normalization: Beyond Standardization towards Efficient Whitening
2108.05293 (2021-08)
Few-Shot Segmentation with Global and Local Contrastive Learning
and 7 more not named here
vgg11_bn 9 other papers
3 dated earlier, 5 later, 1 same year, order not determinable
1904.03441 (2019-04)
Iterative Normalization: Beyond Standardization towards Efficient Whitening
2108.05293 (2021-08)
Few-Shot Segmentation with Global and Local Contrastive Learning
136780474 (2022 (year only))
Hourglass Attention Network for Image Inpainting
and 6 more not named here
densenet121 6 other papers
5 dated earlier, 1 later
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1812.00332 (2018-12)
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
1904.03441 (2019-04)
Iterative Normalization: Beyond Standardization towards Efficient Whitening
and 3 more not named here
densenet169 6 other papers
5 dated earlier, 1 later
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1812.00332 (2018-12)
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
1904.03441 (2019-04)
Iterative Normalization: Beyond Standardization towards Efficient Whitening
and 3 more not named here
densenet201 6 other papers
5 dated earlier, 1 later
1608.06993 (2016-08)
Densely Connected Convolutional Networks
1812.00332 (2018-12)
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
1904.03441 (2019-04)
Iterative Normalization: Beyond Standardization towards Efficient Whitening
and 3 more not named here
inception_v3 6 other papers
5 dated earlier, 1 later
1812.00332 (2018-12)
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
1904.01160 (2019-04)
Curls & Whey: Boosting Black-Box Adversarial Attacks
1904.03441 (2019-04)
Iterative Normalization: Beyond Standardization towards Efficient Whitening
and 3 more not named here
alexnet 5 other papers
4 dated earlier, 1 later
1811.06965 (2018-11)
GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism
1812.00332 (2018-12)
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
1904.03441 (2019-04)
Iterative Normalization: Beyond Standardization towards Efficient Whitening
and 2 more not named here
decode_params 1 other paper
1 dated earlier
2006.11280 (2020-06)
Self-PU: Self Boosted and Calibrated Positive-Unlabeled Training
mpncovresnet101 1 other paper
1 dated earlier
1904.06836 (2019-04)
Deep CNNs Meet Global Covariance Pooling: Better Representation and Generalization

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

The Fine-Grained Visual Categorization (FGVC) is challenging because the subtle inter-class variations are difficult to be captured. One notable research line uses the Global Covariance Pooling (GCP) layer to learn powerful representations with second-order statistics, which can effectively model inter-class differences. In our previous conference paper, we show that truncating small eigenvalues of the GCP covariance can attain smoother gradient and improve the performance on large-scale benchmarks. However, on fine-grained datasets, truncating the small eigenvalues would make the model fail to converge. This observation contradicts the common assumption that the small eigenvalues merely correspond to the noisy and unimportant information. Consequently, ignoring them should have little influence on the performance. To diagnose this peculiar behavior, we propose two attribution methods whose visualizations demonstrate that the seemingly unimportant small eigenvalues are crucial as they are in charge of extracting the discriminative class-specific features. Inspired by this observation, we propose a network branch dedicated to magnifying the importance of small eigenvalues. Without introducing any additional parameters, this branch simply amplifies the small eigenvalues and achieves state-of-the-art performances of GCP methods on three fine-grained benchmarks. Furthermore, the performance is also competitive against other FGVC approaches on larger datasets. Code is available at \href{https://github.com/KingJamesSong/DifferentiableSVD}{https://github.com/KingJamesSong/DifferentiableSVD}.

For agents

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

get_harvested_code_for_paper("2205.13282")
get_code_for_paper("2205.13282")
have("2205.13282")

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

curl https://syntology.ai/api/ran/2205.13282.json

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

[![Syntology run record](https://syntology.ai/api/ran/2205.13282.svg)](https://syntology.ai/paper/2205.13282)

Connect an agent — have() is free.