SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2003.14297 · 2020

Generative Latent Implicit Conditional Optimization when Learning from Small Sample

arXiv · PDF · Open in the Atlas

Code that ran

10 of 16 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, 2 violated it and 8 have no contract verdict; 0 failures of our instrument; 6 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/2003.14297.json · badge: /api/ran/2003.14297.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
IdanAzuri/glico-learning-small-sample canonical (classifier label)
classifier basis: census
10 ran · 6 no recorded run
FunctionStatusWhere it lives
fewshot_setting Ran
ran on a synthesized input
IdanAzuri/glico-learning-small-sample/glico_model/evaluation.py
code served (permissive licence) · get_code("a86a74fe9070c8ff")
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 int, ndim 0, dtype family int, 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 numpy as np

torch.manual_seed(0)

# Analyze the function:
# - samples_per_class: int, used in division and int() conversion
# - im_size: int, used in equality comparison (== 32)

ARGS = {
    'samples_per_class': 5,
    'im_size': 32
}
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 int, ndim 0, dtype family int, 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
has_file_allowed_extension Ran
ran, violated the contract we drafted
IdanAzuri/glico-learning-small-sample/glico_model/imagent_folder_loader.py
code served (permissive licence) · get_code("46a47194a5a214b5")
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 bool, 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 numpy as np

# Construct arguments for has_file_allowed_extension
filename = "example_image.jpg"
extensions = (".jpg", ".png", ".gif")

ARGS = {
    "filename": filename,
    "extensions": extensions,
}
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 bool, 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
is_image_file Ran
ran, violated the contract we drafted
IdanAzuri/glico-learning-small-sample/glico_model/imagent_folder_loader.py
code served (permissive licence) · get_code("3ed19559ece460a6")
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 bool, 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 numpy as np

# Define test arguments for is_image_file function
ARGS = {
    "filename": "test_image.jpg"
}
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 bool, 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
lerp_mat Ran
ran on a synthesized input
IdanAzuri/glico-learning-small-sample/glico_model/interpolate.py
code served (permissive licence) · get_code("cd853a74f6084460")
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 (10, 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
import numpy as np

torch.manual_seed(0)

# The function lerp_mat takes:
# - start: a tensor (used in torch.lerp)
# - end: a tensor (used in torch.lerp)
# - n_steps: an integer (used in np.linspace)

# Create small tensors for start and end
start = torch.randn(2, 4)  # shape (2, 4)
end = torch.randn(2, 4)    # shape (2, 4)

# Create n_steps as an integer
n_steps = 5

ARGS = {
    "start": start,
    "end": end,
    "n_steps": n_steps,
}
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 Tensor, ndim 2, dtype family float, shape (10, 4) — shape (10, 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
make_dataset Ran
ran on a synthesized input
IdanAzuri/glico-learning-small-sample/glico_model/imagent_folder_loader.py
code served (permissive licence) · get_code("f846a37597cc33b1")
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 2, 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 os
import tempfile
import shutil

# Create a temporary directory structure for testing
test_dir = tempfile.mkdtemp()

# Create class subdirectories with some files
class_names = ['class_a', 'class_b']
for class_name in class_names:
    class_dir = os.path.join(test_dir, class_name)
    os.makedirs(class_dir, exist_ok=True)
    # Create some test files with valid extensions
    for i in range(3):
        with open(os.path.join(class_dir, f'image_{i}.jpg'), 'w') as f:
            f.write('dummy')
        with open(os.path.join(class_dir, f'image_{i}.txt'), 'w') as f:
            f.write('dummy')

# Create class_to_idx mapping
class_to_idx = {
    'class_a': 0,
    'class_b': 1
}

# Define valid extensions
extensions = ('.jpg', '.jpeg', '.png')

# Create an ignore_samples_by_path dict (empty for this test)
ignore_samples_by_path = {}

ARGS = {
    'dir': test_dir,
    'class_to_idx': class_to_idx,
    'extensions': extensions,
    'is_valid_file': None,
    'class_cap': None,
    'ignore_samples_by_path': ignore_samples_by_path
}
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 list, ndim 2, 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
sample_from_iter Ran
ran on a synthesized input
IdanAzuri/glico-learning-small-sample/glico_model/model.py
code served (permissive licence) · get_code("247e9650fd4374b4")
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 2, 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
import itertools
import random

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

# Create a simple iterable of tuples containing numeric values
iterable = [
    (1.5, 2.3, 3.7),
    (4.1, 5.9, 6.2),
    (7.8, 8.4, 9.1),
    (10.5, 11.2, 12.9),
    (13.3, 14.6, 15.8)
]

# Define arguments for the function
ARGS = {
    'iterable': iterable,
    'k': 3
}
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 2, 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
slerp Ran
ran on a synthesized input
IdanAzuri/glico-learning-small-sample/glico_model/interpolate.py
code served (permissive licence) · get_code("b31461d93847c383")
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 float, shape (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 numpy as np
from math import sin

# Construct arguments for slerp function
# The function performs spherical linear interpolation between two vectors

# val: interpolation parameter (scalar between 0 and 1)
val = 0.5

# low: first vector (should be normalized or at least non-zero)
low = np.array([1.0, 2.0, 3.0, 4.0])

# high: second vector (should be normalized or at least non-zero)
high = np.array([4.0, 3.0, 2.0, 1.0])

ARGS = {
    "val": val,
    "low": low,
    "high": high,
}
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 ndarray, ndim 1, dtype family float, shape (4,) — shape (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
slerp_torch Ran
ran on a synthesized input
IdanAzuri/glico-learning-small-sample/glico_model/interpolate.py
code served (permissive licence) · get_code("1fdd932b4ae04228")
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 small tensors with batch size 2 and dimension 4
# low and high are the interpolation endpoints
low = torch.randn(2, 4)
high = torch.randn(2, 4)

# val is a scalar interpolation parameter between 0 and 1
val = 0.5

ARGS = {
    "val": val,
    "low": low,
    "high": high
}
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
train_val_split Ran
ran on a synthesized input
IdanAzuri/glico-learning-small-sample/glico_model/cifar10.py
code served (permissive licence) · get_code("a727540bc318bdd2")
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

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

# Create labels for a 10-class dataset
# We need enough samples per class to satisfy both n_labeled_per_class and n_unlabeled_per_class
# Let's create a dataset with 100 samples per class (1000 total)
labels = np.concatenate([np.full(100, i) for i in range(10)])
np.random.shuffle(labels)

# Define the arguments
ARGS = {
    "labels": labels,
    "n_labeled_per_class": 10,
    "n_unlabled_per_class": 20
}
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
train_val_split Ran
ran on a synthesized input
IdanAzuri/glico-learning-small-sample/glico_model/cifar100.py
code served (permissive licence) · get_code("3fb532813fcf4bb5")
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

np.random.seed(0)

# Create labels for 100 classes with varying numbers of samples per class
# We need at least n_labeled_per_class + n_unlabled_per_class samples per class
labels = []
for class_id in range(100):
    # 50 samples per class to have enough for splitting
    labels.extend([class_id] * 50)

labels = np.array(labels)

n_labeled_per_class = 5
n_unlabled_per_class = 10

ARGS = {
    "labels": labels,
    "n_labeled_per_class": n_labeled_per_class,
    "n_unlabled_per_class": n_unlabled_per_class,
}
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
accuracy Not yet run
unverified (harvested, no recorded run)
IdanAzuri/glico-learning-small-sample/glico_model/evaluation.py
code served (permissive licence) · get_code("66f7263a9f98872e")
How it ran

no invocation recorded.

get_cifar10 Not yet run
unverified (harvested, no recorded run)
IdanAzuri/glico-learning-small-sample/glico_model/cifar10.py
code served (permissive licence) · get_code("492e6a1a39e01db2")
How it ran

no invocation recorded.

get_cifar100 Not yet run
unverified (harvested, no recorded run)
IdanAzuri/glico-learning-small-sample/glico_model/cifar100.py
code served (permissive licence) · get_code("70c82888b308c191")
How it ran

no invocation recorded.

get_code Not yet run
unverified (harvested, no recorded run)
IdanAzuri/glico-learning-small-sample/glico_model/brwose_data_and_FID.py
code served (permissive licence) · get_code("76ecc13d5690af8e")
How it ran

no invocation recorded.

normalise Not yet run
unverified (harvested, no recorded run)
IdanAzuri/glico-learning-small-sample/glico_model/cifar10.py
code served (permissive licence) · get_code("e6db98c4f0c516df")
How it ran

no invocation recorded.

normalise Not yet run
unverified (harvested, no recorded run)
IdanAzuri/glico-learning-small-sample/glico_model/cifar100.py
code served (permissive licence) · get_code("f0a379836cfa494d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 1 the same month, 3 later) and 13 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
is_image_file 11 other papers
4 dated earlier, 1 the same month, 6 later
1812.10366 (2018-12)
A Poisson-Gaussian Denoising Dataset with Real Fluorescence Microscopy Images
1902.02476 (2019-02)
A Simple Baseline for Bayesian Uncertainty in Deep Learning
1904.04717 (2019-04)
Label Propagation for Deep Semi-supervised Learning
and 8 more not named here
normalise 8 other papers
1 dated earlier, 7 later
1905.02249 (2019-05)
MixMatch: A Holistic Approach to Semi-Supervised Learning
2006.12135 (2020-06)
Learning to Generate Noise for Multi-Attack Robustness
2103.14222 (2021-03)
Adversarial Attacks are Reversible with Natural Supervision
and 5 more not named here
has_file_allowed_extension 4 other papers
1 dated earlier, 3 later
1905.11946 (2019-05)
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
2105.04580 (2021-05)
Towards Discovery and Attribution of Open-world GAN Generated Images
2306.02031 (2023-06)
DOS: Diverse Outlier Sampling for Out-of-Distribution Detection
and 1 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

We revisit the long-standing problem of learning from a small sample, to which end we propose a novel method called GLICO (Generative Latent Implicit Conditional Optimization). GLICO learns a mapping from the training examples to a latent space and a generator that generates images from vectors in the latent space. Unlike most recent works, which rely on access to large amounts of unlabeled data, GLICO does not require access to any additional data other than the small set of labeled points. In fact, GLICO learns to synthesize completely new samples for every class using as little as 5 or 10 examples per class, with as few as 10 such classes without imposing any prior. GLICO is then used to augment the small training set while training a classifier on the small sample. To this end, our proposed method samples the learned latent space using spherical interpolation, and generates new examples using the trained generator. Empirical results show that the new sampled set is diverse enough, leading to improvement in image classification in comparison with the state of the art, when trained on small samples obtained from CIFAR-10, CIFAR-100, and CUB-200.

For agents

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

get_harvested_code_for_paper("2003.14297")
get_code_for_paper("2003.14297")
have("2003.14297")

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

curl https://syntology.ai/api/ran/2003.14297.json

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

[![Syntology run record](https://syntology.ai/api/ran/2003.14297.svg)](https://syntology.ai/paper/2003.14297)

Connect an agent — have() is free.