SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1807.09937 · 2018

HiDDeN: Hiding Data With Deep Networks

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/1807.09937.json · badge: /api/ran/1807.09937.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
ando-khachatryan/HiDDeN community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
dungpham98/Hidden_Blocking_Artifact_Reduction community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
FunctionStatusWhere it lives
blocking_value Ran
ran on a synthesized input
dungpham98/Hidden_Blocking_Artifact_Reduction/blocking_calculation.py
code served (permissive licence) · get_code("60fd8f53409daa97")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function expects:
# - encoded_imgs: a list of tensors, where each element is indexed [i][idx][0]
#   and can be converted to numpy via .cpu().detach().numpy()
#   Each resulting numpy array should be shape (block_size, block_size)
# - batch: an integer representing batch size
# - block_size: an integer (32 in the example)
# - block_number: an integer used for indexing

block_size = 32
batch = 2
block_number = 3

# Create encoded_imgs as a list of tensors
# Each tensor should support indexing [idx][0] and have .cpu().detach().numpy()
# Shape: (batch, 1, block_size, block_size) to support [idx][0] indexing
# We need at least block_number + 1 elements in the list for the loops to work
encoded_imgs = [
    torch.randn(batch, 1, block_size, block_size) for _ in range(block_number + 2)
]

ARGS = {
    "encoded_imgs": encoded_imgs,
    "batch": batch,
    "block_size": block_size,
    "block_number": block_number,
}
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 float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_random_rectangle_inside Ran
ran on a synthesized input
ando-khachatryan/HiDDeN/noise_layers/crop.py
code served (permissive licence) · get_code("6dd98b0c20170513")
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 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

torch.manual_seed(0)
np.random.seed(0)

# Create a small image tensor with shape (batch, channels, height, width)
# Using distinct sizes: batch=2, channels=3, height=32, width=48
image = torch.randn(2, 3, 32, 48)

# Define ratio ranges for height and width
height_ratio_range = (0.5, 0.9)
width_ratio_range = (0.6, 0.95)

ARGS = {
    "image": image,
    "height_ratio_range": height_ratio_range,
    "width_ratio_range": width_ratio_range,
}
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 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
parse_pair Ran
ran on a synthesized input
ando-khachatryan/HiDDeN/noise_argparser.py
code served (permissive licence) · get_code("da226f72e51ba9e6")
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 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 torch
import numpy as np

def parse_pair(match_groups):
    heights = match_groups[0].split(',')
    hmin = float(heights[0])
    hmax = float(heights[1])
    widths = match_groups[1].split(',')
    wmin = float(widths[0])
    wmax = float(widths[1])
    return ((hmin, hmax), (wmin, wmax))

# match_groups is expected to be a sequence of strings
# Each string contains comma-separated values that can be converted to floats
match_groups = [
    "10.5,20.3",  # heights as a comma-separated string
    "5.1,15.9"    # widths as a comma-separated string
]

ARGS = {
    "match_groups": match_groups
}
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 2, 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
random_float Ran
ran, honoured the contract we drafted
ando-khachatryan/HiDDeN/noise_layers/crop.py
code served (permissive licence) · get_code("d87db75c2765498f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

ARGS = {
    "min": 0.0,
    "max": 10.0
}
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 float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
weighted_mse_loss Ran
ran on a synthesized input
dungpham98/Hidden_Blocking_Artifact_Reduction/model/hidden.py
code served (permissive licence) · get_code("4cb7480b48bada32")
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 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

torch.manual_seed(0)

# Create small tensors for input and target
# The function expects (batch, channels, height, width) based on the unsqueeze operations
batch_size = 2
channels = 4
height = 3
width = 3

input_tensor = torch.randn(batch_size, channels, height, width)
target_tensor = torch.randn(batch_size, channels, height, width)

# Create weights tensor - should be (batch_size,) since it's unsqueezed 3 times
# to match the 4D shape of input and target
weights_tensor = torch.ones(batch_size)

ARGS = {
    "input": input_tensor,
    "target": target_tensor,
    "weights": weights_tensor,
}
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 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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 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
random_float 2 other papers
2 later
Kim_WOUAF_Weight_Modulation_for_User_Attribution_and_Fingerprinting_in_Text-to-Image_CVPR_2024_paper (2024 (year only))
WOUAF: Weight Modulation for User Attribution and Fingerprinting in Text-to-Image Diffusion Models
Wang_SleeperMark_Towards_Robust_Watermark_against_Fine-Tuning_Text-to-image_Diffusion_Models_CVPR_2025_paper (2025 (year only))
SleeperMark: Towards Robust Watermark against Fine-Tuning Text-to-image Diffusion Models

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

Recent work has shown that deep neural networks are highly sensitive to tiny perturbations of input images, giving rise to adversarial examples. Though this property is usually considered a weakness of learned models, we explore whether it can be beneficial. We find that neural networks can learn to use invisible perturbations to encode a rich amount of useful information. In fact, one can exploit this capability for the task of data hiding. We jointly train encoder and decoder networks, where given an input message and cover image, the encoder produces a visually indistinguishable encoded image, from which the decoder can recover the original message. We show that these encodings are competitive with existing data hiding algorithms, and further that they can be made robust to noise: our models learn to reconstruct hidden information in an encoded image despite the presence of Gaussian blurring, pixel-wise dropout, cropping, and JPEG compression. Even though JPEG is non-differentiable, we show that a robust model can be trained using differentiable approximations. Finally, we demonstrate that adversarial training improves the visual quality of encoded images.

For agents

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

get_harvested_code_for_paper("1807.09937")
get_code_for_paper("1807.09937")
have("1807.09937")

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

curl https://syntology.ai/api/ran/1807.09937.json

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

[![Syntology run record](https://syntology.ai/api/ran/1807.09937.svg)](https://syntology.ai/paper/1807.09937)

Connect an agent — have() is free.