SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.02335 · 2024

Linguistic Profiling of Deepfakes: An Open Database for Next-Generation Deepfake Detection

arXiv · PDF · Open in the Atlas

Code that ran

5 of 8 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 0 have no contract verdict; 5 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/2401.02335.json · badge: /api/ran/2401.02335.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
dflip3k/dflip-3k canonical (classifier label)
classifier basis: census
5 ran · 3 no recorded run
FunctionStatusWhere it lives
create_dataloader Ran
ran; our contract draft was wrong, not the code
dflip3k/dflip-3k/src/safety_detector/safety_checker.py
code served (permissive licence) · get_code("fcf321aa47474c75")
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 DataLoader. 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 os
import json
import tempfile
from PIL import Image
import torch
import numpy as np
from torchvision import transforms
from torch.utils.data import Dataset, DataLoader

torch.manual_seed(0)

# Create a temporary directory for test images
temp_dir = tempfile.mkdtemp()

# Create a few dummy images
image_array_1 = np.random.randint(0, 256, (64, 64, 3), dtype=np.uint8)
image_1 = Image.fromarray(image_array_1)
image_1.save(os.path.join(temp_dir, "image1.jpg"))

image_array_2 = np.random.randint(0, 256, (64, 64, 3), dtype=np.uint8)
image_2 = Image.fromarray(image_array_2)
image_2.save(os.path.join(temp_dir, "image2.png"))

image_array_3 = np.random.randint(0, 256, (64, 64, 3), dtype=np.uint8)
image_3 = Image.fromarray(image_array_3)
image_3.save(os.path.join(temp_dir, "image3.jpeg"))

# Create a JSON file with metadata
json_file = os.path.join(temp_dir, "metadata.json")
metadata = {
    "image1.jpg": {"Prompt": "a cat"},
    "image2.png": {"Prompt": "a dog"},
    "image3.jpeg": {"Prompt": ""}
}
with open(json_file, 'w') as f:
    json.dump(metadata, f)

ARGS = {
    "dirpath": temp_dir,
    "jsonfile": json_file,
    "output_size": (260, 260)
}
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 DataLoader — a DataLoader output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataLoader 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
load_model Ran
ran; our contract draft was wrong, not the code
dflip3k/dflip-3k/utils/ASimilarityCalculatior.py
code served (permissive licence) · get_code("df55583b9c1822dd")
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 torch
from pathlib import Path
from tempfile import TemporaryDirectory
import os

torch.manual_seed(0)

# Create a temporary directory and a fake model file
temp_dir = TemporaryDirectory()
temp_path = Path(temp_dir.name)

# Create a .safetensors file
safetensors_path = temp_path / "model.safetensors"
# Create a minimal valid safetensors file by using torch tensors
from safetensors.torch import save_file
test_tensors = {
    "layer1.weight": torch.randn(4, 3),
    "layer1.bias": torch.randn(4),
}
save_file(test_tensors, str(safetensors_path))

# Also create a .pt checkpoint file for alternative testing
pt_path = temp_path / "model.pt"
state_dict = {
    "layer1.weight": torch.randn(4, 3),
    "layer1.bias": torch.randn(4),
}
torch.save({"state_dict": state_dict}, str(pt_path))

# Use the safetensors file as the default test path
ARGS = {
    "path": safetensors_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
safetensors, torch — 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
model_hash Ran
ran; our contract draft was wrong, not the code
dflip3k/dflip-3k/utils/ASimilarityCalculatior.py
code served (permissive licence) · get_code("4b4d79cac96cbf27")
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 str. 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 tempfile
import os
import hashlib

# Create a temporary file with known content
temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.bin')
temp_filename = temp_file.name

# Write enough data so that the function can seek to position 1048576 and read 65536 bytes
# We need at least 1048576 + 65536 bytes
required_size = 1048576 + 65536
data = b'x' * required_size
temp_file.write(data)
temp_file.close()

ARGS = {
    'filename': temp_filename
}

# Clean up will happen after function execution
# The temporary file will be deleted by the OS eventually, or we can add cleanup code
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
preprocess Ran
ran; our contract draft was wrong, not the code
dflip3k/dflip-3k/read_doc.py
code served (permissive licence) · get_code("a7dc320a88be2f18")
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 str. 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 re

def preprocess(text):
    text = text.replace('\n', ' ')
    text = re.sub('\\s+', ' ', text)
    return text

ARGS = {
    'text': "Hello\nworld  \n  this   is\ta test"
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
read_json_file Ran
ran; our contract draft was wrong, not the code
dflip3k/dflip-3k/utils/download_models.py
code served (permissive licence) · get_code("00b94adf3a8669df")
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 json
import tempfile
import os

# Create a temporary JSON file with valid data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_data.json")

test_data = {
    "name": "test",
    "values": [1, 2, 3, 4],
    "nested": {
        "key": "value"
    }
}

with open(temp_file, 'w') as f:
    json.dump(test_data, f)

ARGS = {
    "file_path": 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
cal_cross_attn Not yet run
unverified (harvested, no recorded run)
dflip3k/dflip-3k/utils/ASimilarityCalculatior.py
code served (permissive licence) · get_code("b079cc15fdea6e31")
How it ran

no invocation recorded.

get_aesthetic_model Not yet run
unverified (harvested, no recorded run)
dflip3k/dflip-3k/utils/evaluate.py
code served (permissive licence) · get_code("87a7a854ddf1745e")
How it ran

no invocation recorded.

get_download_urls Not yet run
unverified (harvested, no recorded run)
dflip3k/dflip-3k/utils/download_models.py
code served (permissive licence) · get_code("01fd34228f279be3")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 8 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.

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 emergence of text-to-image generative models has revolutionized the field of deepfakes, enabling the creation of realistic and convincing visual content directly from textual descriptions. However, this advancement presents considerably greater challenges in detecting the authenticity of such content. Existing deepfake detection datasets and methods often fall short in effectively capturing the extensive range of emerging deepfakes and offering satisfactory explanatory information for detection. To address the significant issue, this paper introduces a deepfake database (DFLIP-3K) for the development of convincing and explainable deepfake detection. It encompasses about 300K diverse deepfake samples from approximately 3K generative models, which boasts the largest number of deepfake models in the literature. Moreover, it collects around 190K linguistic footprints of these deepfakes. The two distinguished features enable DFLIP-3K to develop a benchmark that promotes progress in linguistic profiling of deepfakes, which includes three sub-tasks namely deepfake detection, model identification, and prompt prediction. The deepfake model and prompt are two essential components of each deepfake, and thus dissecting them linguistically allows for an invaluable exploration of trustworthy and interpretable evidence in deepfake detection, which we believe is the key for the next-generation deepfake detection. Furthermore, DFLIP-3K is envisioned as an open database that fosters transparency and encourages collaborative efforts to further enhance its growth. Our extensive experiments on the developed benchmark verify that our DFLIP-3K database is capable of serving as a standardized resource for evaluating and comparing linguistic-based deepfake detection, identification, and prompt prediction techniques.

For agents

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

get_harvested_code_for_paper("2401.02335")
get_code_for_paper("2401.02335")
have("2401.02335")

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

curl https://syntology.ai/api/ran/2401.02335.json

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

[![Syntology run record](https://syntology.ai/api/ran/2401.02335.svg)](https://syntology.ai/paper/2401.02335)

Connect an agent — have() is free.