SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.23623 · NeurIPS · 2024

On Learning Multi-Modal Forgery Representation for Diffusion Generated Video Detection

Lei Bai, Xiaoming Liu, Xiaohong Liu, Guangtao Zhai, Xiufeng Song, Xiao Guo, Jiache Zhang, Qirui Li

arXiv · PDF · Open in the Atlas

Code that ran

20 of 24 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; 2 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 17 failures of our instrument; 4 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/2410.23623.json · badge: /api/ran/2410.23623.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
sparklexfantasy/mm-det canonical (classifier label)
classifier basis: census
20 ran · 4 no recorded run
FunctionStatusWhere it lives
collate_video_folder_dataset Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/dataset/vdataset.py
code served (permissive licence) · get_code("a5b506100733c765")
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 torch

torch.manual_seed(0)

# The function expects a batch which is iterable over tuples of (data, label)
# where data and label can be stacked with torch.stack
# We create a list of tuples, each containing two tensors

batch = [
    (torch.randn(4, 3), torch.tensor(0)),
    (torch.randn(4, 3), torch.tensor(1)),
]

ARGS = {
    "batch": batch,
}
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
collate_video_folder_for_recons_dataset_with_fn Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/dataset/vdataset.py
code served (permissive licence) · get_code("96047ea1180d5ab6")
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 torch

torch.manual_seed(0)

# The function expects a batch where each item is a tuple of (fn, data, label)
# fn: a filename string
# data: a tuple of (original_tensor, recons_tensor)
# label: a tensor

batch = [
    ("video_001.mp4", (torch.randn(4, 3), torch.randn(4, 3)), torch.tensor([1.0, 0.0])),
    ("video_002.mp4", (torch.randn(4, 3), torch.randn(4, 3)), torch.tensor([0.0, 1.0])),
]

ARGS = {
    "batch": batch
}
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
create_aa Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/models/vit/resnet.py
code served (permissive licence) · get_code("f4ea0a8791d7d9d4")
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 CustomAALayer. 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
from typing import Type

torch.manual_seed(0)

# Create a custom anti-aliasing layer that accepts channels and stride
class CustomAALayer(nn.Module):
    def __init__(self, channels: int, stride: int):
        super().__init__()
        self.channels = channels
        self.stride = stride
    
    def forward(self, x):
        return x

# Create arguments for create_aa function
aa_layer_avgpool = nn.AvgPool2d
aa_layer_custom = CustomAALayer
channels = 64
stride = 2
enable = True

ARGS = {
    "aa_layer": aa_layer_custom,
    "channels": channels,
    "stride": stride,
    "enable": enable
}
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 CustomAALayer — a CustomAALayer output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a CustomAALayer 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
denormalize_batch_t Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/prepare_reconstructed_dataset.py
code served (permissive licence) · get_code("e12eaf563f19986b")
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 4, dtype family float, shape (2, 3, 4, 5) 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 a batch of normalized images
# Shape: (batch_size, channels, height, width)
batch_size = 2
channels = 3
height = 4
width = 5

img_t = torch.randn(batch_size, channels, height, width)

# Create mean and std for each channel
mean = [0.485, 0.456, 0.406]
std = [0.229, 0.224, 0.225]

ARGS = {
    'img_t': img_t,
    'mean': mean,
    'std': std,
}
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 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 5) 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
downsample_conv Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/models/vit/resnet.py
code served (permissive licence) · get_code("7b8a9c41ab918430")
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)

ARGS = {
    "in_channels": 3,
    "out_channels": 64,
    "kernel_size": 3,
    "stride": 2,
    "dilation": 1,
    "first_dilation": None,
    "norm_layer": nn.BatchNorm2d,
}
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
filter_pretrained_cfg Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/models/vit/_pretrained.py
code served (permissive licence) · get_code("defd8a40081ed476")
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
import torch.nn as nn

torch.manual_seed(0)

# Construct a sample pretrained config dictionary
cfg = {
    'url': 'https://example.com/model.pth',
    'file': 'model.pth',
    'hf_hub_id': 'user/model',
    'hf_hub_filename': 'pytorch_model.bin',
    'source': 'timm',
    'pool_size': (7, 7),
    'first_conv': None,
    'classifier': None,
    'num_classes': 1000,
    'input_size': (3, 224, 224),
    'interpolation': 'bicubic',
    'mean': [0.485, 0.456, 0.406],
    'std': [0.229, 0.224, 0.225],
    'crop_pct': 0.875,
    'test_input_size': (3, 256, 256),
    'test_crop_pct': 1.0,
    'license': 'apache-2.0',
    'description': 'ResNet-50 model',
    'origin': None,
}

ARGS = {
    'cfg': cfg,
    'remove_source': True,
    'remove_null': 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 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
gaussian_blur Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
sparklexfantasy/mm-det/dataset/process.py
code served (permissive licence) · get_code("e2b206e653829624")
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 3, dtype family float, shape (4, 5, 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 numpy as np

# Create a small 3-channel image (height=4, width=5, channels=3)
img = np.random.rand(4, 5, 3).astype(np.float32)

# Define sigma for gaussian blur
sigma = 1.0

ARGS = {
    "img": img,
    "sigma": sigma,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (4, 5, 3) — shape (4, 5, 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
get_arch_name Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/models/vit/_registry.py
code served (permissive licence) · get_code("71420893c20b5733")
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 torch
import torch.nn as nn

def split_model_name_tag(model_name: str, no_tag: str='') -> tuple:
    model_name, *tag_list = model_name.split('.', 1)
    tag = tag_list[0] if tag_list else no_tag
    return (model_name, tag)

def get_arch_name(model_name: str) -> str:
    return split_model_name_tag(model_name)[0]

# Test setup - the function takes a model_name string and extracts the architecture name
# by splitting on '.' and taking the first part

ARGS = {
    'model_name': 'resnet50.v1'
}
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 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
get_dataset_meta Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/prepare_mm_representations.py
code served (permissive licence) · get_code("8c1ea8436a697a27")
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 os
import tempfile
import shutil

# Create a temporary directory with test files
temp_dir = tempfile.mkdtemp()

# Create test files with naming pattern: data_id_suffix
test_files = [
    "sample_001_chunk1",
    "sample_001_chunk2",
    "sample_002_chunk1",
    "sample_003_chunk1",
    "sample_003_chunk2",
    "sample_003_chunk3",
]

for fn in test_files:
    open(os.path.join(temp_dir, fn), 'w').close()

ARGS = {
    "dataset_path": temp_dir
}
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
get_default_transformation_cfg Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/dataset/utils.py
code served (permissive licence) · get_code("495932ca869a515d")
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
import numpy as np

torch.manual_seed(0)

# The function get_default_transformation_cfg takes one parameter: mode
# It accepts 'train' or 'test' as valid values
# The function simply returns a copy of one of two predefined config dicts based on mode

ARGS = {
    'mode': 'train'
}
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 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
get_padding Ran
ran, honoured the contract we drafted
sparklexfantasy/mm-det/models/vit/resnet.py
code served (permissive licence) · get_code("10e6af69cd89654c")
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

torch.manual_seed(0)

ARGS = {
    "kernel_size": 3,
    "stride": 1,
    "dilation": 1,
}
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 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
model_parameters Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/models/vit/_manipulate.py
code served (permissive licence) · get_code("3a0b8256d1b7f59b")
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 generator. 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 torch import nn

torch.manual_seed(0)

# Build a simple model with parameters
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.layer1 = nn.Linear(4, 3)
        self.layer2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = self.layer1(x)
        x = self.layer2(x)
        return x

model = SimpleModel()

ARGS = {
    "model": model,
    "exclude_head": 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 generator — a generator output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a generator 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
named_apply Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/models/vit/_manipulate.py
code served (permissive licence) · get_code("cd025b18122f9a79")
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 SimpleNet. 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 torch import nn

torch.manual_seed(0)

# Create a simple nested module structure
class SimpleNet(nn.Module):
    def __init__(self):
        super().__init__()
        self.layer1 = nn.Linear(4, 3)
        self.layer2 = nn.Sequential(
            nn.Linear(3, 2),
            nn.ReLU()
        )
    
    def forward(self, x):
        return x

module = SimpleNet()

# Create a callback function that the named_apply will call
call_log = []

def callback_fn(module, name):
    """Callback that records which modules were visited"""
    call_log.append((name, type(module).__name__))

# Construct ARGS dict
ARGS = {
    'fn': callback_fn,
    'module': module,
    'name': '',
    'depth_first': True,
    'include_root': 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 SimpleNet — a SimpleNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleNet 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
parse_args Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/prepare_mm_representations.py
code served (permissive licence) · get_code("5756c0240fd19e25")
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 Namespace. 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 argparse

class MockOpt:
    def __init__(self):
        self.parser = argparse.ArgumentParser()

opt = MockOpt()

ARGS = {
    'opt': opt
}
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 Namespace — a Namespace output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Namespace 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
pretrained_cfg_for_features Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/models/vit/_builder.py
code served (permissive licence) · get_code("08cf29ad66c15a74")
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 copy import deepcopy

# Construct a valid pretrained_cfg dict
# The function will deepcopy it and remove certain keys
pretrained_cfg = {
    'num_classes': 1000,
    'classifier': 'linear',
    'global_pool': 'avg',
    'architecture': 'resnet50',
    'input_size': (3, 224, 224),
    'mean': [0.485, 0.456, 0.406],
    'std': [0.229, 0.224, 0.225],
}

ARGS = {
    'pretrained_cfg': pretrained_cfg,
}
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 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
random_split_dataset Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/dataset/vdataset.py
code served (permissive licence) · get_code("4ed1b18f04138d0d")
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. 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 torch.utils.data import Dataset, random_split

torch.manual_seed(0)

class SimpleDataset(Dataset):
    def __init__(self, size=10):
        self.data = torch.randn(size, 4)
        self.labels = torch.randint(0, 2, (size,))
    
    def __len__(self):
        return len(self.data)
    
    def __getitem__(self, idx):
        return self.data[idx], self.labels[idx]

dataset = SimpleDataset(size=10)
split_ratio = [6, 3, 1]
seed = 100

ARGS = {
    "dataset": dataset,
    "split_ratio": split_ratio,
    "seed": seed
}
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 — a list 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
sample_by_interval Ran
ran, honoured the contract we drafted
sparklexfantasy/mm-det/prepare_mm_representations.py
code served (permissive licence) · get_code("03e52c96c55bb4b1")
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 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

torch.manual_seed(0)

# Analyzing the function:
# - frame_count: an integer representing total number of frames
# - interval: an integer representing the sampling interval (default 200)
# The function iterates through frame indices starting at 1, incrementing by interval each time,
# until reaching frame_count, and returns a list of sampled indices.

ARGS = {
    "frame_count": 1000,
    "interval": 200
}
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 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
sample_continuous Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
sparklexfantasy/mm-det/dataset/process.py
code served (permissive licence) · get_code("2d279d661251af46")
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 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 random

def sample_continuous(s):
    if len(s) == 1:
        return s[0]
    if len(s) == 2:
        rg = s[1] - s[0]
        return random.random() * rg + s[0]
    raise ValueError('Length of iterable s should be 1 or 2.')

# Test case 1: single element
ARGS = {
    's': [5.0]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 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
sample_discrete Ran
ran, violated the contract we drafted
sparklexfantasy/mm-det/dataset/process.py
code served (permissive licence) · get_code("d9f67916280ae1e3")
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 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 random

def sample_discrete(s):
    if len(s) == 1:
        return s[0]
    return random.choice(s)

# Test with a list of discrete values
ARGS = {
    "s": [1, 2, 3, 4, 5]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 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
split_model_name_tag Ran
ran; our contract draft was wrong, not the code
sparklexfantasy/mm-det/models/vit/_registry.py
code served (permissive licence) · get_code("de9c9579e64472b9")
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

def split_model_name_tag(model_name: str, no_tag: str='') -> tuple:
    model_name, *tag_list = model_name.split('.', 1)
    tag = tag_list[0] if tag_list else no_tag
    return (model_name, tag)

# Test case 1: model_name with a tag
model_name_1 = "resnet50.v1"
no_tag_1 = ""

# Test case 2: model_name without a tag
model_name_2 = "bert"
no_tag_2 = "default"

# Test case 3: model_name with multiple dots (only first dot should be used as separator)
model_name_3 = "gpt2.large.fp16"
no_tag_3 = ""

ARGS = {
    "model_name": "resnet50.v2",
    "no_tag": "default"
}
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, 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
build_model_with_cfg Not yet run
unverified (harvested, no recorded run)
sparklexfantasy/mm-det/models/vit/_builder.py
code served (permissive licence) · get_code("542765e1fd8ceb42")
How it ran

no invocation recorded.

get_default_transformation Not yet run
unverified (harvested, no recorded run)
sparklexfantasy/mm-det/dataset/utils.py
code served (permissive licence) · get_code("295f60cdd29ae30f")
How it ran

no invocation recorded.

group_with_matcher Not yet run
unverified (harvested, no recorded run)
sparklexfantasy/mm-det/models/vit/_manipulate.py
code served (permissive licence) · get_code("787ccd49175173ad")
How it ran

no invocation recorded.

resolve_pretrained_cfg Not yet run
unverified (harvested, no recorded run)
sparklexfantasy/mm-det/models/vit/_builder.py
code served (permissive licence) · get_code("29a3e687db3fce97")
How it ran

no invocation recorded.

Also present in other papers' code

Of 24 samples lifted from this paper's linked repositories, 24 carry a function identity and 0 do not. Of the 24 distinct functions with an identity, 12 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 10 later) and 12 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
named_apply 3 other papers
2 dated earlier, 1 later
2402.12138 (2024-02)
Perceiving Longer Sequences With Bi-Directional Cross-Attention Transformers
2407.19985 (2024-07)
Mixture of Nested Experts: Adaptive Processing of Visual Tokens
2412.10569 (2024-12)
Learning to Merge Tokens via Decoupled Embedding for Efficient Vision Transformers
sample_continuous 2 other papers
2 later
2503.18286 (2025-03)
CO-SPY: Combining Semantic and Pixel Features to Detect Synthetic Images by AI
2503.20188 (2025-03)
Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face Detector
sample_discrete 2 other papers
2 later
2503.18286 (2025-03)
CO-SPY: Combining Semantic and Pixel Features to Detect Synthetic Images by AI
2503.20188 (2025-03)
Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face Detector
filter_pretrained_cfg 1 other paper
1 later
2412.10569 (2024-12)
Learning to Merge Tokens via Decoupled Embedding for Efficient Vision Transformers
gaussian_blur 1 other paper
1 later
2503.20188 (2025-03)
Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face Detector
get_arch_name 1 other paper
1 later
2412.10569 (2024-12)
Learning to Merge Tokens via Decoupled Embedding for Efficient Vision Transformers
get_default_transformation 1 other paper
1 later
2503.20188 (2025-03)
Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face Detector
group_with_matcher 1 other paper
1 later
2412.10569 (2024-12)
Learning to Merge Tokens via Decoupled Embedding for Efficient Vision Transformers
model_parameters 1 other paper
1 later
2412.10569 (2024-12)
Learning to Merge Tokens via Decoupled Embedding for Efficient Vision Transformers
pretrained_cfg_for_features 1 other paper
1 dated earlier
2406.00830 (2024-06)
Collaborative Novel Object Discovery and Box-Guided Cross-Modal Alignment for Open-Vocabulary 3D Object Detection
resolve_pretrained_cfg 1 other paper
1 dated earlier
2406.00830 (2024-06)
Collaborative Novel Object Discovery and Box-Guided Cross-Modal Alignment for Open-Vocabulary 3D Object Detection
split_model_name_tag 1 other paper
1 later
2412.10569 (2024-12)
Learning to Merge Tokens via Decoupled Embedding for Efficient Vision Transformers

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

Large numbers of synthesized videos from diffusion models pose threats to information security and authenticity, leading to an increasing demand for generated content detection. However, existing video-level detection algorithms primarily focus on detecting facial forgeries and often fail to identify diffusion-generated content with a diverse range of semantics. To advance the field of video forensics, we propose an innovative algorithm named Multi-Modal Detection(MM-Det) for detecting diffusion-generated videos. MM-Det utilizes the profound perceptual and comprehensive abilities of Large Multi-modal Models (LMMs) by generating a Multi-Modal Forgery Representation (MMFR) from LMM's multi-modal space, enhancing its ability to detect unseen forgery content. Besides, MM-Det leverages an In-and-Across Frame Attention (IAFA) mechanism for feature augmentation in the spatio-temporal domain. A dynamic fusion strategy helps refine forgery representations for the fusion. Moreover, we construct a comprehensive diffusion video dataset, called Diffusion Video Forensics (DVF), across a wide range of forgery videos. MM-Det achieves state-of-the-art performance in DVF, demonstrating the effectiveness of our algorithm. Both source code and DVF are available at link.

For agents

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

get_harvested_code_for_paper("2410.23623")
get_code_for_paper("2410.23623")
have("2410.23623")

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

curl https://syntology.ai/api/ran/2410.23623.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.23623.svg)](https://syntology.ai/paper/2410.23623)

Connect an agent — have() is free.