SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · aaai_25362 · AAAI · 2023

Global-Local Characteristic Excited Cross-Modal Attacks from Images to Videos

Yuanfang Guo, Yunhong Wang, Ruikui Wang

Open in the Atlas

Code that ran

5 of 6 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; 1 no recorded run), measured 2026-11-22; 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/aaai_25362.json · badge: /api/ran/aaai_25362.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
lwmming/Cross-Modal-Attack canonical (classifier label)
classifier basis: census
5 ran · 1 no recorded run
FunctionStatusWhere it lives
change_cfg Ran
ran on a synthesized input
lwmming/Cross-Modal-Attack/main_k400.py
code served (permissive licence) · get_code("1fa72ed0ce38f36b")
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 MockConfig. 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

class MockConfig:
    def __init__(self):
        self.CONFIG = MockConfigInner()

class MockConfigInner:
    def __init__(self):
        self.DATA = MockData()
        self.MODEL = MockModel()
        self.VAL = MockVal()

class MockData:
    def __init__(self):
        self.VAL_DATA_PATH = None
        self.VAL_ANNO_PATH = None

class MockModel:
    def __init__(self):
        self.PRETRAINED = False

class MockVal:
    def __init__(self):
        self.BATCH_SIZE = None

torch.manual_seed(0)

cfg = MockConfig()
batch_size = 32
random = True

ARGS = {
    "cfg": cfg,
    "batch_size": batch_size,
    "random": random
}
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 MockConfig — a MockConfig output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MockConfig 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
change_cfg Ran
ran on a synthesized input
lwmming/Cross-Modal-Attack/utils.py
code served (permissive licence) · get_code("ba2aa6cefe15ae47")
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 StandInConfig. 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)

class StandInConfig:
    def __init__(self):
        self.CONFIG = StandInConfigNested()

class StandInConfigNested:
    def __init__(self):
        self.DATA = StandInData()
        self.MODEL = StandInModel()
        self.VAL = StandInVal()

class StandInData:
    def __init__(self):
        self.VAL_DATA_PATH = ""
        self.VAL_ANNO_PATH = ""

class StandInModel:
    def __init__(self):
        self.PRETRAINED = False

class StandInVal:
    def __init__(self):
        self.BATCH_SIZE = 1

cfg = StandInConfig()
batch_size = 8

ARGS = {
    "cfg": cfg,
    "batch_size": batch_size
}
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 StandInConfig — a StandInConfig output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a StandInConfig 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
pil_loader Ran
ran, honoured the contract we drafted
lwmming/Cross-Modal-Attack/transforms_ucf101.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from YuemingJin/MTRCNet-CL, is pointer only (licence: NONE) · get_code("f13b031e63ffbc06")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from YuemingJin/MTRCNet-CL, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

transform_video Ran
ran on a synthesized input
lwmming/Cross-Modal-Attack/main_k400.py
code served (permissive licence) · get_code("c64f0d04d77bfd4c")
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 3, dtype family float, shape (3, 4, 5) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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 small video tensor with shape [C, H, W] where C=3 (RGB channels)
# Using shape [3, 4, 5] for distinct dimensions
video = torch.randn(3, 4, 5, dtype=torch.float32)

ARGS = {
    'video': video,
    'mode': 'forward'
}
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 3, dtype family float, shape (3, 4, 5) — shape (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
video_loader Ran
ran on a synthesized input
lwmming/Cross-Modal-Attack/transforms_ucf101.py
code served (permissive licence) · get_code("969914385b6159ef")
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 4, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile
from PIL import Image
import numpy as np

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

# Create a few test images
for i in range(0, 3):
    img_array = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)
    img = Image.fromarray(img_array.transpose(1, 2, 0), 'RGB')
    img_path = os.path.join(temp_dir, f'image_{i:05d}.jpg')
    img.save(img_path)

# Define a simple image loader function
def simple_image_loader(image_path):
    img = Image.open(image_path)
    return np.array(img)

# Define arguments
ARGS = {
    'video_dir_path': temp_dir,
    'frame_indices': [0, 1, 2],
    'image_loader': simple_image_loader
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 4, 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
accimage_loader Not yet run
unverified (harvested, no recorded run)
lwmming/Cross-Modal-Attack/transforms_ucf101.py
code served (permissive licence) · get_code("53558f474d00dc28")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 1 later) and 3 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
pil_loader 28 other papers
19 dated earlier, 5 later, 4 same year, order not determinable
1702.00758 (2017-02)
HashNet: Deep Learning to Hash by Continuation
1805.07932 (2018-05)
Bilinear Attention Networks
1807.06587 (2018-07)
Deep Exemplar-based Colorization
and 25 more not named here
accimage_loader 3 other papers
3 dated earlier
1912.05534 (2019-12)
Why Can't I Dance in the Mall? Learning to Mitigate Scene Bias in Action Recognition
2009.09805 (2020-09)
Active Contrastive Learning of Audio-Visual Video Representations
2111.13675 (2021-11)
Weakly-guided Self-supervised Pretraining for Temporal Activity Detection
video_loader 2 other papers
2 dated earlier
1912.05534 (2019-12)
Why Can't I Dance in the Mall? Learning to Mitigate Scene Bias in Action Recognition
2009.09805 (2020-09)
Active Contrastive Learning of Audio-Visual Video Representations

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 transferability of adversarial examples is the key property in practical black-box scenarios. Currently, numerous methods improve the transferability across different models trained on the same modality of data. The investigation of generating video adversarial examples with imagebased substitute models to attack the target video models, i.e., cross-modal transferability of adversarial examples, is rarely explored. A few works on cross-modal transferability directly apply image attack methods for each frame and no factors especial for video data are considered, which limits the cross-modal transferability of adversarial examples. In this paper, we propose an effective cross-modal attack method which considers both the global and local characteristics of video data. Firstly, from the global perspective, we introduce inter-frame interaction into attack process to induce more diverse and stronger gradients rather than perturb each frame separately. Secondly, from the local perspective, we disrupt the inherently local correlation of frames within a video, which prevents black-box video model from capturing valuable temporal clues. Extensive experiments on the UCF-101 and Kinetics-400 validate the proposed method significantly improves cross-modal transferability and even surpasses stronger baseline using video models as substitute model. Our source codes are available at https://github.com/lwmming/Cross-Modal-Attack.

For agents

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

get_harvested_code_for_paper("aaai_25362")
get_code_for_paper("aaai_25362")
have("aaai_25362")

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

curl https://syntology.ai/api/ran/aaai_25362.json

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

[![Syntology run record](https://syntology.ai/api/ran/aaai_25362.svg)](https://syntology.ai/paper/aaai_25362)

Connect an agent — have() is free.