SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1806.10348 · 2018

Learning Visually-Grounded Semantics from Contrastive Adversarial Samples

arXiv · PDF · Open in the Atlas

Code that ran

14 of 15 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 12 have no contract verdict; 2 failures of our instrument; 1 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/1806.10348.json · badge: /api/ran/1806.10348.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
ExplorerFreda/VSE-C canonical (classifier label)
classifier basis: census
14 ran · 1 no recorded run
FunctionStatusWhere it lives
EncoderImage Ran
ran on a synthesized input
ExplorerFreda/VSE-C/VSE_C/model.py
code served (permissive licence) · get_code("16c5b66114f6387f")
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 EncoderImagePrecomp. 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 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
import numpy as np

torch.manual_seed(0)

ARGS = {
    'data_name': 'coco_precomp',
    'img_dim': 4,
    'embed_size': 8,
    'finetune': False,
    'cnn_type': 'vgg19',
    'use_abs': False,
    'no_imgnorm': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EncoderImagePrecomp — a EncoderImagePrecomp output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a EncoderImagePrecomp 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
English Ran
ran on a synthesized input
ExplorerFreda/VSE-C/adversarial_attack/noun.py
code served (permissive licence) · get_code("e143fe1425b0601d")
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

torch.manual_seed(0)

# The function English(word) checks if a word contains only lowercase letters or spaces
# It takes a single parameter 'word' which should be a string

ARGS = {
    'word': 'hello world'
}
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 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
English Ran
ran on a synthesized input
ExplorerFreda/VSE-C/adversarial_attack/relation.py
code served (permissive licence) · get_code("472607bbb393c46f")
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

torch.manual_seed(0)

def English(word):
    for k_ in range(len(word)):
        if word[k_] != ' ' and (not 'a' <= word[k_] <= 'z'):
            return False
    return True

ARGS = {
    'word': 'hello world'
}
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 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
collate_fn Ran
ran on a synthesized input
ExplorerFreda/VSE-C/VSE_C/data.py
code served (permissive licence) · get_code("6961e1859e3c7fe9")
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 list of tuples where each tuple contains:
# - image: torch tensor of shape (3, 256, 256)
# - caption: torch tensor of shape (variable length,)
# - id: some identifier
# - img_id: some image identifier

# Create minimal sample data
image1 = torch.randn(3, 256, 256)
image2 = torch.randn(3, 256, 256)

caption1 = torch.tensor([1, 2, 3, 4, 5])
caption2 = torch.tensor([6, 7, 8])

id1 = 0
id2 = 1

img_id1 = "img_0"
img_id2 = "img_1"

# Build the data list as tuples of (image, caption, id, img_id)
data = [
    (image1, caption1, id1, img_id1),
    (image2, caption2, id2, img_id2),
]

ARGS = {
    "data": data
}
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_fn_train_text Ran
ran on a synthesized input
ExplorerFreda/VSE-C/VSE_C/data.py
code served (permissive licence) · get_code("1a58cc01e62897f2")
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
import torch.utils.data as data

torch.manual_seed(0)

# Create sample data
# Each item in the data list is a tuple of:
# (image, caption, id, img_id, extended_captions)

# Create 2 sample images (3, 256, 256)
image1 = torch.randn(3, 256, 256)
image2 = torch.randn(3, 256, 256)

# Create 2 sample captions (variable length tensors)
caption1 = torch.tensor([1, 2, 3, 4, 5])  # length 5
caption2 = torch.tensor([6, 7, 8])  # length 3

# Create extended captions (list of caption tensors for each sample)
extended_captions1 = [
    torch.tensor([10, 11, 12, 13, 14]),
    torch.tensor([15, 16, 17, 18, 19])
]
extended_captions2 = [
    torch.tensor([20, 21, 22]),
    torch.tensor([23, 24, 25])
]

# Create ids and img_ids
id1 = "sample_id_1"
id2 = "sample_id_2"
img_id1 = 101
img_id2 = 102

# Construct the data list
data_list = [
    (image1, caption1, id1, img_id1, extended_captions1),
    (image2, caption2, id2, img_id2, extended_captions2)
]

ARGS = {
    "data": data_list
}
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
cosine_sim Ran
ran; our contract draft was wrong, not the code
ExplorerFreda/VSE-C/VSE_C/model.py
pointer only (licence: MIT); the stored copy get_code serves, from kadarakos/mulisera, is pointer only (licence: NONE) · get_code("9b0a787b92a87023")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from kadarakos/mulisera, 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.

detect_noun_chunks Ran
ran on a synthesized input
ExplorerFreda/VSE-C/adversarial_attack/relation.py
code served (permissive licence) · get_code("d113e4ae72bec1ef")
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
import numpy as np
from unittest.mock import Mock

# Create a mock spacy Doc-like object with noun_chunks
class MockToken:
    def __init__(self, text, idx):
        self.text = text
        self.idx = idx

class MockSpan:
    def __init__(self, start, end, text):
        self.start = start
        self.end = end
        self.text = text

class MockDoc:
    def __init__(self):
        # Create noun chunks with different spans
        self.noun_chunks = [
            MockSpan(0, 2, "the dog"),
            MockSpan(2, 4, "the cat"),
            MockSpan(5, 7, "the bird"),
        ]
        
        # Create tokens to be indexed
        self.tokens = [
            MockToken("the", 0),
            MockToken("dog", 1),
            MockToken("the", 2),
            MockToken("cat", 3),
            MockToken("with", 4),
            MockToken("the", 5),
            MockToken("bird", 6),
        ]
    
    def __getitem__(self, idx):
        if idx < len(self.tokens):
            return self.tokens[idx]
        # Return a mock token for out of bounds
        return MockToken(",  ", idx)

text_info = MockDoc()

ARGS = {
    "text_info": text_info
}
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 — 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
encode_data Ran
ran on a synthesized input
ExplorerFreda/VSE-C/VSE_C/evaluation.py
code served (permissive licence) · get_code("5eef3e620d10d47b")
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 3, 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 torch.nn as nn
import numpy as np
from collections import OrderedDict

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 8)
    
    def val_start(self):
        self.eval()
    
    def forward_emb(self, images, captions, lengths, volatile=False):
        # Return image embeddings and caption embeddings
        # images shape: (batch_size, 4), captions shape: (batch_size, 4)
        img_emb = self.linear(images).float()  # (batch_size, 8)
        cap_emb = self.linear(captions).float()  # (batch_size, 8)
        return img_emb, cap_emb

class SimpleDataset:
    def __init__(self, num_samples=6):
        self.num_samples = num_samples
    
    def __len__(self):
        return self.num_samples

class SimpleDataLoader:
    def __init__(self, num_batches=3, batch_size=2):
        self.num_batches = num_batches
        self.batch_size = batch_size
        self.dataset = SimpleDataset(num_samples=num_batches * batch_size)
    
    def __iter__(self):
        for i in range(self.num_batches):
            images = torch.randn(self.batch_size, 4)
            captions = torch.randn(self.batch_size, 4)
            lengths = torch.tensor([3, 4] * (self.batch_size // 2 + 1))[:self.batch_size]
            ids = np.array([i * self.batch_size + j for j in range(self.batch_size)])
            yield images, captions, lengths, ids
    
    def __len__(self):
        return self.num_batches

model = SimpleModel()
data_loader = SimpleDataLoader(num_batches=3, batch_size=2)

ARGS = {
    'model': model,
    'data_loader': data_loader,
    'log_step': 2,
    'logging': print,
}
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 3, 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
from_flickr_json Ran
ran on a synthesized input
ExplorerFreda/VSE-C/VSE_C/vocab.py
code served (permissive licence) · get_code("4d60d6b4f58a577f")
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 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 json
import tempfile
import os

# Create a temporary JSON file with Flickr-style data
temp_dir = tempfile.mkdtemp()
flickr_data = {
    "images": [
        {
            "sentences": [
                {"raw": "A dog playing in the park"},
                {"raw": "The dog is brown"}
            ]
        },
        {
            "sentences": [
                {"raw": "A cat sitting on a mat"},
                {"raw": "The cat is orange"}
            ]
        }
    ]
}

temp_path = os.path.join(temp_dir, "flickr_data.json")
with open(temp_path, 'w') as f:
    json.dump(flickr_data, f)

ARGS = {
    "path": temp_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 1, 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
from_txt Ran
ran; our contract draft was wrong, not the code
ExplorerFreda/VSE-C/VSE_C/vocab.py
code served (permissive licence) · get_code("ef05dda21c1cc726")
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 list, 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 tempfile
import os

# Create a temporary text file with some caption data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write(b"caption one\n".decode())
temp_file.write(b"caption two\n".decode())
temp_file.write(b"caption three\n".decode())
temp_file.close()

# Create the ARGS dictionary
ARGS = {
    'txt': temp_file.name
}

# Clean up will happen after function execution
# Note: In production, you'd want to clean up with os.unlink(temp_file.name)
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 list, ndim 1, 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
get_paths Ran
ran on a synthesized input
ExplorerFreda/VSE-C/VSE_C/data.py
code served (permissive licence) · get_code("b61084903f7e3069")
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 os
import numpy as np
import tempfile

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

# Create subdirectories
os.makedirs(os.path.join(tmpdir, 'images', 'train2014'), exist_ok=True)
os.makedirs(os.path.join(tmpdir, 'images', 'val2014'), exist_ok=True)
os.makedirs(os.path.join(tmpdir, 'annotations'), exist_ok=True)

# Create dummy numpy files that the function tries to load
np.save(os.path.join(tmpdir, 'annotations', 'coco_train_ids.npy'), np.array([1, 2, 3, 4, 5]))
np.save(os.path.join(tmpdir, 'annotations', 'coco_dev_ids.npy'), np.array([6, 7, 8, 9, 10]))
np.save(os.path.join(tmpdir, 'annotations', 'coco_test_ids.npy'), np.array([11, 12, 13, 14, 15]))
np.save(os.path.join(tmpdir, 'annotations', 'coco_restval_ids.npy'), np.array([16, 17, 18, 19, 20]))

# Create dummy annotation files
with open(os.path.join(tmpdir, 'annotations', 'captions_train2014.json'), 'w') as f:
    f.write('{}')
with open(os.path.join(tmpdir, 'annotations', 'captions_val2014.json'), 'w') as f:
    f.write('{}')

# Create dummy dataset files for other datasets
with open(os.path.join(tmpdir, 'dataset_flickr8k.json'), 'w') as f:
    f.write('{}')
with open(os.path.join(tmpdir, 'dataset_flickr30k.json'), 'w') as f:
    f.write('{}')

ARGS = {
    'path': tmpdir,
    'name': 'coco',
    'use_restval': False
}
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, 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
isalpha Ran
ran on a synthesized input
ExplorerFreda/VSE-C/evaluation/completion/completion_datamaker.py
code served (permissive licence) · get_code("7a8088198ac7ad87")
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

ARGS = {
    "word": "hello world"
}
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 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
l2norm Ran
ran on a synthesized input
ExplorerFreda/VSE-C/VSE_C/model.py
code served (permissive licence) · get_code("1ff2d83dc35a67ce")
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)

# Construct a small tensor for L2 normalization
# Shape: (2, 4) - batch size 2, feature dimension 4
X = torch.randn(2, 4, dtype=torch.float32)

ARGS = {
    "X": X
}
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
match Ran
ran on a synthesized input
ExplorerFreda/VSE-C/adversarial_attack/relation.py
code served (permissive licence) · get_code("373106f951241575")
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

torch.manual_seed(0)

# The function 'match' compares two sequences element-by-element
# It iterates through indices and checks if elements match
# Both parameters should be sequences (strings, lists, etc.)

orig_text = "hello"
pattern_text = "hello"

ARGS = {
    "orig_text": orig_text,
    "pattern_text": pattern_text
}
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 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
valid Not yet run
unverified (harvested, no recorded run)
ExplorerFreda/VSE-C/adversarial_attack/noun.py
code served (permissive licence) · get_code("4839bd92d6cdf63b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 3 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
cosine_sim 12 other papers
1 dated earlier, 11 later
1707.05612 (2017-07)
VSE++: Improving Visual-Semantic Embeddings with Hard Negatives
1809.06181 (2018-09)
Dual Encoding for Zero-Example Video Retrieval
1909.05506 (2019-09)
CAMP: Cross-Modal Adaptive Message Passing for Text-Image Retrieval
and 9 more not named here
from_txt 3 other papers
1 dated earlier, 2 later
1803.08024 (2018-03)
Stacked Cross Attention for Image-Text Matching
2010.03403 (2020-10)
Universal Weighting Metric Learning for Cross-Modal Matching
2506.23502 (2025-06)
LLM-enhanced Action-aware Multi-modal Prompt Tuning for Image-Text Matching
from_flickr_json 1 other paper
1 later
1909.05506 (2019-09)
CAMP: Cross-Modal Adaptive Message Passing for Text-Image Retrieval

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 study the problem of grounding distributional representations of texts on the visual domain, namely visual-semantic embeddings (VSE for short). Begin with an insightful adversarial attack on VSE embeddings, we show the limitation of current frameworks and image-text datasets (e.g., MS-COCO) both quantitatively and qualitatively. The large gap between the number of possible constitutions of real-world semantics and the size of parallel data, to a large extent, restricts the model to establish the link between textual semantics and visual concepts. We alleviate this problem by augmenting the MS-COCO image captioning datasets with textual contrastive adversarial samples. These samples are synthesized using linguistic rules and the WordNet knowledge base. The construction procedure is both syntax- and semantics-aware. The samples enforce the model to ground learned embeddings to concrete concepts within the image. This simple but powerful technique brings a noticeable improvement over the baselines on a diverse set of downstream tasks, in addition to defending known-type adversarial attacks. We release the codes at https://github.com/ExplorerFreda/VSE-C.

For agents

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

get_harvested_code_for_paper("1806.10348")
get_code_for_paper("1806.10348")
have("1806.10348")

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

curl https://syntology.ai/api/ran/1806.10348.json

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

[![Syntology run record](https://syntology.ai/api/ran/1806.10348.svg)](https://syntology.ai/paper/1806.10348)

Connect an agent — have() is free.