SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.04135 · 2022

VoLTA: Vision-Language Transformer with Weakly-Supervised Local-Feature Alignment

arXiv · PDF · Open in the Atlas

Code that ran

12 of 14 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 6 have no contract verdict; 5 failures of our instrument; 2 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/2210.04135.json · badge: /api/ran/2210.04135.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
ShramanPramanick/VoLTA canonical (classifier label)
classifier basis not recorded on the link
10 ran · 2 no recorded run
copy not recorded community
not in the paper's declared repositories
2 ran · 0 no recorded run
FunctionStatusWhere it lives
IPOT_distance_torch Ran
ran on a synthesized input
ShramanPramanick/VoLTA/Pre-training/OT_torch_.py
code served (permissive licence) · get_code("b1e24c9598daad69")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. 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)

# The function IPOT_distance_torch expects:
# C: a cost matrix (2D tensor)
# n: number of rows in C
# m: number of columns in C
# miu: a vector of shape (n,) representing row marginals
# nu: a vector of shape (m,) representing column marginals

# Create a small cost matrix C with shape (4, 3)
C = torch.randn(4, 3)

# n is the number of rows
n = 4

# m is the number of columns
m = 3

# miu: row marginals, shape (n,)
miu = torch.ones(n) / n

# nu: column marginals, shape (m,)
nu = torch.ones(m) / m

ARGS = {
    'C': C,
    'n': n,
    'm': m,
    'miu': miu,
    'nu': nu
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
IPOT_torch Ran
ran on a synthesized input
ShramanPramanick/VoLTA/Pre-training/OT_torch_.py
code served (permissive licence) · get_code("674f9140f14e54a6")
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 (3, 4) 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
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# - C: cost matrix, used with torch.exp and matrix operations, shape (n, m)
# - n: number of rows, int
# - m: number of columns, int
# - miu: vector of size n, used in division
# - nu: vector of size m, used in operations
# - beta: scalar float, default 0.5

# Create small tensors with distinct dimensions
n = 3
m = 4

# C: cost matrix of shape (n, m)
C = torch.randn(n, m).float()

# miu: vector of size n
miu = torch.ones(n).float()

# nu: vector of size m
nu = torch.ones(m).float()

# beta: scalar
beta = 0.5

ARGS = {
    "C": C,
    "n": n,
    "m": m,
    "miu": miu,
    "nu": nu,
    "beta": beta
}
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 Tensor, ndim 2, dtype family float, shape (3, 4) — shape (3, 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
cost_matrix_torch Ran
ran; our contract draft was wrong, not the code
ShramanPramanick/VoLTA/Pre-training/OT_torch_.py
code served (permissive licence) · get_code("d79fd2ac61421292")
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, 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 torch

torch.manual_seed(0)

# Create two tensors with distinct dimensions
# x: shape (4, 3) - 4 features, 3 samples
# y: shape (4, 2) - 4 features, 2 samples
x = torch.randn(4, 3)
y = torch.randn(4, 2)

ARGS = {
    "x": x,
    "y": y,
}
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, 3) — shape (2, 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
create_position_ids_from_input_ids Ran
ran; our contract draft was wrong, not the code
ShramanPramanick/VoLTA/Pre-training/roberta.py
code served (permissive licence) · get_code("da1c54dffed5d60c")
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 Tensor, ndim 2, dtype family int, 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)

# Create input_ids tensor with shape (batch_size=2, seq_length=4)
# Values should include some padding tokens and some non-padding tokens
input_ids = torch.tensor([
    [0, 1, 2, 0],      # batch 1: has padding_idx (0) at positions 0 and 3
    [1, 2, 0, 1]       # batch 2: has padding_idx (0) at position 2
])

# padding_idx is the token ID that represents padding
padding_idx = 0

# past_key_values_length represents how many tokens were processed before
past_key_values_length = 0

ARGS = {
    "input_ids": input_ids,
    "padding_idx": padding_idx,
    "past_key_values_length": past_key_values_length
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 int, 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
create_positive_map Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from microsoft/glip
pointer only · get_code("525b77c721203779")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

create_positive_map_label_to_token_from_positive_map Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from microsoft/glip
pointer only · get_code("ea2c188d68558611")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

get_id_list_coco2014 Ran
ran on a synthesized input
ShramanPramanick/VoLTA/Pre-training/get_id_list.py
code served (permissive licence) · get_code("c98d2b87e8adc438")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import json
import tempfile
import os
from collections import defaultdict

# Create a temporary directory structure for COCO2014
temp_dir = tempfile.mkdtemp()
root_dir = temp_dir + '/'

# Create train2014 and val2014 directories
os.makedirs(f'{root_dir}train2014', exist_ok=True)
os.makedirs(f'{root_dir}val2014', exist_ok=True)

# Create dummy image files
open(f'{root_dir}train2014/COCO_train2014_000000000001.jpg', 'w').close()
open(f'{root_dir}train2014/COCO_train2014_000000000002.jpg', 'w').close()
open(f'{root_dir}val2014/COCO_val2014_000000000003.jpg', 'w').close()
open(f'{root_dir}val2014/COCO_val2014_000000000004.jpg', 'w').close()

# Create dataset_coco.json with proper structure
dataset_coco = {
    'images': [
        {
            'filename': 'COCO_train2014_000000000001.jpg',
            'split': 'train',
            'sentences': [
                {'raw': 'A dog running in the park'},
                {'raw': 'The dog is playing'},
            ]
        },
        {
            'filename': 'COCO_train2014_000000000002.jpg',
            'split': 'train',
            'sentences': [
                {'raw': 'A cat on a mat'},
            ]
        },
        {
            'filename': 'COCO_val2014_000000000003.jpg',
            'split': 'restval',
            'sentences': [
                {'raw': 'A bird flying high'},
                {'raw': 'The bird soars'},
            ]
        },
        {
            'filename': 'COCO_val2014_000000000004.jpg',
            'split': 'val',
            'sentences': [
                {'raw': 'A flower blooming'},
            ]
        },
    ]
}

with open(f'{root_dir}dataset_coco.json', 'w') as fp:
    json.dump(dataset_coco, fp)

ARGS = {
    'root_dir': root_dir,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_id_list_separate Ran
ran on a synthesized input
ShramanPramanick/VoLTA/Pre-training/get_id_list.py
code served (permissive licence) · get_code("33e998e445a0eee0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tempfile
import os

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

# Create a subdirectory structure
subdir = os.path.join(temp_dir, 'images')
os.makedirs(subdir, exist_ok=True)

# Create test image and text files
test_images = []
test_texts = []

for i in range(2):
    # Create .jpg file (just a dummy file, not actual image data)
    jpg_path = os.path.join(subdir, f'image_{i}.jpg')
    with open(jpg_path, 'w') as f:
        f.write('dummy image data')
    test_images.append(jpg_path)
    
    # Create corresponding .txt file with caption
    txt_path = os.path.join(subdir, f'image_{i}.txt')
    caption = f'This is a test caption number {i}'
    with open(txt_path, 'w') as f:
        f.write(caption)
    test_texts.append(caption)

ARGS = {
    'root_dir': 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 tuple, ndim 2, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_state_dict Ran
ran on a synthesized input
ShramanPramanick/VoLTA/Pre-training/swin_helpers.py
code served (permissive licence) · get_code("bc3a13e8155e8140")
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 OrderedDict. 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 os
import tempfile
from collections import OrderedDict

torch.manual_seed(0)

# Create a temporary checkpoint file with a valid state dict
temp_dir = tempfile.mkdtemp()
checkpoint_path = os.path.join(temp_dir, "checkpoint.pt")

# Create a sample state dict
state_dict = OrderedDict([
    ("layer1.weight", torch.randn(4, 3)),
    ("layer1.bias", torch.randn(4)),
    ("module.layer2.weight", torch.randn(5, 4)),
])

# Create checkpoint dict with both regular and EMA state dicts
checkpoint = {
    "state_dict": state_dict,
    "state_dict_ema": OrderedDict([
        ("layer1.weight", torch.randn(4, 3)),
        ("layer1.bias", torch.randn(4)),
        ("layer2.weight", torch.randn(5, 4)),
    ]),
    "epoch": 10,
}

# Save checkpoint
torch.save(checkpoint, checkpoint_path)

ARGS = {
    "checkpoint_path": checkpoint_path,
    "use_ema": 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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
swin_adapt_position_encoding Ran
ran on a synthesized input
ShramanPramanick/VoLTA/Pre-training/swin_helpers.py
code served (permissive licence) · get_code("3c0bf9aa4752cff5")
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)

# The function expects a dict-like object (model state dict)
# It accesses keys and modifies values which are tensors
# It looks for keys ending with 'relative_position_bias_table', 'attn_mask', 'relative_position_index'

# Create a mock model state dict with position embedding tables
model = {
    'layer1.relative_position_bias_table': torch.randn(23, 8),  # (2*12-1)^2 = 23, num_heads=8
    'layer2.relative_position_bias_table': torch.randn(23, 8),
    'layer1.attn_mask': torch.randn(2, 2),  # Will be popped
    'layer1.relative_position_index': torch.randn(2, 2),  # Will be popped
}

ARGS = {
    'model': model,
    'before': 384,
    'patch_size': 32,
    'after': 384,
    'suffix': 'relative_position_bias_table'
}
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
window_partition Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ShramanPramanick/VoLTA/Pre-training/swin_transformer.py
pointer only (licence: MIT); the stored copy get_code serves, from heekhero/DTL, is pointer only (licence: NONE) · get_code("993ae96666b00cb5")
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 heekhero/DTL, 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.

window_reverse Ran
ran; our contract draft was wrong, not the code
ShramanPramanick/VoLTA/Pre-training/swin_transformer.py
code served (permissive licence) · get_code("265c72d3ccc64544")
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, 8, 8, 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 torch

torch.manual_seed(0)

# Parameters for window_reverse
# windows: (num_windows*B, window_size, window_size, C)
# The function expects windows to be partitioned from an image of size (B, H, W, C)
# where the image is divided into non-overlapping windows of size window_size x window_size

B = 2  # batch size
H = 8  # image height
W = 8  # image width
window_size = 4  # window size
C = 3  # number of channels

# Calculate num_windows per batch: (H/window_size) * (W/window_size)
num_windows_per_batch = (H // window_size) * (W // window_size)  # (8/4) * (8/4) = 4
num_windows_total = num_windows_per_batch * B  # 4 * 2 = 8

# Create windows tensor with shape (num_windows*B, window_size, window_size, C)
# Shape: (8, 4, 4, 3)
windows = torch.randn(num_windows_total, window_size, window_size, C)

ARGS = {
    "windows": windows,
    "window_size": window_size,
    "H": H,
    "W": W,
}
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, 8, 8, 3) — shape (2, 8, 8, 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
apply_chunking_to_forward Not yet run
unverified (harvested, no recorded run)
ShramanPramanick/VoLTA/Pre-training/roberta.py
code served (permissive licence) · get_code("7a80cf522c830aeb")
How it ran

no invocation recorded.

resume_checkpoint Not yet run
unverified (harvested, no recorded run)
ShramanPramanick/VoLTA/Pre-training/swin_helpers.py
pointer only (licence: MIT); the stored copy get_code serves, from Wenhao-Chi/UNIEGO, is pointer only (licence: NOASSERTION) · get_code("102b75f3481035d1")
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 Wenhao-Chi/UNIEGO, and GitHub could not classify the licence of that repository (NOASSERTION). 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. No invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 11 are also present in other papers whose code we hold (10 with at least one paper dated earlier, 1 the same month, 6 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
window_partition 31 other papers
3 dated earlier, 1 the same month, 27 later
2111.11418 (2021-11)
MetaFormer Is Actually What You Need for Vision
2203.12119 (2022-03)
Visual Prompt Tuning
2204.03645 (2022-04)
DaViT: Dual Attention Vision Transformers
and 28 more not named here
create_position_ids_from_input_ids 28 other papers
10 dated earlier, 18 later
2107.00910 (2021-07)
Learned Token Pruning for Transformers
2107.14035 (2021-07)
ProtoTransformer: A Meta-Learning Approach to Providing Student Feedback
2201.11732 (2022-01)
IGLUE: A Benchmark for Transfer Learning across Modalities, Tasks, and Languages
and 25 more not named here
window_reverse 14 other papers
2 dated earlier, 12 later
1706.03762 (2017-06)
Attention Is All You Need
2204.03645 (2022-04)
DaViT: Dual Attention Vision Transformers
Li_I-ViT_Integer-only_Quantization_for_Efficient_Vision_Transformer_Inference_ICCV_2023_paper (2023 (year only))
I-ViT: Integer-only Quantization for Efficient Vision Transformer Inference
and 11 more not named here
resume_checkpoint 6 other papers
3 dated earlier, 3 later
2110.07641 (2021-10)
Non-deep Networks
2203.10233 (2022-03)
DirecFormer: A Directed Attention in Transformer Approach to Robust Action Recognition
2206.03452 (2022-06)
Can CNNs Be More Robust Than Transformers?
and 3 more not named here
create_positive_map 3 other papers
3 dated earlier
2112.03857 (2021-12)
Grounded Language-Image Pre-training
2206.05836 (2022-06)
GLIPv2: Unifying Localization and VL Understanding
2206.07643 (2022-06)
Coarse-to-Fine Vision-Language Pre-training with Fusion in the Backbone
create_positive_map_label_to_token_from_positive_map 3 other papers
3 dated earlier
2112.03857 (2021-12)
Grounded Language-Image Pre-training
2206.05836 (2022-06)
GLIPv2: Unifying Localization and VL Understanding
2206.07643 (2022-06)
Coarse-to-Fine Vision-Language Pre-training with Fusion in the Backbone
cost_matrix_torch 2 other papers
1 dated earlier, 1 later
2006.14744 (2020-06)
Graph Optimal Transport for Cross-Domain Alignment
2407.01851 (2024-07)
Meerkat: Audio-Visual Large Language Model for Grounding in Space and Time
IPOT_distance_torch 1 other paper
1 dated earlier
2006.14744 (2020-06)
Graph Optimal Transport for Cross-Domain Alignment
IPOT_torch 1 other paper
1 dated earlier
2006.14744 (2020-06)
Graph Optimal Transport for Cross-Domain Alignment
apply_chunking_to_forward 1 other paper
1 later
2311.18259 (2023-11)
Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives
load_state_dict 1 other paper
1 dated earlier
2110.07641 (2021-10)
Non-deep Networks

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

Vision-language pre-training (VLP) has recently proven highly effective for various uni- and multi-modal downstream applications. However, most existing end-to-end VLP methods use high-resolution image-text box data to perform well on fine-grained region-level tasks, such as object detection, segmentation, and referring expression comprehension. Unfortunately, such high-resolution images with accurate bounding box annotations are expensive to collect and use for supervision at scale. In this work, we propose VoLTA (Vision-Language Transformer with weakly-supervised local-feature Alignment), a new VLP paradigm that only utilizes image-caption data but achieves fine-grained region-level image understanding, eliminating the use of expensive box annotations. VoLTA adopts graph optimal transport-based weakly-supervised alignment on local image patches and text tokens to germinate an explicit, self-normalized, and interpretable low-level matching criterion. In addition, VoLTA pushes multi-modal fusion deep into the uni-modal backbones during pre-training and removes fusion-specific transformer layers, further reducing memory requirements. Extensive experiments on a wide range of vision- and vision-language downstream tasks demonstrate the effectiveness of VoLTA on fine-grained applications without compromising the coarse-grained downstream performance, often outperforming methods using significantly more caption and box annotations.

For agents

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

get_harvested_code_for_paper("2210.04135")
get_code_for_paper("2210.04135")
have("2210.04135")

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

curl https://syntology.ai/api/ran/2210.04135.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.04135.svg)](https://syntology.ai/paper/2210.04135)

Connect an agent — have() is free.