SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.07502 · NeurIPS · 2021

MULTIBENCH: Multiscale Benchmarks for Multimodal Representation Learning

Paul Liang, Ruslan Salakhutdinov, Louis-Philippe Morency, U Austin, Yuke Zhu, Cmu, Yiwei Lyu, Xiang Fan, Zetian Wu, Yun Cheng, Jason Wu, Leslie Chen, and 3 more

arXiv · PDF · Open in the Atlas

Code that ran

11 of 16 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 9 have no contract verdict; 1 failure of our instrument; 5 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/2107.07502.json · badge: /api/ran/2107.07502.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
mit-lcp/mimic-code canonical (classifier label)
classifier basis: census
4 ran · 2 no recorded run
pliang279/factorized canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
pliang279/MultiBench canonical (classifier label)
classifier basis: census
2 ran · 1 no recorded run
VahidooX/DeepCCA canonical (classifier label)
classifier basis: census
1 ran · 2 no recorded run
brentyi/multimodalfilter canonical (classifier label)
classifier basis: census
1 ran · 0 no recorded run
FunctionStatusWhere it lives
compare_table Ran
ran on a synthesized input
mit-lcp/mimic-code/src/mimic_utils/compare_concepts.py
code served (permissive licence) · get_code("07d31fd5cd3b490b")
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 pandas as pd
import numpy as np

# Create two dataframes with matching columns and data types
pg_df = pd.DataFrame({
    'id': [1, 2, 3, 4],
    'value': [1.5, 2.5, 3.5, 4.5],
    'date': pd.to_datetime(['2021-01-01', '2021-01-02', '2021-01-03', '2021-01-04']),
    'name': ['alice', 'bob', 'charlie', 'david']
})

duck_df = pd.DataFrame({
    'id': [1, 2, 3, 4],
    'value': [1.500001, 2.500001, 3.500001, 4.500001],
    'date': pd.to_datetime(['2021-01-01', '2021-01-02', '2021-01-03', '2021-01-04']),
    'name': ['alice', 'bob', 'charlie', 'david']
})

ARGS = {
    'pg_df': pg_df,
    'duck_df': duck_df,
    'rtol': 1e-5,
    'atol': 1e-5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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
compute_kernel Ran
ran; our contract draft was wrong, not the code
pliang279/factorized/mfm_model.py
pointer only (licence: MIT); the stored copy get_code serves, from osu-srml/cf_representation_learning, is pointer only (licence: NONE) · get_code("e0a46babceb34647")
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 osu-srml/cf_representation_learning, 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.

getallparams Ran
ran, honoured the contract we drafted
pliang279/MultiBench/special/kinetics_audio_unimodal.py
code served (permissive licence) · get_code("f86744b3991d473e")
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
import torch.nn as nn

torch.manual_seed(0)

# Create small neural network modules
class TinyModel1(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        return self.linear2(self.linear1(x))

class TinyModel2(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv = nn.Conv1d(2, 3, kernel_size=2)
        self.linear = nn.Linear(3, 1)
    
    def forward(self, x):
        return self.linear(self.conv(x))

class TinyModel3(nn.Module):
    def __init__(self):
        super().__init__()
        self.embedding = nn.Embedding(10, 5)
        self.linear = nn.Linear(5, 2)
    
    def forward(self, x):
        return self.linear(self.embedding(x))

# Create instances
model1 = TinyModel1()
model2 = TinyModel2()
model3 = TinyModel3()

# Build the list of modules
module_list = [model1, model2, model3]

# Define ARGS
ARGS = {
    "li": module_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 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
list_rindex Ran
ran on a synthesized input
mit-lcp/mimic-code/mimic-iv-cxr/txt/create_section_files.py
code served (permissive licence) · get_code("4d6b6588f49b7842")
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

def list_rindex(l, s):
    """Helper function: *last* matching element in a list"""
    return len(l) - l[-1::-1].index(s) - 1

# Construct valid arguments
# l: a list (any list will work)
# s: an element to search for (should be in the list)

l = [1, 2, 3, 2, 4, 2, 5]
s = 2

ARGS = {
    'l': l,
    's': s
}
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
loss_KLD Ran
ran on a synthesized input
pliang279/factorized/mfm_model.py
code served (permissive licence) · get_code("68f220ce165da1bf")
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. 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 small tensors for mu and logvar
# mu: mean of latent distribution, shape (batch_size, latent_dim)
# logvar: log variance of latent distribution, shape (batch_size, latent_dim)
mu = torch.randn(2, 4)  # batch_size=2, latent_dim=4
logvar = torch.randn(2, 4)  # batch_size=2, latent_dim=4

ARGS = {
    "mu": mu,
    "logvar": logvar,
}
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
loss_MMD Ran
ran on a synthesized input
pliang279/factorized/mfm_model.py
code served (permissive licence) · get_code("db4e55195900c65a")
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
from torch.autograd import Variable

torch.manual_seed(0)

# The function loss_MMD takes a single parameter: zy
# zy is used as:
# - zy.size() is called to create a tensor of same size
# - compute_kernel(zy, zy) is called, which expects a 2D tensor
# - compute_kernel(zy_real_gauss, zy) is called, which expects 2D tensors

# Create a small 2D tensor: batch_size=2, dim=4
zy = Variable(torch.randn(2, 4))

ARGS = {
    "zy": zy
}
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
multimodalcompute Ran
ran on a synthesized input
pliang279/MultiBench/special/kinetics_gradient_blend.py
code served (permissive licence) · get_code("2fe3e23ec5e8959a")
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
import torch.nn as nn

torch.manual_seed(0)

# Create simple neural network models
class SimpleModel(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, x):
        return self.linear(x)

# Create models for each modality
model1 = SimpleModel(4, 3)
model2 = SimpleModel(4, 5)
model3 = SimpleModel(4, 6)

models = [model1, model2, model3]

# Create corresponding input tensors for each model
# Each tensor has batch size 2 and feature dimension 4
train_x = [
    torch.randn(2, 4),
    torch.randn(2, 4),
    torch.randn(2, 4)
]

ARGS = {
    "models": models,
    "train_x": train_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 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
normalize_section_names Ran
ran on a synthesized input
mit-lcp/mimic-code/mimic-iv-cxr/txt/section_parser.py
code served (permissive licence) · get_code("49696211667f2162")
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 re

def normalize_section_names(section_names):
    section_names = [s.lower().strip() for s in section_names]
    frequent_sections = {'preamble': 'preamble', 'impression': 'impression', 'comparison': 'comparison', 'indication': 'indication', 'findings': 'findings', 'examination': 'examination', 'technique': 'technique', 'history': 'history', 'comparisons': 'comparison', 'clinical history': 'history', 'reason for examination': 'indication', 'notification': 'notification', 'reason for exam': 'indication', 'clinical information': 'history', 'exam': 'examination', 'clinical indication': 'indication', 'conclusion': 'impression', 'chest, two views': 'findings', 'recommendation(s)': 'recommendations', 'type of examination': 'examination', 'reference exam': 'comparison', 'patient history': 'history', 'addendum': 'addendum', 'comparison exam': 'comparison', 'date': 'date', 'comment': 'comment', 'findings and impression': 'impression', 'wet read': 'wet read', 'comparison film': 'comparison', 'recommendations': 'recommendations', 'findings/impression': 'impression', 'pfi': 'history', 'recommendation': 'recommendations', 'wetread': 'wet read', 'ndication': 'impression', 'impresson': 'impression', 'imprression': 'impression', 'imoression': 'impression', 'impressoin': 'impression', 'imprssion': 'impression', 'impresion': 'impression', 'imperssion': 'impression', 'mpression': 'impression', 'impession': 'impression', 'findings/ impression': 'impression', 'finding': 'findings', 'findins': 'findings', 'findindgs': 'findings', 'findgings': 'findings', 'findngs': 'findings', 'findnings': 'findings', 'finidngs': 'findings', 'idication': 'indication', 'reference findings': 'findings', 'comparision': 'comparison', 'comparsion': 'comparison', 'comparrison': 'comparison', 'comparisions': 'comparison'}
    p_findings = ['chest', 'portable', 'pa and lateral', 'lateral and pa', 'ap and lateral', 'lateral and ap', 'frontal and', 'two views', 'frontal view', 'pa view', 'ap view', 'one view', 'lateral view', 'bone window', 'frontal upright', 'frontal semi-upright', 'ribs', 'pa and lat']
    p_findings = re.compile('({})'.format('|'.join(p_findings)))
    main_sections = ['impression', 'findings', 'history', 'comparison', 'addendum']
    for i, s in enumerate(section_names):
        if s in frequent_sections:
            section_names[i] = frequent_sections[s]
            continue
        main_flag = False
        for m in main_sections:
            if m in s:
                section_names[i] = m
                main_flag = True
                break
        if main_flag:
            continue
        m = p_findings.search(s)
        if m is not None:
            section_names[i] = 'findings'
    return section_names

ARGS = {
    "section_names": [
        "Impression",
        "  FINDINGS  ",
        "Clinical History",
        "PA and Lateral",
        "Unknown Section",
        "chest portable",
        "Comparisons",
        "Wet Read"
    ]
}
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
process_dataframe Ran
ran on a synthesized input
mit-lcp/mimic-code/mimic-iv/buildmimic/sqlite/import.py
code served (permissive licence) · get_code("8457a015d3002b75")
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 DataFrame. 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 pandas as pd
import typing as t

# Create a sample DataFrame with various columns including time/date columns
df = pd.DataFrame({
    'subject_id': [1, 2, 3, 1, 2],
    'measurement_time': ['2023-01-01T10:00:00', '2023-01-02T11:00:00', '2023-01-03T12:00:00', '2023-01-04T13:00:00', '2023-01-05T14:00:00'],
    'created_date': ['2023-01-01', '2023-01-02', '2023-01-03', '2023-01-04', '2023-01-05'],
    'value': [10.5, 20.3, 15.7, 22.1, 18.9],
    'name': ['alice', 'bob', 'charlie', 'alice', 'bob']
})

# Create a list of subject IDs to filter by
subjects = [1, 2]

ARGS = {
    'df': df,
    'subjects': subjects
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
svm_classify Ran
ran on a synthesized input
VahidooX/DeepCCA/utils.py
code served (permissive licence) · get_code("a53476caeab3f698")
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 numpy as np
from sklearn import svm
from sklearn.metrics import accuracy_score

# Create small synthetic data
np.random.seed(0)

# Training data: (features, _, labels)
train_features = np.random.randn(10, 4)  # 10 samples, 4 features
train_labels = np.random.randint(0, 2, (10, 1))  # Binary labels

# Validation data
valid_features = np.random.randn(6, 4)
valid_labels = np.random.randint(0, 2, (6, 1))

# Test data
test_features = np.random.randn(8, 4)
test_labels = np.random.randint(0, 2, (8, 1))

# Construct data parameter: list of 3 tuples (train, valid, test)
# Each tuple is (features, _, labels)
data = [
    (train_features, None, train_labels),
    (valid_features, None, valid_labels),
    (test_features, None, test_labels)
]

ARGS = {
    'data': data,
    'C': 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — 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
weighted_average Ran
ran on a synthesized input
brentyi/multimodalfilter/crossmodal/base_models/utility.py
code served (permissive licence) · get_code("4da5d48e575e59b6")
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 1, dtype family float, shape (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 small tensors with distinct dimensions
# predictions: shape (2, 4) - 2 samples, 4 dimensions
predictions = torch.randn(2, 4)

# weights: shape (2, 4) - must match predictions shape
weights = torch.rand(2, 4) + 0.1  # Add 0.1 to ensure non-zero weights

ARGS = {
    "predictions": predictions,
    "weights": weights,
}
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 1, dtype family float, shape (4,) — shape (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
CreateOp Not yet run
unverified (harvested, no recorded run)
pliang279/MultiBench/utils/aux_models.py
code served (permissive licence) · get_code("7ee0d14726bfe8b5")
How it ran

no invocation recorded.

compare_concepts Not yet run
unverified (harvested, no recorded run)
mit-lcp/mimic-code/src/mimic_utils/compare_concepts.py
code served (permissive licence) · get_code("e09d350937192171")
How it ran

no invocation recorded.

linear_cca Not yet run
unverified (harvested, no recorded run)
VahidooX/DeepCCA/linear_cca.py
code served (permissive licence) · get_code("102a690f83452199")
How it ran

no invocation recorded.

section_text Not yet run
unverified (harvested, no recorded run)
mit-lcp/mimic-code/mimic-iv-cxr/txt/section_parser.py
code served (permissive licence) · get_code("6bac6c790b33b690")
How it ran

no invocation recorded.

train_model Not yet run
unverified (harvested, no recorded run)
VahidooX/DeepCCA/DeepCCA.py
code served (permissive licence) · get_code("ce081e7ccd4cc697")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 6 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 4 later) and 10 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
normalize_section_names 3 other papers
2 later, 1 same year, order not determinable
2021.acl-long.459 (2021 (year only))
Cross-modal Memory Networks for Radiology Report Generation
2209.07162 (2022-09)
Brain Imaging Generation with Latent Diffusion Models
2304.08295 (2023-04)
Interactive and Explainable Region-guided Radiology Report Generation
CreateOp 2 other papers
2 later
2309.06612 (2023-09)
Harmonic-NAS: Hardware-Aware Multimodal Neural Architecture Search on Resource-constrained Devices
2405.17613 (2024-05)
Jointly Modeling Inter-& Intra-Modality Dependencies for Multi-modal Learning
compute_kernel 2 other papers
1 dated earlier, 1 later
1904.08194 (2019-04)
Effective Estimation of Deep Generative Language Models
2311.05420 (2023-11)
Counterfactually Fair Representation
getallparams 2 other papers
2 later
2212.01241 (2022-12)
MMBench: Benchmarking End-to-End Multi-modal DNNs and Understanding Their Hardware-Software Implications
2405.17613 (2024-05)
Jointly Modeling Inter-& Intra-Modality Dependencies for Multi-modal Learning
list_rindex 1 other paper
1 same year, order not determinable
2021.acl-long.459 (2021 (year only))
Cross-modal Memory Networks for Radiology Report Generation
svm_classify 1 other paper
1 dated earlier
2106.07479 (2021-06)
An Online Riemannian PCA for Stochastic Canonical Correlation Analysis

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

Learning multimodal representations involves integrating information from multiple heterogeneous sources of data. It is a challenging yet crucial area with numerous real-world applications in multimedia, affective computing, robotics, finance, human-computer interaction, and healthcare. Unfortunately, multimodal research has seen limited resources to study (1) generalization across domains and modalities, (2) complexity during training and inference, and (3) robustness to noisy and missing modalities. In order to accelerate progress towards understudied modalities and tasks while ensuring real-world robustness, we release MULTIBENCH, a systematic and unified large-scale benchmark for multimodal learning spanning 15 datasets, 10 modalities, 20 prediction tasks, and 6 research areas. MULTIBENCH provides an automated end-to-end machine learning pipeline that simplifies and standardizes data loading, experimental setup, and model evaluation. To enable holistic evaluation, MULTIBENCH offers a comprehensive methodology to assess (1) generalization, (2) time and space complexity, and (3) modality robustness. MULTIBENCH introduces impactful challenges for future research, including scalability to large-scale multimodal datasets and robustness to realistic imperfections. To accompany this benchmark, we also provide a standardized implementation of 20 core approaches in multimodal learning spanning innovations in fusion paradigms, optimization objectives, and training approaches. Simply applying methods proposed in different research areas can improve the state-of-the-art performance on 9 15 datasets. Therefore, MULTIBENCH presents a milestone in unifying disjoint efforts in multimodal machine learning research and paves the way towards a better understanding of the capabilities and limitations of multimodal models, all the while ensuring ease of use, accessibility, and reproducibility. MULTIBENCH, our standardized implementations, and leaderboards are publicly available, will be regularly updated, and welcomes inputs from the community.

For agents

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

get_harvested_code_for_paper("2107.07502")
get_code_for_paper("2107.07502")
have("2107.07502")

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

curl https://syntology.ai/api/ran/2107.07502.json

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

[![Syntology run record](https://syntology.ai/api/ran/2107.07502.svg)](https://syntology.ai/paper/2107.07502)

Connect an agent — have() is free.