SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2201.05609 · 2022

Multilingual Open Text Release 1: Public Domain News in 44 Languages

arXiv · PDF · Open in the Atlas

Code that ran

13 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; 0 honoured the contract we drafted, 0 violated it and 13 have no contract verdict; 0 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/2201.05609.json · badge: /api/ran/2201.05609.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
bltlab/mot canonical (classifier label)
classifier basis not recorded on the link
12 ran · 1 no recorded run
VietHoang1512/khmer-nltk community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
FunctionStatusWhere it lives
author_allowed Ran
ran on a synthesized input
bltlab/mot/extraction/dump_documents.py
code served (permissive licence) · get_code("ac259c52c934e834")
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
import numpy as np
from typing import Dict, AbstractSet, Any

# Construct test data for author_allowed function

# Create a sample document with utag_data and authors
doc = {
    'utag_data': {
        'byline': 'John Smith, Jane Doe'
    },
    'authors': ['John Smith', 'Jane Doe', 'Reuters']
}

# Create a set of bad bylines to check against
bad_bylines = {'AFP', 'AP', 'Reuters'}

# Define ARGS mapping all parameters
ARGS = {
    'doc': doc,
    'bad_bylines': bad_bylines
}
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 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
average_chars Ran
ran on a synthesized input
bltlab/mot/qualitychecks/character_average.py
code served (permissive licence) · get_code("6bd8e8b915fc59a5")
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 json
import tempfile
import os

# Create a temporary JSON file with the required structure
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_data.json")

test_data = {
    "site_language": "English",
    "n_chars": 42
}

with open(temp_file, 'w') as f:
    json.dump(test_data, f)

ARGS = {
    "path": temp_file
}
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 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
clean_fields Ran
ran on a synthesized input
bltlab/mot/extraction/downloadsitemaps.py
code served (permissive licence) · get_code("1cf40a8bef19a322")
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 torch
import torch.nn as nn
from typing import List, Optional, Generator, Iterable, Dict

torch.manual_seed(0)

# Construct a value for the 'fields' parameter
# The function expects a List[str] and processes each by stripping whitespace
fields_value = [
    "  hello  ",
    "world",
    "   ",
    "\ttabbed\n",
    "",
    "no_whitespace"
]

ARGS = {
    "fields": fields_value
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family 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
confident_multiple_languages Ran
ran on a synthesized input
bltlab/mot/extraction/extracttext.py
code served (permissive licence) · get_code("3c6c78775c3fb35c")
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.

# Setup code for confident_multiple_languages function

ARGS = {
    "cld3_predictions": {
        "en": {
            "probability": 0.95,
            "proportion": 0.08
        },
        "fr": {
            "probability": 0.92,
            "proportion": 0.06
        },
        "de": {
            "probability": 0.85,
            "proportion": 0.03
        }
    }
}
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 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
count_chars Ran
ran on a synthesized input
bltlab/mot/qualitychecks/character_count.py
code served (permissive licence) · get_code("0dcddfcbdeb0e1b8")
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 json
import tempfile
import os

# Create a temporary JSON file with the required structure
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test_data.json")

test_data = {
    "site_language": "English",
    "n_chars": 1024
}

with open(temp_file, 'w', encoding='utf8') as f:
    json.dump(test_data, f)

ARGS = {
    "path": temp_file
}
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 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
filter_confident_langs Ran
ran on a synthesized input
bltlab/mot/extraction/extracttext.py
code served (permissive licence) · get_code("cb050e89f262217f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

# Create sample cld3 predictions dictionary
# Structure: {lang_code: {'probability': float, 'proportion': float, ...}}
cld3_predictions = {
    'en': {'probability': 0.95, 'proportion': 0.75, 'language': 'English'},
    'es': {'probability': 0.88, 'proportion': 0.005, 'language': 'Spanish'},
    'fr': {'probability': 0.92, 'proportion': 0.02, 'language': 'French'},
    'de': {'probability': 0.85, 'proportion': 0.008, 'language': 'German'},
    'it': {'probability': 0.91, 'proportion': 0.015, 'language': 'Italian'},
}

ARGS = {
    'cld3_predictions': cld3_predictions,
    'probability_threshold': 0.9,
    'proportion_threshold': 0.01,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_publication_date_from_utag Ran
ran on a synthesized input
bltlab/mot/extraction/utils.py
code served (permissive licence) · get_code("83d45f7a39a14210")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

from datetime import datetime
from typing import Dict, Optional

# Test case 1: Valid complete utag_data with all fields
utag_data_complete = {
    'pub_year': '2016',
    'pub_month': '03',
    'pub_day': '15',
    'pub_hour': '10',
    'pub_minute': '30'
}

# Test case 2: Valid utag_data with only year, month, day
utag_data_partial = {
    'pub_year': '2020',
    'pub_month': '07',
    'pub_day': '22'
}

# Test case 3: Valid utag_data with only year and month (should fallback)
utag_data_minimal = {
    'pub_year': '2018',
    'pub_month': '05'
}

# Test case 4: Empty/invalid utag_data
utag_data_empty = {}

# Use the complete valid case for ARGS
ARGS = {
    'utag_data': utag_data_complete
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
gunzip Ran
ran on a synthesized input
bltlab/mot/extraction/downloadsitemaps.py
code served (permissive licence) · get_code("8905f3eb7b69542e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import gzip
import os
import tempfile
import shutil

# Create a temporary directory for our test
temp_dir = tempfile.mkdtemp()

# Create a test file and gzip it
test_content = b"This is test content for gunzip"
original_file = os.path.join(temp_dir, "test_file.txt")
gzip_file = os.path.join(temp_dir, "test_file.txt.gz")

# Write original content and compress it
with open(original_file, 'wb') as f:
    f.write(test_content)

with open(original_file, 'rb') as f_in:
    with gzip.open(gzip_file, 'wb') as f_out:
        shutil.copyfileobj(f_in, f_out)

# Remove the original file so we only have the .gz version
os.remove(original_file)

# Now set up the ARGS dict with the filepath to the gzip file
ARGS = {
    'filepath': gzip_file
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
is_valid Ran
ran on a synthesized input
bltlab/mot/extraction/scraper.py
code served (permissive licence) · get_code("7916cce5b903f3c4")
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 = {
    "text": "This is a valid paragraph of text that should pass the validation check."
}
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
post_process Ran
ran on a synthesized input
VietHoang1512/khmer-nltk/khmernltk/utils/data.py
code served (permissive licence) · get_code("365063f7b2778eda")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def post_process(text, separator):
    text = text.strip(separator)
    return re.sub(f'(?:{separator}| )+', separator, text)

# Setup code
text = "  hello---world---test  "
separator = "-"

ARGS = {
    "text": text,
    "separator": separator
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
predict_language Ran
ran on a synthesized input
bltlab/mot/extraction/extracttext.py
code served (permissive licence) · get_code("0df7e49ac86de638")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from typing import Dict, Any

# Test case 1: Single language with high confidence
cld3_predictions_single = {
    'eng': {
        'probability': 0.95,
        'proportion': 0.98
    }
}

# Test case 2: Multiple languages with high confidence
cld3_predictions_multiple = {
    'eng': {
        'probability': 0.92,
        'proportion': 0.50
    },
    'fra': {
        'probability': 0.88,
        'proportion': 0.45
    }
}

# Test case 3: Empty predictions
cld3_predictions_empty = {}

# Test case 4: English prediction with non-english ISO
cld3_predictions_eng_only = {
    'eng': {
        'probability': 0.93,
        'proportion': 0.99
    }
}

# Define ARGS with all parameter values for predict_language function
ARGS = {
    'iso': 'fra',
    'cld3_predictions': cld3_predictions_single
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_voa_domains Ran
ran on a synthesized input
bltlab/mot/extraction/downloadsitemaps.py
code served (permissive licence) · get_code("6b3a7b351209d121")
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 tempfile
import os

# Create a temporary TSV file with valid VOA domain data
temp_dir = tempfile.mkdtemp()
domain_tsv_path = os.path.join(temp_dir, "domains.tsv")

# Write a sample VOA domains TSV file
with open(domain_tsv_path, 'w', encoding='utf8') as f:
    f.write("Region\tISO\tLanguage\tSite Name\tNews\tRadio\tVideo\tTV\tDirect\n")
    f.write("Africa\ten\tEnglish\tVOA English\thttps://www.voanews.com\thttps://www.voanews.com/radio\thttps://www.voanews.com/video\thttps://www.voanews.com/tv\thttps://www.voanews.com/direct\n")
    f.write("\tpt\tPortuguese\tVOA Portuguese\thttps://www.voaportugues.com\t\t\t\t\n")
    f.write("Asia\tzh\tChinese\tVOA Chinese\thttps://www.voachina.com\thttps://www.voachina.com/radio\t\t\t\n")

ARGS = {
    "domain_tsv_path": domain_tsv_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
attr — 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
stringify_mongovalues Ran
ran on a synthesized input
bltlab/mot/extraction/dump_documents.py
code served (permissive licence) · get_code("911b9a345e873933")
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.

from datetime import datetime
from typing import Any

def stringify_mongovalues(v: Any):
    """
    Mostly the values are fine, but at least datetime needs handled.
    Also need to handle datetimes embedded in dicts, so use recursion to get them.
    """
    if type(v) == datetime:
        return v.isoformat()
    elif type(v) == dict:
        return {key: stringify_mongovalues(value) for key, value in v.items()}
    else:
        return v


# Test case 1: datetime object
test_datetime = datetime(2023, 5, 15, 10, 30, 45)

# Test case 2: dict with nested datetime
test_dict_with_datetime = {
    "name": "test",
    "timestamp": datetime(2023, 5, 15, 10, 30, 45),
    "nested": {
        "created": datetime(2023, 1, 1, 0, 0, 0),
        "value": 42
    }
}

# Test case 3: simple value (string, int, etc.)
test_simple_value = "simple_string"

# For the function parameter 'v', we'll use a complex dict with nested datetimes
ARGS = {
    "v": {
        "id": 12345,
        "timestamp": datetime(2023, 5, 15, 10, 30, 45),
        "data": {
            "created_at": datetime(2023, 1, 1, 12, 0, 0),
            "name": "example",
            "nested_more": {
                "updated": datetime(2023, 6, 20, 14, 30, 0),
                "count": 100
            }
        },
        "simple_field": "value"
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
doc_to_dict Not yet run
unverified (harvested, no recorded run)
bltlab/mot/extraction/dump_documents.py
code served (permissive licence) · get_code("4adb3f5772a9e273")
How it ran

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, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 14 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.

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 present Multilingual Open Text (MOT), a new multilingual corpus containing text in 44 languages, many of which have limited existing text resources for natural language processing. The first release of the corpus contains over 2.8 million news articles and an additional 1 million short snippets (photo captions, video descriptions, etc.) published between 2001--2022 and collected from Voice of America's news websites. We describe our process for collecting, filtering, and processing the data. The source material is in the public domain, our collection is licensed using a creative commons license (CC BY 4.0), and all software used to create the corpus is released under the MIT License. The corpus will be regularly updated as additional documents are published.

For agents

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

get_harvested_code_for_paper("2201.05609")
get_code_for_paper("2201.05609")
have("2201.05609")

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

curl https://syntology.ai/api/ran/2201.05609.json

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

[![Syntology run record](https://syntology.ai/api/ran/2201.05609.svg)](https://syntology.ai/paper/2201.05609)

Connect an agent — have() is free.