SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.04672 · 2022

No Language Left Behind: Scaling Human-Centered Machine Translation

U Berkeley, James Cross, Angela Fan, Vedanuj Goswami, Philipp Koehn, Kenneth Heafield, Onur Çelebi, Sergey Edunov, Shruti Bhosale, Francisco Guzmán, Janice Lam, Alexandre Mourachko, and 42 more

arXiv · PDF · Open in the Atlas

Code that ran

6 of 10 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 6 have no contract verdict; 0 failures of our instrument; 4 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/2207.04672.json · badge: /api/ran/2207.04672.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
ragerri/antidote-projections community (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
zurichnlp/nmtscore community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
xhluca/dl-translate community (classifier label)
classifier basis not recorded on the link
0 ran · 3 no recorded run
FunctionStatusWhere it lives
count_number_of_sentences Ran
ran on a synthesized input
ragerri/antidote-projections/cross-lingual-argument-mining/correction/manual_correction/sentence_reader.py
code served (permissive licence) · get_code("b69c18b1f7c30d9e")
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 os
import tempfile
from typing import TextIO

# Create a temporary file with sample NER data in BIO format
def create_sample_ner_file():
    temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt', encoding='utf8')
    
    # Write sample sentences in BIO format (word tag pairs)
    # Sentence 1
    temp_file.write("The O\n")
    temp_file.write("cat B-ANIMAL\n")
    temp_file.write("sat O\n")
    temp_file.write("on O\n")
    temp_file.write("the O\n")
    temp_file.write("mat O\n")
    temp_file.write("\n")  # Empty line separates sentences
    
    # Sentence 2
    temp_file.write("John B-PERSON\n")
    temp_file.write("walked O\n")
    temp_file.write("to O\n")
    temp_file.write("Paris B-LOCATION\n")
    temp_file.write("\n")
    
    # Sentence 3
    temp_file.write("She O\n")
    temp_file.write("saw B-PERSON\n")
    temp_file.write("Mary I-PERSON\n")
    temp_file.write("yesterday O\n")
    temp_file.write("\n")
    
    temp_file.close()
    return temp_file.name

# Create the sample file
sample_file_path = create_sample_ner_file()

ARGS = {
    "input_path": sample_file_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
get_line Ran
ran on a synthesized input
ragerri/antidote-projections/cross-lingual-argument-mining/correction/manual_correction/sentence_reader.py
code served (permissive licence) · get_code("98026a3a5a6a4584")
How it ran

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

import io
from typing import List, TextIO

# Create a StringIO object to simulate a file
file_content = "  hello   world  \n"
file_obj = io.StringIO(file_content)

ARGS = {
    "file": file_obj
}
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 — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_sentence Ran
ran on a synthesized input
ragerri/antidote-projections/cross-lingual-argument-mining/correction/manual_correction/sentence_reader.py
code served (permissive licence) · get_code("f972e3a23421e756")
How it ran

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

import io
from typing import List, TextIO

# Create a TextIO object with sample NER-formatted data
sample_data = """The B-PER
Smith B-PER
is O
a O
doctor B-PROF
in O
New B-LOC
York B-LOC
City I-LOC
"""

file_obj = io.StringIO(sample_data)

ARGS = {
    "file": file_obj
}
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 — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_json Ran
ran on a synthesized input
zurichnlp/nmtscore/src/nmtscore/models/tokenization_small100.py
code served (permissive licence) · get_code("aba2bc51d649c01d")
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 json
import tempfile
import os

# Create a temporary JSON file for testing
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.json")

# Create a valid JSON file with test data
test_data = {"key": "value", "number": 42, "list": [1, 2, 3]}
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 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
read_file Ran
ran on a synthesized input
ragerri/antidote-projections/cross-lingual-argument-mining/correction/post_processing/projection_corrections.py
code served (permissive licence) · get_code("8f9afa32c236ac4c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 3, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tempfile
import os

# Create a temporary file with the expected format
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("token1 tag1\n")
temp_file.write("token2 tag2\n")
temp_file.write("\n")
temp_file.write("token3 tag3\n")
temp_file.write("token4 tag4\n")
temp_file.write("\n")
temp_file.flush()
temp_file.close()

ARGS = {
    'file': temp_file.name
}
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 3, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
to_sentence Ran
ran on a synthesized input
ragerri/antidote-projections/cross-lingual-argument-mining/translation/translate.py
code served (permissive licence) · get_code("cbe5fb8f67ccfbed")
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 TextIOWrapper. 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 tempfile
import os

# Create temporary files for input and output
temp_dir = tempfile.mkdtemp()
input_file = os.path.join(temp_dir, "input.tsv")
output_file = os.path.join(temp_dir, "output.txt")

# Create a sample input TSV file with tokens and newlines
with open(input_file, 'w') as f:
    f.write("hello world\n")
    f.write("this is\n")
    f.write("a test\n")
    f.write("\n")
    f.write("another sentence\n")
    f.write("here\n")
    f.write("\n")

ARGS = {
    'input_file': input_file,
    'output_file': output_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 TextIOWrapper — a TextIOWrapper 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
available_codes Not yet run
unverified (harvested, no recorded run)
xhluca/dl-translate/dl_translate/utils.py
code served (permissive licence) · get_code("b5baae4eda803c23")
How it ran

no invocation recorded.

available_languages Not yet run
unverified (harvested, no recorded run)
xhluca/dl-translate/dl_translate/utils.py
code served (permissive licence) · get_code("ed005bfb5ec22688")
How it ran

no invocation recorded.

count_lines Not yet run
unverified (harvested, no recorded run)
ragerri/antidote-projections/cross-lingual-argument-mining/correction/manual_correction/utils.py
code served (permissive licence) · get_code("0605d52f9bcf9662")
How it ran

no invocation recorded.

get_lang_code_map Not yet run
unverified (harvested, no recorded run)
xhluca/dl-translate/dl_translate/utils.py
code served (permissive licence) · get_code("f1688a767a8dce8c")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 9 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
load_json 2 other papers
1 dated earlier, 1 later
2010.11125 (2020-10)
Beyond English-Centric Multilingual Machine Translation
2210.11621 (2022-10)
SMaLL-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languages

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

Driven by the goal of eradicating language barriers on a global scale, machine translation has solidified itself as a key focus of artificial intelligence research today. However, such efforts have coalesced around a small subset of languages, leaving behind the vast majority of mostly low-resource languages. What does it take to break the 200 language barrier while ensuring safe, high quality results, all while keeping ethical considerations in mind? In No Language Left Behind, we took on this challenge by first contextualizing the need for low-resource language translation support through exploratory interviews with native speakers. Then, we created datasets and models aimed at narrowing the performance gap between low and high-resource languages. More specifically, we developed a conditional compute model based on Sparsely Gated Mixture of Experts that is trained on data obtained with novel and effective data mining techniques tailored for low-resource languages. We propose multiple architectural and training improvements to counteract overfitting while training on thousands of tasks. Critically, we evaluated the performance of over 40,000 different translation directions using a human-translated benchmark, Flores-200, and combined human evaluation with a novel toxicity benchmark covering all languages in Flores-200 to assess translation safety. Our model achieves an improvement of 44% BLEU relative to the previous state-of-the-art, laying important groundwork towards realizing a universal translation system. Finally, we open source all contributions described in this work, accessible at https://github.com/facebookresearch/fairseq/tree/nllb.

For agents

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

get_harvested_code_for_paper("2207.04672")
get_code_for_paper("2207.04672")
have("2207.04672")

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

curl https://syntology.ai/api/ran/2207.04672.json

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

[![Syntology run record](https://syntology.ai/api/ran/2207.04672.svg)](https://syntology.ai/paper/2207.04672)

Connect an agent — have() is free.