SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.15263 · EMNLP · 2024

Back to School: Translation Using Grammar Books

Antonios Anastasopoulos, Jonathan Hus

arXiv · PDF · Open in the Atlas

Code that ran

2 of 6 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 0 have no contract verdict; 2 failures of our instrument; 4 no recorded run), measured 2026-11-22; 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/2410.15263.json · badge: /api/ran/2410.15263.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
jonathanhus/back-to-school canonical (classifier label)
classifier basis: census
2 ran · 4 no recorded run
FunctionStatusWhere it lives
count_grammar_book Ran
ran; our contract draft was wrong, not the code
jonathanhus/back-to-school/utilities/count_data.py
code served (permissive licence) · get_code("63195ea86c7b2b4f")
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 os
import tempfile

# Create a temporary directory structure to simulate the resources
temp_dir = tempfile.mkdtemp()
resources_dir = os.path.join(temp_dir, 'resources', 'english')
os.makedirs(resources_dir, exist_ok=True)

# Create a sample grammar book file
grammar_book_path = os.path.join(resources_dir, 'grammar_book_for_claude_long.txt')
with open(grammar_book_path, 'w') as f:
    f.write("The quick brown fox jumps over the lazy dog\n")
    f.write("Grammar is the study of words and their functions\n")
    f.write("This file contains sample text for testing\n")

# Patch the resources_dir lookup to use our temp directory
import sys
from unittest.mock import patch

def count_grammar_book_wrapper(language):
    resources_dir = os.path.join(temp_dir, 'resources', language) + '/'
    stats = {'language': language}
    token_count = 0
    try:
        with open(resources_dir + 'grammar_book_for_claude_long.txt') as file:
            for line in file:
                tokens = line.split()
                token_count += len(tokens)
            stats['words'] = token_count
    except FileNotFoundError:
        stats['words'] = 'No grammar book'
    return stats

ARGS = {
    'language': 'english'
}
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
write_common_output Ran
ran; our contract draft was wrong, not the code
jonathanhus/back-to-school/utilities/get_common_sentences.py
code served (permissive licence) · get_code("b12d2c2268d2c019")
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 json
import os
import tempfile

# Create a temporary directory for test files
test_dir = tempfile.mkdtemp()

# Create test data file
test_data = [
    {"id": 1, "text": "example one"},
    {"id": 2, "text": "example two"},
    {"id": 3, "text": "example three"},
    {"id": 4, "text": "example four"},
]
test_filename = "test_data.json"
test_filepath = os.path.join(test_dir, test_filename)
with open(test_filepath, 'w') as f:
    json.dump(test_data, f)

# Define arguments
common_filename = "common_output.json"
common_set = [0, 2]  # indices to extract
dict_map = {0: 0, 2: 2}  # mapping from common_set elements to test_data indices

ARGS = {
    "test_dir": test_dir + "/",
    "test_filename": test_filename,
    "common_filename": common_filename,
    "common_set": common_set,
    "dict_map": dict_map,
}
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
count_sentences Not yet run
unverified (harvested, no recorded run)
jonathanhus/back-to-school/utilities/count_data.py
code served (permissive licence) · get_code("bd1960932c46719e")
How it ran

no invocation recorded.

count_words Not yet run
unverified (harvested, no recorded run)
jonathanhus/back-to-school/utilities/count_data.py
code served (permissive licence) · get_code("bdd4c45f6daf0dab")
How it ran

no invocation recorded.

get_indices_dict Not yet run
unverified (harvested, no recorded run)
jonathanhus/back-to-school/utilities/get_common_sentences.py
code served (permissive licence) · get_code("b00b30f2d1ae0f3d")
How it ran

no invocation recorded.

summarize_grammar Not yet run
unverified (harvested, no recorded run)
jonathanhus/back-to-school/utilities/format_grammar_book.py
code served (permissive licence) · get_code("7d95d91a47fdb095")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 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 6 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

Machine translation systems for high resource languages perform exceptionally well and produce high quality translations. Unfortunately, the vast majority of languages lack the quantity of parallel sentences needed to train such systems. These under-represented languages are not entirely without resources, as bilingual dictionaries and grammar books may be available as linguistic reference material. With current large language models (LLMs) supporting near book-length contexts, we can use the available material to ensure advancements are shared among all of the world's languages. In this paper, we use dictionaries and grammar books to improve machine translation. We evaluate on 16 typologically diverse low-resource languages, showing encouraging improvements. 1

For agents

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

get_harvested_code_for_paper("2410.15263")
get_code_for_paper("2410.15263")
have("2410.15263")

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

curl https://syntology.ai/api/ran/2410.15263.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.15263.svg)](https://syntology.ai/paper/2410.15263)

Connect an agent — have() is free.