SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1906.02780 · 2019

Syntactically Supervised Transformers for Faster Neural Machine Translation

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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 4 have no contract verdict; 1 failure 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/1906.02780.json · badge: /api/ran/1906.02780.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
dojoteef/synst canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
FunctionStatusWhere it lives
add_parse_transformer_args Ran
ran on a synthesized input
dojoteef/synst/args.py
code served (permissive licence) · get_code("0827b4cdb7e8b84d")
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 ArgGroup. 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 argparse
import torch

torch.manual_seed(0)

# Create a minimal argparse parser with argument groups
parser = argparse.ArgumentParser()

# This is the only parameter to add_parse_transformer_args
ARGS = {
    'parser': parser
}
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 ArgGroup — a ArgGroup 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
add_transformer_args Ran
ran on a synthesized input
dojoteef/synst/args.py
code served (permissive licence) · get_code("34bb40508e810900")
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 ArgGroup. 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 argparse
import json

class ArgGroup(object):
    """ A simple class wrapping argparse groups """

    def __init__(self, group):
        self.args = []
        self.dict = {}
        self.group = group

    def __getattr__(self, name):
        """ Try to read an argument """
        if name == 'dict':
            return self.dict
        else:
            return self.dict.get(name)

    def __str__(self):
        """ Return a json representation of the arguments """
        return json.dumps(self.dict, indent=1)

    def set_defaults(self, **kwargs):
        """ Set the defaults based on the kwargs """
        self.dict.update(kwargs)

    def read(self, args):
        """ Read in the parsed args that are part of this group """
        self.dict.update({k: args.__dict__[k] for k in self.args if k in args.__dict__})
        return self

    def add_argument(self, *args, **kwargs):
        """ Add an argument to group """
        actions = set(self.group._actions)
        self.group.add_argument(*args, **kwargs)
        for action in set(self.group._actions) - actions:
            self.args.append(action.dest)

def add_transformer_args(parser):
    """ Defines Transformer model specific arguments """
    group = ArgGroup(parser.add_argument_group('Transformer Model'))
    group.add_argument('--num-layers', type=int, default=6, help='Number of layers in each Transformer stack')
    group.add_argument('--num-heads', type=int, default=8, help='Number of heads in each Transformer layer for multi-headed attention')
    group.add_argument('--embedding-size', type=int, default=512, help='The size of the Transformer model dimension')
    group.add_argument('--hidden-dim', type=int, default=2048, help='The size of the Transformer feed-forward hidden layer')
    group.add_argument('--span', type=int, default=1, help='How many tokens to decode at once')
    return group

parser = argparse.ArgumentParser()

ARGS = {
    'parser': parser
}
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 ArgGroup — a ArgGroup 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
checkpoint Ran
ran; our contract draft was wrong, not the code
dojoteef/synst/models/utils.py
code served (permissive licence) · get_code("189298fec1e80029")
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 torch.nn as nn
import tempfile
import os

torch.manual_seed(0)

# Create a simple nn.Module for testing
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

class SimpleOptimizer:
    def __init__(self):
        self.param_groups = [{'params': []}]
    
    def state_dict(self):
        return {'state': {}, 'param_groups': self.param_groups}

# Create temporary directory for checkpoint storage
temp_dir = tempfile.mkdtemp()

# Construct the modules dict with objects that support state_dict()
model = SimpleModel()
optimizer = SimpleOptimizer()

modules = {
    'model': model,
    'optimizer': optimizer
}

ARGS = {
    'epoch': 5,
    'step': 1000,
    'modules': modules,
    'directory': temp_dir,
    'filename': 'checkpoint.pt',
    'max_checkpoints': 5
}
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 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
compute_f1 Ran
ran on a synthesized input
dojoteef/synst/analysis.py
code served (permissive licence) · get_code("315cda1182f75c8e")
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 float, 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
from collections import Counter

torch.manual_seed(0)

# The function compute_f1 takes two parameters:
# - reference: a sequence (list, tuple, or similar) that Counter can consume
# - prediction: a sequence (list, tuple, or similar) that Counter can consume
# 
# The function uses:
# - Counter(prediction) & Counter(reference) - intersection operation
# - len(reference) - length of reference
# - len(prediction) - length of prediction
# - sum(matches.values()) - sum of counter values
#
# So both parameters need to be sequences of hashable items

reference = ['cat', 'dog', 'bird', 'cat', 'fish']
prediction = ['cat', 'dog', 'lizard', 'cat', 'dog']

ARGS = {
    'reference': reference,
    'prediction': prediction
}
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 float, 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
parse_args Ran
ran on a synthesized input
dojoteef/synst/analysis.py
code served (permissive licence) · get_code("3c81cc5b8129690e")
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 Namespace. 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 argparse
import tempfile
import os

# Create temporary files for reference and prediction
with tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt') as ref_file:
    ref_file.write("This is a reference text for testing.\n")
    reference_path = ref_file.name

with tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt') as pred_file:
    pred_file.write("This is a predicted text for testing.\n")
    prediction_path = pred_file.name

with tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt') as bpe_file:
    bpe_file.write("l o w </w>\n")
    bpe_file.write("l o w e r </w>\n")
    bpe_path = bpe_file.name

# Build argument list for parse_args
ARGS = {
    'argv': [
        '-b', bpe_path,
        '-s', '8',
        '--preprocess-buffer-size', '2000',
        '-vv',
        reference_path,
        prediction_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 Namespace — a Namespace output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Namespace 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
restore Ran
ran, honoured the contract we drafted
dojoteef/synst/models/utils.py
code served (permissive licence) · get_code("c21d5cfb9ef52750")
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 int; 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 os
import torch
from torch import nn
import tempfile

torch.manual_seed(0)

# Create a simple model
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Create a simple optimizer state dict container
class SimpleOptimizer:
    def __init__(self):
        self.state_dict_data = {'lr': 0.001}
    
    def load_state_dict(self, state):
        self.state_dict_data = state

# Create temporary checkpoint file
model = SimpleModel()
optimizer = SimpleOptimizer()

# Create state dict to save
state = {
    'model': model.state_dict(),
    'optimizer': {'lr': 0.001},
    'epoch': 5,
    'step': 100
}

# Write to temporary file
with tempfile.NamedTemporaryFile(mode='wb', delete=False, suffix='.pt') as f:
    checkpoint_path = f.name
    torch.save(state, checkpoint_path)

# Create fresh model and optimizer for restoration
model_for_restore = SimpleModel()
optimizer_for_restore = SimpleOptimizer()

# Construct the modules dict
modules = {
    'model': model_for_restore,
    'optimizer': optimizer_for_restore
}

ARGS = {
    'path': checkpoint_path,
    'modules': modules,
    'num_checkpoints': 1,
    'map_location': None,
    'strict': True
}
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 tuple, ndim 1, dtype family int — 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_final_state Not yet run
unverified (harvested, no recorded run)
dojoteef/synst/models/utils.py
code served (permissive licence) · get_code("9fd00da14a2fbd08")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 4 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
checkpoint 2 other papers
2 later
2104.03474 (2021-04)
Revisiting Simple Neural Probabilistic Language Models
2204.10878 (2022-04)
CHAPTERBREAK: A Challenge Dataset for Long-Range Language Models
get_final_state 1 other paper
1 later
2104.03474 (2021-04)
Revisiting Simple Neural Probabilistic Language Models
restore 1 other paper
1 later
2104.03474 (2021-04)
Revisiting Simple Neural Probabilistic Language Models

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

Standard decoders for neural machine translation autoregressively generate a single target token per time step, which slows inference especially for long outputs. While architectural advances such as the Transformer fully parallelize the decoder computations at training time, inference still proceeds sequentially. Recent developments in non- and semi- autoregressive decoding produce multiple tokens per time step independently of the others, which improves inference speed but deteriorates translation quality. In this work, we propose the syntactically supervised Transformer (SynST), which first autoregressively predicts a chunked parse tree before generating all of the target tokens in one shot conditioned on the predicted parse. A series of controlled experiments demonstrates that SynST decodes sentences ~ 5x faster than the baseline autoregressive Transformer while achieving higher BLEU scores than most competing methods on En-De and En-Fr datasets.

For agents

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

get_harvested_code_for_paper("1906.02780")
get_code_for_paper("1906.02780")
have("1906.02780")

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

curl https://syntology.ai/api/ran/1906.02780.json

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

[![Syntology run record](https://syntology.ai/api/ran/1906.02780.svg)](https://syntology.ai/paper/1906.02780)

Connect an agent — have() is free.