SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.03474 · NAACL · 2021

Revisiting Simple Neural Probabilistic Language Models

Mohit Iyyer, Simeng Sun

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 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 0 have no contract verdict; 1 failure of our instrument; 1 no recorded run), measured 2026-11-21; 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/2104.03474.json · badge: /api/ran/2104.03474.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
SimengSun/revisit-nplm canonical (classifier label)
classifier basis: census
2 ran · 1 no recorded run
FunctionStatusWhere it lives
checkpoint Ran
ran; our contract draft was wrong, not the code
SimengSun/revisit-nplm/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
restore Ran
ran, honoured the contract we drafted
SimengSun/revisit-nplm/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)
SimengSun/revisit-nplm/models/utils.py
code served (permissive licence) · get_code("9fd00da14a2fbd08")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 1 later) and 0 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
1 dated earlier, 1 later
1906.02780 (2019-06)
Syntactically Supervised Transformers for Faster Neural Machine Translation
2204.10878 (2022-04)
CHAPTERBREAK: A Challenge Dataset for Long-Range Language Models
get_final_state 1 other paper
1 dated earlier
1906.02780 (2019-06)
Syntactically Supervised Transformers for Faster Neural Machine Translation
restore 1 other paper
1 dated earlier
1906.02780 (2019-06)
Syntactically Supervised Transformers for Faster Neural Machine Translation

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recent progress in language modeling has been driven not only by advances in neural architectures, but also through hardware and optimization improvements. In this paper, we revisit the neural probabilistic language model (NPLM) of Bengio et al. (2003), which simply concatenates word embeddings within a fixed window and passes the result through a feed-forward network to predict the next word. When scaled up to modern hardware, this model (despite its many limitations) performs much better than expected on word-level language model benchmarks. Our analysis reveals that the NPLM achieves lower perplexity than a baseline Transformer with short input contexts but struggles to handle long-term dependencies. Inspired by this result, we modify the Transformer by replacing its first selfattention layer with the NPLM's local concatenation layer, which results in small but consistent perplexity decreases across three wordlevel language modeling datasets.

For agents

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

get_harvested_code_for_paper("2104.03474")
get_code_for_paper("2104.03474")
have("2104.03474")

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

curl https://syntology.ai/api/ran/2104.03474.json

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

[![Syntology run record](https://syntology.ai/api/ran/2104.03474.svg)](https://syntology.ai/paper/2104.03474)

Connect an agent — have() is free.