SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1909.03004 · 2019

Show Your Work: Improved Reporting of Experimental Results

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 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 7 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/1909.03004.json · badge: /api/ran/1909.03004.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
allenai/allentune canonical (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
DerwenAI/pytextrank community (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
FunctionStatusWhere it lives
default_scrubber Ran
ran on a synthesized input
DerwenAI/pytextrank/deprecated/oldsrc.py
code served (permissive licence) · get_code("06162a75989f9f16")
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

def default_scrubber(text):
    """
    remove spurious punctuation (for English)
    """
    return text.lower().replace("'", '')

ARGS = {
    "text": "Hello, World! It's a beautiful day."
}
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
get_pyvis_html Ran
ran on a synthesized input
DerwenAI/pytextrank/bin/vis_doc.py
code served (permissive licence) · get_code("d3d93d483c6b1ccb")
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

PAT_SOURCE = re.compile('\\s+src\\=\\"(\\S+)\\"')

def get_pyvis_html(iframe):
    """
    located the HTML file generated by PyVis, if any
    """
    source_html = None
    m_source = PAT_SOURCE.search(iframe)
    if m_source:
        source_html = m_source.group(1)
        if 'tmp.fig' not in source_html:
            source_html = None
    return source_html

# Test case 1: iframe with tmp.fig in src attribute
iframe_with_tmp_fig = '  src="path/to/tmp.fig_12345.html"'

# Test case 2: iframe with src attribute but no tmp.fig
iframe_without_tmp_fig = '  src="path/to/other_file.html"'

# Test case 3: iframe without src attribute
iframe_no_src = '  data="some_data"'

ARGS = {
    "iframe": iframe_with_tmp_fig
}
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
groupby_apply Ran
ran on a synthesized input
DerwenAI/pytextrank/pytextrank/util.py
code served (permissive licence) · get_code("e6ebd165e111d23f")
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 2, 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 numpy as np
from typing import List, Tuple, Any, Callable, Iterable

# Set seed for reproducibility
torch.manual_seed(0)

# Create sample data: list of dictionaries with 'category' and 'value' keys
data = [
    {'category': 'A', 'value': 10},
    {'category': 'B', 'value': 20},
    {'category': 'A', 'value': 15},
    {'category': 'C', 'value': 30},
    {'category': 'B', 'value': 25},
    {'category': 'A', 'value': 12},
]

# Key function: extract the 'category' field
def keyfunc(item: dict) -> str:
    return item['category']

# Apply function: sum all values in a group
def applyfunc(group: Iterable[dict]) -> int:
    return sum(item['value'] for item in group)

ARGS = {
    'data': data,
    'keyfunc': keyfunc,
    'applyfunc': applyfunc,
}
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 list, ndim 2, 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
maniacal_scrubber Ran
ran on a synthesized input
DerwenAI/pytextrank/deprecated/oldsrc.py
code served (permissive licence) · get_code("3838530f25ae6667")
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 unicodedata

# The function takes a single parameter: text (a string)
# It performs various text cleaning operations on the input

ARGS = {
    "text": """
    This is some "garbled" text with fancy quotes ''and` dashes–here.
    It has multiple lines
    and weird unicode characters…
    """
}
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
replace_sys_header Ran
ran on a synthesized input
DerwenAI/pytextrank/bin/vis_doc.py
code served (permissive licence) · get_code("b0ffcfed6e97e271")
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

PAT_HEADER = re.compile('^(```python\\n\\# for use.*production:\\n.*\\n```\\n)', re.MULTILINE)

def replace_sys_header(text, stem):
    """
Replace the `sys.path` header for example notebooks
    """
    output = []
    for chunk in PAT_HEADER.split(text):
        m_iframe = PAT_HEADER.match(chunk)
        if m_iframe:
            header = '\n!!! note\n    To run this notebook in JupyterLab, load [`examples/{stem}.ipynb`](https://github.com/DerwenAI/pytextrank/blob/main/examples/{stem}.ipynb)\n\n'.format(stem=stem)
            output.append(header)
        else:
            output.append(chunk)
    return '\n'.join(output)


# Setup code to construct valid arguments
text_with_header = """```python
# for use in production:
import sys


This is some content after the header.

More text here."""

stem_value = "example_notebook"

ARGS = {
    "text": text_with_header,
    "stem": stem_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 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
samplemax Ran
ran on a synthesized input
allenai/allentune/allentune/commands/plot.py
code served (permissive licence) · get_code("091c323ea322b6ab")
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 numpy as np
import scipy.special

ARGS = {
    'validation_performance': [0.72, 0.65, 0.81, 0.58, 0.91],
    'with_replacement': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — 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
td_format Ran
ran on a synthesized input
allenai/allentune/allentune/commands/plot.py
code served (permissive licence) · get_code("44f8471c142ddd5c")
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 timedelta

# The function expects a timedelta object with a total_seconds() method
td_object = timedelta(days=2, hours=3, minutes=45, seconds=30)

ARGS = {
    'td_object': td_object
}
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
filter_quotes Not yet run
unverified (harvested, no recorded run)
DerwenAI/pytextrank/deprecated/oldsrc.py
code served (permissive licence) · get_code("1fe08169a36b3ea0")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 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.

FunctionAlso present inEarliest-datedOthers named
samplemax 1 other paper
1 later
2005.00115 (2020-05)
Learning to Faithfully Rationalize by Construction
td_format 1 other paper
1 later
2005.00115 (2020-05)
Learning to Faithfully Rationalize by Construction

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

Research in natural language processing proceeds, in part, by demonstrating that new models achieve superior performance (e.g., accuracy) on held-out test data, compared to previous results. In this paper, we demonstrate that test-set performance scores alone are insufficient for drawing accurate conclusions about which model performs best. We argue for reporting additional details, especially performance on validation data obtained during model development. We present a novel technique for doing so: expected validation performance of the best-found model as a function of computation budget (i.e., the number of hyperparameter search trials or the overall training time). Using our approach, we find multiple recent model comparisons where authors would have reached a different conclusion if they had used more (or less) computation. Our approach also allows us to estimate the amount of computation required to obtain a given accuracy; applying it to several recently published results yields massive variation across papers, from hours to weeks. We conclude with a set of best practices for reporting experimental results which allow for robust future comparisons, and provide code to allow researchers to use our technique.

For agents

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

get_harvested_code_for_paper("1909.03004")
get_code_for_paper("1909.03004")
have("1909.03004")

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

curl https://syntology.ai/api/ran/1909.03004.json

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

[![Syntology run record](https://syntology.ai/api/ran/1909.03004.svg)](https://syntology.ai/paper/1909.03004)

Connect an agent — have() is free.