SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2306.07261 · 2023

Unprocessing Seven Years of Algorithmic Fairness

arXiv · PDF · Open in the Atlas

Code that ran

12 of 14 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 12 have no contract verdict; 0 failures of our instrument; 2 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/2306.07261.json · badge: /api/ran/2306.07261.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
socialfoundations/error-parity canonical (classifier label)
classifier basis: census
12 ran · 2 no recorded run
FunctionStatusWhere it lives
calc_cost_of_point Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/roc_utils.py
code served (permissive licence) · get_code("373e6932d3ceb321")
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 float64, ndim 0, 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 numpy as np

ARGS = {
    "fpr": 0.15,
    "fnr": 0.08,
    "prevalence": 0.3,
    "false_pos_cost": 2.0,
    "false_neg_cost": 5.0,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
compute_binary_predictions Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/binarize.py
code served (permissive licence) · get_code("bea30f0c4191c10a")
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 ndarray, ndim 1, dtype family int, shape (10,) 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 numpy as np

# Create small numpy arrays for binary classification
np.random.seed(42)

# Create true binary labels (10 samples, mix of 0s and 1s)
y_true = np.array([1, 0, 1, 1, 0, 0, 1, 0, 1, 0])

# Create predicted scores between 0 and 1 (10 samples)
y_pred_scores = np.array([0.9, 0.2, 0.8, 0.75, 0.3, 0.15, 0.85, 0.25, 0.95, 0.1])

# Define ARGS dict with exactly one of the four options specified
ARGS = {
    'y_true': y_true,
    'y_pred_scores': y_pred_scores,
    'threshold': 0.5,
    'tpr': None,
    'fpr': None,
    'ppr': None,
    'random_seed': 42
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family int, shape (10,) — shape (10,) 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
compute_global_roc_from_groupwise Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/roc_utils.py
code served (permissive licence) · get_code("9f867f80ffa8598c")
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 ndarray, ndim 1, dtype family float, shape (2,) 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 numpy as np

# Create sample data for the function
n_groups = 3
n_roc_dims = 2

# groupwise_roc_points: shape (n_groups, n_roc_dims)
# ROC points with FPR and TPR values
groupwise_roc_points = np.array([
    [0.1, 0.8],
    [0.2, 0.85],
    [0.15, 0.75]
])

# groupwise_label_pos_weight: shape (n_groups,)
# Relative size of each group in positive samples
groupwise_label_pos_weight = np.array([0.4, 0.35, 0.25])

# groupwise_label_neg_weight: shape (n_groups,)
# Relative size of each group in negative samples
groupwise_label_neg_weight = np.array([0.3, 0.5, 0.2])

ARGS = {
    "groupwise_roc_points": groupwise_roc_points,
    "groupwise_label_pos_weight": groupwise_label_pos_weight,
    "groupwise_label_neg_weight": groupwise_label_neg_weight
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) 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
compute_halfspace_inequality Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/cvxpy_utils.py
code served (permissive licence) · get_code("cba32d01413e5673")
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 1, 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 numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Create two distinct 2D points
p1 = np.array([1.0, 2.0])
p2 = np.array([4.0, 5.0])

ARGS = {
    "p1": p1,
    "p2": p2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, 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
compute_line Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/cvxpy_utils.py
code served (permissive licence) · get_code("d00d9d69efa4b9ee")
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 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 numpy as np

# Create two distinct 2D points
p1 = np.array([1.0, 2.0])
p2 = np.array([3.0, 6.0])

ARGS = {
    "p1": p1,
    "p2": p2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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 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
compute_roc_point_from_predictions Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/roc_utils.py
code served (permissive licence) · get_code("95616abc4aac0887")
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 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 numpy as np

# Set random seed for reproducibility
np.random.seed(0)

# Create small sample data for binary classification
# y_true: true binary labels (0 or 1)
y_true = np.array([0, 1, 1, 0, 1, 0, 1, 1])

# y_pred_binary: predicted binary labels (0 or 1)
y_pred_binary = np.array([0, 1, 0, 0, 1, 1, 1, 1])

ARGS = {
    "y_true": y_true,
    "y_pred_binary": y_pred_binary,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — 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 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
evaluate_fairness Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/evaluation.py
code served (permissive licence) · get_code("17715daaa9040165")
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

# Create sample data for fairness evaluation
np.random.seed(0)

# y_true: binary class labels (0 or 1)
y_true = np.array([0, 1, 1, 0, 1, 1, 0, 0, 1, 0])

# y_pred: discretized predictions (0 or 1)
y_pred = np.array([0, 1, 1, 0, 0, 1, 0, 1, 1, 0])

# sensitive_attribute: group membership (needs at least 2 unique values)
# Using values 0 and 1 to represent two different groups
sensitive_attribute = np.array([0, 0, 0, 0, 0, 1, 1, 1, 1, 1])

# return_groupwise_metrics: optional boolean flag
return_groupwise_metrics = False

ARGS = {
    'y_true': y_true,
    'y_pred': y_pred,
    'sensitive_attribute': sensitive_attribute,
    'return_groupwise_metrics': return_groupwise_metrics
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — 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
evaluate_performance Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/evaluation.py
code served (permissive licence) · get_code("39cf6fea798aec32")
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

# Set random seed for reproducibility
np.random.seed(0)

# Create simple binary classification data
# y_true: true binary labels (0 or 1)
y_true = np.array([0, 1, 1, 0, 1, 0, 1, 1, 0, 0])

# y_pred: predicted binary labels (0 or 1)
y_pred = np.array([0, 1, 1, 0, 0, 0, 1, 1, 0, 1])

ARGS = {
    'y_true': y_true,
    'y_pred': y_pred,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — 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
get_cost_envelope Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/_commons.py
code served (permissive licence) · get_code("2c63f257596f927f")
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 ndarray, ndim 2, dtype family float, shape (4, 2) 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 numpy as np

# Create a sample costs array with multiple points and objectives
# Shape: (n_points, n_objectives)
# We'll create points where some are Pareto efficient and some are on the convex hull
costs = np.array([
    [1.0, 5.0],
    [2.0, 3.0],
    [3.0, 2.0],
    [5.0, 1.0],
    [2.5, 2.5],  # This point is dominated by others
    [1.5, 4.0],  # This point is dominated
], dtype=np.float64)

ARGS = {
    "costs": costs
}
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 ndarray, ndim 2, dtype family float, shape (4, 2) — shape (4, 2) 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
is_pareto_efficient Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/_commons.py
code served (permissive licence) · get_code("63d28ab3fce7bb7d")
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 ndarray, ndim 1, dtype family float, shape (5,) 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 numpy as np

# Create a simple test case for Pareto efficiency
# We need an (n_points, n_costs) array where smaller values are better
# Let's create 5 points with 3 cost dimensions

costs = np.array([
    [1.0, 2.0, 3.0],      # Point 0: potentially efficient
    [2.0, 1.0, 4.0],      # Point 1: potentially efficient
    [3.0, 3.0, 3.0],      # Point 2: dominated by point 0
    [1.5, 1.5, 1.5],      # Point 3: potentially efficient
    [5.0, 5.0, 5.0],      # Point 4: dominated
], dtype=np.float64)

ARGS = {
    "costs": costs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (5,) — shape (5,) 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
is_part_of_convexhull Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/_commons.py
code served (permissive licence) · get_code("9f640a34c00a2a55")
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 ndarray, ndim 1, dtype family float, shape (4,) 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 numpy as np
import logging

# Set up logging to avoid warnings
logging.basicConfig(level=logging.INFO)

# Create a small set of 2D points where we can verify convex hull behavior
# Using 4 points in 2D space - some will be on the hull, some won't
costs = np.array([
    [0.0, 0.0],      # vertex 0 - corner
    [1.0, 0.0],      # vertex 1 - corner
    [0.0, 1.0],      # vertex 2 - corner
    [0.5, 0.5],      # vertex 3 - interior point
], dtype=np.float64)

ARGS = {
    "costs": costs
}
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 ndarray, ndim 1, dtype family float, shape (4,) — shape (4,) 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
make_cvxpy_halfspace_inequality Ran
ran on a synthesized input
socialfoundations/error-parity/error_parity/cvxpy_utils.py
code served (permissive licence) · get_code("01c4318cc6a2bdb8")
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 Inequality. 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
from cvxpy.expressions.variable import Variable

# Create two 2D points for the halfspace inequality
p1 = np.array([0.0, 0.0])
p2 = np.array([1.0, 1.0])

# Create a cvxpy Variable representing a 2D point
cvxpy_point = Variable(2)

ARGS = {
    "p1": p1,
    "p2": p2,
    "cvxpy_point": cvxpy_point,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cvxpy, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Inequality — a Inequality output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Inequality 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
eval_accuracy_and_equalized_odds Not yet run
unverified (harvested, no recorded run)
socialfoundations/error-parity/error_parity/evaluation.py
code served (permissive licence) · get_code("f23c0de3ee013200")
How it ran

no invocation recorded.

get_envelope_of_postprocessing_frontier Not yet run
unverified (harvested, no recorded run)
socialfoundations/error-parity/error_parity/pareto_curve.py
code served (permissive licence) · get_code("6b67fcc092be21ed")
How it ran

no invocation recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 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 14 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

Seven years ago, researchers proposed a postprocessing method to equalize the error rates of a model across different demographic groups. The work launched hundreds of papers purporting to improve over the postprocessing baseline. We empirically evaluate these claims through thousands of model evaluations on several tabular datasets. We find that the fairness-accuracy Pareto frontier achieved by postprocessing contains all other methods we were feasibly able to evaluate. In doing so, we address two common methodological errors that have confounded previous observations. One relates to the comparison of methods with different unconstrained base models. The other concerns methods achieving different levels of constraint relaxation. At the heart of our study is a simple idea we call unprocessing that roughly corresponds to the inverse of postprocessing. Unprocessing allows for a direct comparison of methods using different underlying models and levels of relaxation.

For agents

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

get_harvested_code_for_paper("2306.07261")
get_code_for_paper("2306.07261")
have("2306.07261")

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

curl https://syntology.ai/api/ran/2306.07261.json

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

[![Syntology run record](https://syntology.ai/api/ran/2306.07261.svg)](https://syntology.ai/paper/2306.07261)

Connect an agent — have() is free.