SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.12489 · ECCV · 2022

Sound Localization by Self-Supervised Time Delay Estimation

Andrew Owens, David Fouhey, Ziyang Chen

arXiv · PDF · Open in the Atlas

Code that ran

12 of 16 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; 4 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/2204.12489.json · badge: /api/ran/2204.12489.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
IFICL/stereocrw canonical (classifier label)
classifier basis not recorded on the link
12 ran · 4 no recorded run
FunctionStatusWhere it lives
angle Ran
ran on a synthesized input
IFICL/stereocrw/utils/utils.py
code served (permissive licence) · get_code("90ce9429d8e138ca")
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 Tensor, ndim 2, dtype family float; no output shape is stated. It ran in syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1) at 2026-09-26T22:17:51Z. 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

torch.manual_seed(0)

# Create small tensors for real and imaginary components
real = torch.randn(2, 4)
imag = torch.randn(2, 4)

ARGS = {
    "real": real,
    "imag": imag,
}
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 Tensor, ndim 2, dtype family float — no output shape is stated: no later run recorded an output shape for this sample (observed_output and metamorphic_properties carry none; shape_contract and output_contract record ndim only)
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 syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1); 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
apply_cmap Ran
ran on a synthesized input
IFICL/stereocrw/utils/vis_utils.py
code served (permissive licence) · get_code("985be22f95033ef0")
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 3, dtype family float, shape (4, 3, 3) 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 pylab

# Create a small 2D image array (height=4, width=3)
im = np.random.rand(4, 3) * 100

# Define the colormap
cmap = pylab.cm.jet

# Define optional parameters
lo = None
hi = None

# Construct the arguments dictionary
ARGS = {
    'im': im,
    'cmap': cmap,
    'lo': lo,
    'hi': hi
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pylab — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (4, 3, 3) — shape (4, 3, 3) 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
binary_acc Ran
ran on a synthesized input
IFICL/stereocrw/utils/torch_utils.py
code served (permissive licence) · get_code("3da38a399d02010a")
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

# Construct valid arguments for binary_acc function
# The function takes:
# - pred: predictions (compared with > thred)
# - target: ground truth labels (compared with pred)
# - thred: threshold value for binary classification

# Create small numpy arrays for pred and target
pred = np.array([0.1, 0.9, 0.3, 0.8, 0.2, 0.7])
target = np.array([0, 1, 0, 1, 0, 1])

# Create a threshold value
thred = 0.5

ARGS = {
    "pred": pred,
    "target": target,
    "thred": thred
}
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
calc_acc Ran
ran on a synthesized input
IFICL/stereocrw/utils/torch_utils.py
code served (permissive licence) · get_code("e1dc20ea343147aa")
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 Tensor, ndim 0, dtype family float; no output shape is stated. It ran in syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1) at 2026-09-26T22:17:51Z. 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

torch.manual_seed(0)

# Create small tensors for testing
batch_size = 2
num_classes = 4

# prob: probability tensor of shape (batch_size, num_classes)
# The function does argsort on dim=-1, so we need a 2D tensor
prob = torch.randn(batch_size, num_classes)

# labels: label tensor of shape (batch_size,)
# The function reshapes it to (-1, 1) and compares with pred
labels = torch.tensor([1, 3])

# k: number of top predictions to consider
k = 2

ARGS = {
    "prob": prob,
    "labels": labels,
    "k": k
}
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 Tensor, ndim 0, dtype family float — no output shape is stated: no later run recorded an output shape for this sample (observed_output and metamorphic_properties carry none; shape_contract and output_contract record ndim only)
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 syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1); 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
clip_rescale Ran
ran on a synthesized input
IFICL/stereocrw/utils/vis_utils.py
code served (permissive licence) · get_code("85e99894dc6a3ab1")
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; no output shape is stated. It ran in syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1) at 2026-09-26T22:17:51Z. 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

# Construct valid arguments for clip_rescale
# The function takes x (required), lo (optional), hi (optional)
# x should be a numpy array
# lo and hi should be None or numeric values

x = np.array([1.0, 2.0, 3.0, 4.0, 5.0])
lo = None
hi = None

ARGS = {
    "x": x,
    "lo": lo,
    "hi": hi,
}
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 — no output shape is stated: no later run recorded an output shape for this sample (observed_output and metamorphic_properties carry none; shape_contract and output_contract record ndim only)
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 syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1); 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
cmap_im Ran
ran on a synthesized input
IFICL/stereocrw/utils/vis_utils.py
code served (permissive licence) · get_code("954ba75869c341e4")
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 3, dtype family int; no output shape is stated. It ran in syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1) at 2026-09-26T22:17:51Z. 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
import pylab

# Create a small test image (2D array with shape (4, 3))
im = np.random.rand(4, 3) * 100

# Define ARGS mapping all parameters of cmap_im
ARGS = {
    'cmap': pylab.cm.jet,
    'im': im,
    'lo': None,
    'hi': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pylab — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int — no output shape is stated: no later run recorded an output shape for this sample (observed_output and metamorphic_properties carry none; shape_contract and output_contract record ndim only)
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 syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1); 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_regression Ran
ran on a synthesized input
IFICL/stereocrw/vis_scripts/eval_itd_by_itd.py
code served (permissive licence) · get_code("3cecb8e4ec547c9c")
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
from sklearn.metrics import r2_score

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

# Create small synthetic regression data
# pred and target should be 1D arrays or compatible shapes
pred = np.array([1.2, 2.3, 3.1, 4.8, 5.2])
target = np.array([1.0, 2.5, 3.0, 4.9, 5.0])

# prefix can be empty string or a custom string
prefix = 'test'

ARGS = {
    'pred': pred,
    'target': target,
    'prefix': prefix
}
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
load_sound Ran
ran on a synthesized input
IFICL/stereocrw/utils/sound.py
code served (permissive licence) · get_code("cb41f906fb41033a")
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 Sound. It ran in syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1) at 2026-09-26T22:17:51Z. 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 tempfile
import scipy.io.wavfile
import os

# Create a temporary WAV file for testing
temp_dir = tempfile.gettempdir()
wav_fname = os.path.join(temp_dir, "test_sound.wav")

# Create a simple audio signal: 1 second at 44100 Hz
rate = 44100
duration = 1
num_samples = rate * duration
t = np.linspace(0, duration, num_samples)
# Simple sine wave at 440 Hz
frequency = 440
samples = np.sin(2 * np.pi * frequency * t)
# Convert to int16
samples_int16 = np.array(samples * 32767, dtype=np.int16)

# Write to WAV file
scipy.io.wavfile.write(wav_fname, rate, samples_int16)

ARGS = {
    "wav_fname": wav_fname
}
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 Sound — a Sound output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sound output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1); 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
normalize_rms Ran
ran on a synthesized input
IFICL/stereocrw/utils/utils.py
code served (permissive licence) · get_code("1257ffc2d2e10d60")
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 Tensor, ndim 2, dtype family float, shape (2, 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 torch

torch.manual_seed(0)

# Create a small tensor with shape (batch_size=2, features=4)
# The function expects samples to have a dim=1 for RMS calculation
samples = torch.randn(2, 4, dtype=torch.float32)

# desired_rms parameter - a small positive float
desired_rms = 0.1

# eps parameter - small epsilon value to avoid division by zero
eps = 0.0001

ARGS = {
    "samples": samples,
    "desired_rms": desired_rms,
    "eps": eps,
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 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
stft_frame_length Ran
ran on a synthesized input
IFICL/stereocrw/utils/sourcesep.py
code served (permissive licence) · get_code("d276952abad5059f")
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 torch
import numpy as np

class AudioConfig:
    def __init__(self, frame_length_ms, samp_sr):
        self.frame_length_ms = frame_length_ms
        self.samp_sr = samp_sr

pr = AudioConfig(frame_length_ms=25.0, samp_sr=16000)

ARGS = {
    'pr': pr
}
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 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
stft_frame_step Ran
ran on a synthesized input
IFICL/stereocrw/utils/sourcesep.py
code served (permissive licence) · get_code("3c5b0d2377733d6f")
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 torch

class MockAudioConfig:
    def __init__(self):
        self.frame_step_ms = 10
        self.samp_sr = 16000

pr = MockAudioConfig()

ARGS = {
    'pr': pr
}
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 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
stft_num_fft Ran
ran on a synthesized input
IFICL/stereocrw/utils/sourcesep.py
code served (permissive licence) · get_code("28487c00cd7b97ce")
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 numpy as np

class AudioConfig:
    def __init__(self, frame_length_ms, samp_sr):
        self.frame_length_ms = frame_length_ms
        self.samp_sr = samp_sr

pr = AudioConfig(frame_length_ms=25.0, samp_sr=16000)

ARGS = {"pr": pr}
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 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
inference Not yet run
unverified (harvested, no recorded run)
IFICL/stereocrw/vis_scripts/eval_itd_by_itd.py
code served (permissive licence) · get_code("8f4555ec07aef089")
How it ran

no invocation recorded.

init_args Not yet run
unverified (harvested, no recorded run)
IFICL/stereocrw/config/config.py
code served (permissive licence) · get_code("30a8feb0ae31dd2f")
How it ran

no invocation recorded.

load_model Not yet run
unverified (harvested, no recorded run)
IFICL/stereocrw/utils/torch_utils.py
code served (permissive licence) · get_code("dd477c377ef9e596")
How it ran

no invocation recorded.

normalize_rms_np Not yet run
unverified (harvested, no recorded run)
IFICL/stereocrw/utils/utils.py
code served (permissive licence) · get_code("9282d9cd4bf5682e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 0 later) and 12 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
load_sound 1 other paper
1 dated earlier
2111.05846 (2021-11)
Structure from Silence: Learning Scene Structure from Ambient Sound
stft_frame_length 1 other paper
1 dated earlier
2111.05846 (2021-11)
Structure from Silence: Learning Scene Structure from Ambient Sound
stft_frame_step 1 other paper
1 dated earlier
2111.05846 (2021-11)
Structure from Silence: Learning Scene Structure from Ambient Sound
stft_num_fft 1 other paper
1 dated earlier
2111.05846 (2021-11)
Structure from Silence: Learning Scene Structure from Ambient Sound

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

Input video Time delay (ms) L R Time (seconds) Fig. 1: Given a stereo audio recording, we estimate a sound's interaural time delay. Our model learns through self-supervision to find correspondences between the signals in each channel, from which the time delay can be estimated. We show time delay predictions for two scenes, along with their corresponding video frames (not used by the model). In both cases, the sound source changes its position in a scene, resulting in a corresponding change in time delay.

For agents

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

get_harvested_code_for_paper("2204.12489")
get_code_for_paper("2204.12489")
have("2204.12489")

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

curl https://syntology.ai/api/ran/2204.12489.json

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

[![Syntology run record](https://syntology.ai/api/ran/2204.12489.svg)](https://syntology.ai/paper/2204.12489)

Connect an agent — have() is free.