SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · aaai_20056 · AAAI · 2022

Local Similarity Pattern and Cost Self-Reassembling for Deep Stereo Matching Networks

Huimin Yu, Biyang Liu, Yangqi Long

Open in the Atlas

Code that ran

12 of 15 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, 1 violated it and 7 have no contract verdict; 4 failures of our instrument; 3 no recorded run), measured 2026-11-22; 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/aaai_20056.json · badge: /api/ran/aaai_20056.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
SpadeLiu/Lac-GwcNet canonical (classifier label)
classifier basis: census
12 ran · 3 no recorded run
FunctionStatusWhere it lives
conv2d Ran
ran; our contract draft was wrong, not the code
SpadeLiu/Lac-GwcNet/networks/refinement.py
code served (permissive licence) · get_code("2c4f90d71c073dfe")
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 Sequential. 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

torch.manual_seed(0)

ARGS = {
    "in_channels": 3,
    "out_channels": 16,
    "kernel_size": 3,
    "stride": 1,
    "dilation": 1,
    "groups": 1
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
SpadeLiu/Lac-GwcNet/networks/refinement.py
code served (permissive licence) · get_code("9fba66e062846970")
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 Conv2d. 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

torch.manual_seed(0)

ARGS = {
    "in_planes": 3,
    "out_planes": 64,
    "stride": 1,
    "groups": 1,
    "dilation": 1,
    "with_bn_relu": False,
    "leaky_relu": False,
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
convbn Ran
ran; our contract draft was wrong, not the code
SpadeLiu/Lac-GwcNet/networks/submodule.py
pointer only (licence: MIT); the stored copy get_code serves, from 3dcvdeveloper/bgnet, is pointer only (licence: NONE) · get_code("15ab3c76823aef7e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from 3dcvdeveloper/bgnet, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

convbn_3d Ran
ran; our contract draft was wrong, not the code
SpadeLiu/Lac-GwcNet/networks/submodule.py
pointer only (licence: MIT); the stored copy get_code serves, from mickeykang16/temporaleventstereo, is pointer only (licence: NONE) · get_code("46674adf87c5459d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from mickeykang16/temporaleventstereo, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

disp2distribute Ran
ran on a synthesized input
SpadeLiu/Lac-GwcNet/loss_functions.py
code served (permissive licence) · get_code("d0251f3589a6d87c")
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 4, dtype family float, shape (2, 8, 4, 3) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

# disp_gt: disparity ground truth tensor
# Should be a tensor with shape (batch, height, width)
# The function will unsqueeze it to (batch, 1, height, width)
disp_gt = torch.randn(2, 4, 3)  # batch=2, height=4, width=3

# max_disp: maximum disparity value (int)
max_disp = 8

# b: smoothing parameter (float, default=2)
b = 2

ARGS = {
    "disp_gt": disp_gt,
    "max_disp": max_disp,
    "b": b
}
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 4, dtype family float, shape (2, 8, 4, 3) — shape (2, 8, 4, 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
img_loader Ran
ran on a synthesized input
SpadeLiu/Lac-GwcNet/dataloader/KITTI2012loader.py
code served (permissive licence) · get_code("2a811e120e26afac")
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 Image. 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
from PIL import Image
import io
import numpy as np

# Create a minimal RGB image in memory
img_array = np.random.randint(0, 256, (10, 10, 3), dtype=np.uint8)
img = Image.fromarray(img_array, 'RGB')

# Save to bytes buffer
buffer = io.BytesIO()
img.save(buffer, format='PNG')
buffer.seek(0)

# Create a temporary path-like object that PIL can work with
# We'll use a simple approach: create the image bytes and use a data URI approach
# But since PIL.Image.open expects a file path or file-like object,
# we'll pass a BytesIO object wrapped as a path-like interface

class FakeImagePath:
    def __init__(self):
        self.buffer = io.BytesIO()
        img_array = np.random.randint(0, 256, (10, 10, 3), dtype=np.uint8)
        img = Image.fromarray(img_array, 'RGB')
        img.save(self.buffer, format='PNG')
        self.buffer.seek(0)
    
    def __fspath__(self):
        return str(self.buffer)
    
    def __str__(self):
        return self.buffer.getvalue().__repr__()

# Actually, let's use a simpler approach with a file-like object directly
# PIL.Image.open can accept file-like objects
buffer = io.BytesIO()
img_array = np.random.randint(0, 256, (10, 10, 3), dtype=np.uint8)
img = Image.fromarray(img_array, 'RGB')
img.save(buffer, format='PNG')
buffer.seek(0)

ARGS = {
    'path': buffer
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
is_image_file Ran
ran, violated the contract we drafted
SpadeLiu/Lac-GwcNet/dataloader/KITTI2012loader.py
pointer only (licence: MIT); the stored copy get_code serves, from sgoldyaev/DeepFashion.ADGAN, is pointer only (licence: NONE) · get_code("ab4109634b75ef8b")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from sgoldyaev/DeepFashion.ADGAN, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

kt2012_loader Ran
ran on a synthesized input
SpadeLiu/Lac-GwcNet/dataloader/KITTI2012loader.py
code served (permissive licence) · get_code("be1a40bb874f6f8c")
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. 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 os
import tempfile
import shutil

# Create a temporary directory structure that mimics the kt2012 dataset
temp_dir = tempfile.mkdtemp()

# Create subdirectories
colored_0_dir = os.path.join(temp_dir, 'colored_0')
colored_1_dir = os.path.join(temp_dir, 'colored_1')
disp_occ_dir = os.path.join(temp_dir, 'disp_occ')

os.makedirs(colored_0_dir)
os.makedirs(colored_1_dir)
os.makedirs(disp_occ_dir)

# Create dummy files with names containing '_10' (since the function filters for these)
# Create 170 files total (160 for train, 10 for val)
for i in range(170):
    # Create dummy files
    open(os.path.join(colored_0_dir, f'frame_{i:06d}_10.png'), 'w').close()
    open(os.path.join(colored_1_dir, f'frame_{i:06d}_10.png'), 'w').close()
    open(os.path.join(disp_occ_dir, f'frame_{i:06d}_10.png'), 'w').close()

ARGS = {
    'filepath': temp_dir
}
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 tuple — a tuple 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
kt2012_loader Ran
ran on a synthesized input
SpadeLiu/Lac-GwcNet/dataloader/KITTIloader.py
code served (permissive licence) · get_code("329bdb11a2a2bcd7")
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. 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 os
import tempfile
import shutil

# Create a temporary directory structure that mimics the kt2012 dataset
temp_dir = tempfile.mkdtemp()

# Create the required subdirectories
left_dir = os.path.join(temp_dir, 'colored_0')
right_dir = os.path.join(temp_dir, 'colored_1')
displ_dir = os.path.join(temp_dir, 'disp_occ')

os.makedirs(left_dir)
os.makedirs(right_dir)
os.makedirs(displ_dir)

# Create dummy files with names containing '_10' to satisfy the filter condition
# We need at least 160 files for train_name and some for val_name
for i in range(170):
    # Create empty files with names matching the pattern
    left_file = os.path.join(left_dir, f'frame_{i:03d}_10.png')
    right_file = os.path.join(right_dir, f'frame_{i:03d}_10.png')
    displ_file = os.path.join(displ_dir, f'frame_{i:03d}_10.png')
    
    open(left_file, 'a').close()
    open(right_file, 'a').close()
    open(displ_file, 'a').close()

ARGS = {
    'filepath': temp_dir
}
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 tuple — a tuple 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
kt_loader Ran
ran on a synthesized input
SpadeLiu/Lac-GwcNet/dataloader/KITTIloader.py
code served (permissive licence) · get_code("178b73271f5e6fee")
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. 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 os
import tempfile
import shutil

# Create a temporary directory structure that mimics the expected file layout
temp_dir = tempfile.mkdtemp()

# Create subdirectories
image_2_dir = os.path.join(temp_dir, 'image_2')
image_3_dir = os.path.join(temp_dir, 'image_3')
disp_occ_0_dir = os.path.join(temp_dir, 'disp_occ_0')

os.makedirs(image_2_dir, exist_ok=True)
os.makedirs(image_3_dir, exist_ok=True)
os.makedirs(disp_occ_0_dir, exist_ok=True)

# Create dummy files with names containing '_10'
# The function filters for names with '_10' and expects at least 160+ files
for i in range(170):
    filename = f'image_{i:06d}_10.png'
    
    # Create empty files in each directory
    open(os.path.join(image_2_dir, filename), 'w').close()
    open(os.path.join(image_3_dir, filename), 'w').close()
    open(os.path.join(disp_occ_0_dir, filename), 'w').close()

ARGS = {
    'filepath': temp_dir
}
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 tuple — a tuple 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
random_noise Ran
ran on a synthesized input
SpadeLiu/Lac-GwcNet/loss_functions.py
code served (permissive licence) · get_code("b4e1d9a047616b74")
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 Image. 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 PIL import Image
import cv2

# Create a small valid RGB image (4x3x3)
np.random.seed(0)
rgb_array = np.random.randint(100, 200, size=(4, 3, 3), dtype=np.uint8)
img = Image.fromarray(rgb_array, mode='RGB')

ARGS = {
    'img': img,
    'type': 'illumination'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
readPFM Ran
ran on a synthesized input
SpadeLiu/Lac-GwcNet/dataloader/readpfm.py
code served (permissive licence) · get_code("973b549613a75154")
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. 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
import numpy as np
import tempfile
import os

# Create a temporary PFM file for testing
temp_dir = tempfile.mkdtemp()
pfm_file = os.path.join(temp_dir, 'test.pfm')

# Create a valid PFM file (color image, 2x3 pixels)
height, width = 3, 2
with open(pfm_file, 'wb') as f:
    # Header: PF for color
    f.write(b'PF\n')
    # Dimensions: width height
    f.write(f'{width} {height}\n'.encode('utf-8'))
    # Scale (negative indicates little-endian)
    f.write(b'-1.0\n')
    # Image data: height x width x 3 channels, flipped for PFM format
    data = np.random.rand(height, width, 3).astype('<f')
    data = np.flipud(data)
    data.tofile(f)

ARGS = {
    'file': pfm_file
}
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 — a tuple 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
CEloss Not yet run
unverified (harvested, no recorded run)
SpadeLiu/Lac-GwcNet/loss_functions.py
code served (permissive licence) · get_code("53cecb06d0e57df4")
How it ran

no invocation recorded.

reconstruction Not yet run
unverified (harvested, no recorded run)
SpadeLiu/Lac-GwcNet/networks/refinement.py
code served (permissive licence) · get_code("9a40c87795390bdf")
How it ran

no invocation recorded.

sf_loader Not yet run
unverified (harvested, no recorded run)
SpadeLiu/Lac-GwcNet/dataloader/sceneflow_loader.py
code served (permissive licence) · get_code("fbdd87da9c7160a4")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 distinct functions with an identity, 5 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 5 later) and 10 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
is_image_file 62 other papers
22 dated earlier, 30 later, 10 same year, order not determinable
1609.04802 (2016-09)
Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
1703.07511 (2017-03)
Deep Photo Style Transfer
1710.10916 (2017-10)
StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networks
and 59 more not named here
convbn 13 other papers
7 dated earlier, 5 later, 1 same year, order not determinable
1803.08669 (2018-03)
Pyramid Stereo Matching Network
1806.01260 (2018-06)
Digging Into Self-Supervised Monocular Depth Estimation
1905.00538 (2019-05)
DPSNet: End-to-end Deep Plane Sweep Stereo
and 10 more not named here
convbn_3d 9 other papers
7 dated earlier, 2 later
1803.08669 (2018-03)
Pyramid Stereo Matching Network
1806.01260 (2018-06)
Digging Into Self-Supervised Monocular Depth Estimation
1905.00538 (2019-05)
DPSNet: End-to-end Deep Plane Sweep Stereo
and 6 more not named here
conv2d 2 other papers
1 dated earlier, 1 later
2004.09548 (2020-04)
AANet: Adaptive Aggregation Network for Efficient Stereo Matching
2404.06842 (2024-04)
MoCha-Stereo: Motif Channel Attention Network for Stereo Matching
conv3x3 2 other papers
1 dated earlier, 1 later
2004.09548 (2020-04)
AANet: Adaptive Aggregation Network for Efficient Stereo Matching
2404.06842 (2024-04)
MoCha-Stereo: Motif Channel Attention Network for Stereo Matching

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

Although convolutional neural network based stereo matching architectures have made impressive achievements, there are still some limitations: 1) Convolutional Feature (CF) tends to capture appearance information, which is inadequate for accurate matching. 2) Due to the static filters, current convolution based disparity refinement modules often produce over-smooth results. In this paper, we present two schemes to address these issues, where some traditional wisdoms are integrated. Firstly, we introduce a pairwise feature for deep stereo matching networks, named LSP (Local Similarity Pattern). Through explicitly revealing the neighbor relationships, LSP contains rich structural information, which can be leveraged to aid CF for more discriminative feature description. Secondly, we design a dynamic self-reassembling refinement strategy and apply it to the cost distribution and the disparity map respectively. The former could be equipped with the unimodal distribution constraint to alleviate the over-smoothing problem, and the latter is more practical. The effectiveness of the proposed methods is demonstrated via incorporating them into two well-known basic architectures, GwcNet and GANet-deep. Experimental results on the SceneFlow and KITTI benchmarks show that our modules significantly improve the performance of the model. Code is available at https://github.com/SpadeLiu/Lac-GwcNet.

For agents

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

get_harvested_code_for_paper("aaai_20056")
get_code_for_paper("aaai_20056")
have("aaai_20056")

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

curl https://syntology.ai/api/ran/aaai_20056.json

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

[![Syntology run record](https://syntology.ai/api/ran/aaai_20056.svg)](https://syntology.ai/paper/aaai_20056)

Connect an agent — have() is free.