SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1911.03584 · 2019

On the Relationship between Self-Attention and Convolutional Layers

arXiv · PDF · Open in the Atlas

Code that ran

13 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; 1 honoured the contract we drafted, 0 violated it and 9 have no contract verdict; 3 failures of our instrument; 3 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/1911.03584.json · badge: /api/ran/1911.03584.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
epfml/attention-cnn canonical (classifier label)
classifier basis: census
13 ran · 3 no recorded run
FunctionStatusWhere it lives
ResNet10 Ran
ran on a synthesized input
epfml/attention-cnn/models/resnet.py
code served (permissive licence) · get_code("1119795bf433256e")
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 ResNet. 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 = {
    "num_classes": 10,
    "use_batchnorm": True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
ResNet18 Ran
ran on a synthesized input
epfml/attention-cnn/models/resnet.py
code served (permissive licence) · get_code("a4eff019b95fd2ce")
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 ResNet. 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 = {
    "num_classes": 10,
    "use_batchnorm": True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
ResNet34 Ran
ran on a synthesized input
epfml/attention-cnn/models/resnet.py
code served (permissive licence) · get_code("af3a6e8a2702592f")
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 ResNet. 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 = {
    "num_classes": 10,
    "use_batchnorm": True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
gaussian_kernel_2d Ran
ran on a synthesized input
epfml/attention-cnn/models/gaussian.py
code served (permissive licence) · get_code("796bf85d9c48fcf1")
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 (5, 7) 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 numbers

torch.manual_seed(0)

# Construct arguments for gaussian_kernel_2d function
mean = torch.tensor([0.5, 0.3], dtype=torch.float32)
std_inv = torch.tensor([1.0, 1.2], dtype=torch.float32)
size = (5, 7)

ARGS = {
    "mean": mean,
    "std_inv": std_inv,
    "size": size,
}
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 (5, 7) — shape (5, 7) 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
gelu Ran
ran, honoured the contract we drafted
epfml/attention-cnn/models/bert.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from tanmay1618/text_to_sql_bert, is pointer only (licence: NONE) · get_code("fdc64f4c72036ae4")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from tanmay1618/text_to_sql_bert, 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.

get_num_parameter Ran
ran on a synthesized input
epfml/attention-cnn/utils/logging.py
code served (permissive licence) · get_code("d5d2fd533e9dac1d")
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 torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
        self.frozen_param = nn.Parameter(torch.randn(5))
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()
model.frozen_param.requires_grad = False

ARGS = {
    "model": model,
    "trainable": 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 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
human_format Ran
ran; our contract draft was wrong, not the code
epfml/attention-cnn/utils/logging.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from siddk/vqa-outliers, is pointer only (licence: GPL-3.0) · get_code("2844a6a8960060bc")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from siddk/vqa-outliers, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

linear_warmup_cosine_lr_scheduler Ran
ran on a synthesized input
epfml/attention-cnn/utils/learning_rate.py
code served (permissive licence) · get_code("14f069410e56d3ba")
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 LambdaLR. 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
from torch.optim import SGD

torch.manual_seed(0)

# Create a simple model
model = nn.Linear(4, 3)

# Create an optimizer
optimizer = SGD(model.parameters(), lr=0.1)

# Define the arguments
ARGS = {
    "optimizer": optimizer,
    "warmup_time_ratio": 0.1,
    "T_max": 100
}
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 LambdaLR — a LambdaLR output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a LambdaLR 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
parse_cli_overides Ran
ran on a synthesized input
epfml/attention-cnn/utils/config.py
code served (permissive licence) · get_code("c3de407b29c79036")
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 argparse
import enum
from unittest.mock import patch
import sys

class TestEnum(enum.Enum):
    OPTION_A = "option_a"
    OPTION_B = "option_b"

# Create a minimal config dictionary with various types
config = {
    'learning_rate': 0.001,
    'batch_size': 32,
    'model_name': 'bert',
    'use_cuda': False,
    'epochs': 10,
    'dropout': 0.5,
    'enum_param': TestEnum.OPTION_A,
    'optional_param': None,
}

# Mock sys.argv to simulate CLI arguments
# We'll test with no overrides (empty CLI args) so the function uses defaults
mock_argv = ['script.py']

ARGS = {
    'config': config
}

# Patch sys.argv before the function would parse it
with patch.object(sys, 'argv', mock_argv):
    pass
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
termcolor — 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
plot_attention_layer Ran
ran on a synthesized input
epfml/attention-cnn/utils/plotting.py
code served (permissive licence) · get_code("de4b895397bf1813")
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 Axes. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import numpy as np
import matplotlib.pyplot as plt

torch.manual_seed(0)

class AttentionSelf(nn.Module):
    def __init__(self, num_heads=4):
        super().__init__()
        self.num_heads = num_heads
    
    def get_attention_probs(self, height, width):
        # Return attention probabilities with shape: (1, height, width, num_heads, height, width)
        # or (height, width, num_heads, height, width)
        return torch.randn(height, width, self.num_heads, height, width)

class Attention(nn.Module):
    def __init__(self):
        super().__init__()
        self.self = AttentionSelf(num_heads=4)

class TransformerLayer(nn.Module):
    def __init__(self):
        super().__init__()
        self.attention = Attention()

class Encoder(nn.Module):
    def __init__(self, num_layers=2):
        super().__init__()
        self.layer = nn.ModuleList([TransformerLayer() for _ in range(num_layers)])

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.encoder = Encoder(num_layers=3)

model = Model()
layer_idx = 1
width = 8

fig, ax = plt.subplots()

ARGS = {
    "model": model,
    "layer_idx": layer_idx,
    "width": width,
    "ax": ax
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy, scipy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Axes — a Axes output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Axes 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
prune_linear_layer Ran
ran on a synthesized input
epfml/attention-cnn/models/bert.py
code served (permissive licence) · get_code("2f516ae7fc833c82")
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 Linear. 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 torch import nn

torch.manual_seed(0)

# Create a linear layer with input dimension 3 and output dimension 4
# This will have weight shape (4, 3) and bias shape (4,)
layer = nn.Linear(in_features=3, out_features=4, bias=True)

# Create an index tensor to select which output dimensions to keep
# We'll keep dimensions 0 and 2 (out of 0, 1, 2, 3)
index = torch.tensor([0, 2], dtype=torch.long)

# Set dim to 0 to prune along the output dimension
dim = 0

ARGS = {
    "layer": layer,
    "index": index,
    "dim": dim
}
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 Linear — a Linear output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Linear 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
sizeof_fmt Ran
ran; our contract draft was wrong, not the code
epfml/attention-cnn/utils/logging.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from GeorgeLuImmortal/Hierarchical-BERT-Model-with-Limited-Labelled-Data, is pointer only (licence: NONE) · get_code("b6bef72aa7ff57a5")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from GeorgeLuImmortal/Hierarchical-BERT-Model-with-Limited-Labelled-Data, 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.

url_to_filename Ran
ran; our contract draft was wrong, not the code
epfml/attention-cnn/models/bert_utils.py
code served (permissive licence) · get_code("5eff22fa0a651276")
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 hashlib import sha256

ARGS = {
    "url": "https://example.com/model/checkpoint.bin",
    "etag": "abc123def456"
}
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
cached_path Not yet run
unverified (harvested, no recorded run)
epfml/attention-cnn/models/bert_utils.py
code served (permissive licence) · get_code("cf9c448c96f48428")
How it ran

no invocation recorded.

filename_to_url Not yet run
unverified (harvested, no recorded run)
epfml/attention-cnn/models/bert_utils.py
code served (permissive licence) · get_code("a6aab787fcf8b10d")
How it ran

no invocation recorded.

load_tf_weights_in_bert Not yet run
unverified (harvested, no recorded run)
epfml/attention-cnn/models/bert.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from sreyan88/toxicity-detection-in-spoken-utterances, is pointer only (licence: NONE) · get_code("b1a7180868dd8d6a")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from sreyan88/toxicity-detection-in-spoken-utterances, 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. 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, 7 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 5 the same month, 6 later) and 9 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
gelu 198 other papers
28 dated earlier, 3 the same month, 167 later · 32 only as an assembly brick
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.10125 (2019-01)
Glyce: Glyph-vectors for Chinese Character Representations
and 195 more not named here
url_to_filename 48 other papers
11 dated earlier, 1 the same month, 36 later · 8 only as an assembly brick
1610.02424 (2016-10)
Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Models
1809.11096 (2018-09)
Large Scale GAN Training for High Fidelity Natural Image Synthesis
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
and 45 more not named here
cached_path 27 other papers
7 dated earlier, 1 the same month, 19 later · 5 only as an assembly brick
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1901.02860 (2019-01)
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
1908.02265 (2019-08)
ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks
and 24 more not named here
filename_to_url 18 other papers
2 dated earlier, 1 the same month, 15 later
1908.09355 (2019-08)
Patient Knowledge Distillation for BERT Model Compression
1909.02209 (2019-09)
Semantics-aware BERT for Language Understanding
1911.02896 (2019-11)
Contextualized Sparse Representations for Real-Time Open-Domain Question Answering
and 15 more not named here
load_tf_weights_in_bert 15 other papers
2 dated earlier, 1 the same month, 12 later
1810.04805 (2018-10)
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
1908.09355 (2019-08)
Patient Knowledge Distillation for BERT Model Compression
1911.00720 (2019-11)
ZEN: Pre-training Chinese Text Encoder Enhanced by N-gram Representations
and 12 more not named here
human_format 1 other paper
1 later
2107.02331 (2021-07)
Mind Your Outliers! Investigating the Negative Impact of Outliers on Active Learning for Visual Question Answering
sizeof_fmt 1 other paper
1 dated earlier
1905.05583 (2019-05)
How to Fine-Tune BERT for Text Classification?

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

Recent trends of incorporating attention mechanisms in vision have led researchers to reconsider the supremacy of convolutional layers as a primary building block. Beyond helping CNNs to handle long-range dependencies, Ramachandran et al. (2019) showed that attention can completely replace convolution and achieve state-of-the-art performance on vision tasks. This raises the question: do learned attention layers operate similarly to convolutional layers? This work provides evidence that attention layers can perform convolution and, indeed, they often learn to do so in practice. Specifically, we prove that a multi-head self-attention layer with sufficient number of heads is at least as expressive as any convolutional layer. Our numerical experiments then show that self-attention layers attend to pixel-grid patterns similarly to CNN layers, corroborating our analysis. Our code is publicly available.

For agents

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

get_harvested_code_for_paper("1911.03584")
get_code_for_paper("1911.03584")
have("1911.03584")

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

curl https://syntology.ai/api/ran/1911.03584.json

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

[![Syntology run record](https://syntology.ai/api/ran/1911.03584.svg)](https://syntology.ai/paper/1911.03584)

Connect an agent — have() is free.