SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.04759 · NeurIPS · 2024

Probabilistic Weather Forecasting with Hierarchical Graph Neural Networks

Marc Deisenroth, Joel Oskarsson, Fredrik Lindsten, Tomas Landelius

arXiv · PDF · Open in the Atlas

Code that ran

11 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, 0 violated it and 3 have no contract verdict; 8 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/2406.04759.json · badge: /api/ran/2406.04759.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
joeloskarsson/neural-lam canonical (classifier label)
classifier basis: census
6 ran · 3 no recorded run
joeloskarsson/neural-lam-dev community (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
FunctionStatusWhere it lives
from_networkx_with_start_index Ran
ran; our contract draft was wrong, not the code
joeloskarsson/neural-lam-dev/neural_lam/create_graph.py
code served (permissive licence) · get_code("be958c03b6706529")
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 Data. 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 networkx as nx
from torch_geometric.utils.convert import from_networkx

torch.manual_seed(0)

# Create a small NetworkX graph
nx_graph = nx.Graph()
nx_graph.add_edges_from([(0, 1), (1, 2), (2, 0)])

# Define start_index as an integer
start_index = 5

ARGS = {
    "nx_graph": nx_graph,
    "start_index": start_index,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
networkx, torch, torch_geometric — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Data — a Data 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
from_networkx_with_start_index Ran
ran; our contract draft was wrong, not the code
joeloskarsson/neural-lam/neural_lam/create_graph.py
code served (permissive licence) · get_code("a4132a1a7527faea")
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 Data. 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 networkx
import torch_geometric as pyg

# Create a simple NetworkX graph
nx_graph = networkx.Graph()
nx_graph.add_edges_from([(0, 1), (1, 2), (2, 0)])

# Define the start_index offset
start_index = 5

ARGS = {
    "nx_graph": nx_graph,
    "start_index": start_index,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
networkx, torch_geometric — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Data — a Data 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
make_mlp Ran
ran; our contract draft was wrong, not the code
joeloskarsson/neural-lam/neural_lam/utils/networks.py
code served (permissive licence) · get_code("556864668ef1941e")
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
from torch import nn

torch.manual_seed(0)

# Test case 1: Simple 2-layer MLP (1 hidden layer)
blueprint_simple = [4, 8, 2]

# Test case 2: Deeper MLP (2 hidden layers)
blueprint_deep = [3, 5, 7, 4]

# Test case 3: Single layer (no hidden layers)
blueprint_single = [6, 2]

ARGS = {
    "blueprint": [4, 8, 6, 2],
    "layer_norm": 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 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
mask_and_reduce_metric Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
joeloskarsson/neural-lam/neural_lam/metrics.py
code served (permissive licence) · get_code("0baa433c352e2ed1")
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. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create metric_entry_vals tensor with shape (..., N, num_variables)
# Using batch dimensions (2,), N=5 grid nodes, num_variables=3
metric_entry_vals = torch.randn(2, 5, 3)

# Create a boolean mask with shape (N,) selecting some grid nodes
# N=5 to match the second-to-last dimension of metric_entry_vals
mask = torch.tensor([True, False, True, True, False])

# Boolean parameters
average_grid = True
sum_vars = False

ARGS = {
    "metric_entry_vals": metric_entry_vals,
    "mask": mask,
    "average_grid": average_grid,
    "sum_vars": sum_vars,
}
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: observe_signature.py could not observe it (import:AttributeError)
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
mask_and_reduce_metric Ran
ran on a synthesized input
joeloskarsson/neural-lam-dev/neural_lam/metrics.py
code served (permissive licence) · get_code("166d6e6e06a4b039")
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, 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 torch

torch.manual_seed(0)

# metric_entry_vals has shape (..., N, d_state)
# Let's use batch dimensions (...) = (2,), N = 5, d_state = 3
metric_entry_vals = torch.randn(2, 5, 3)

# mask has shape (N,) - boolean mask for grid nodes
# N = 5, so mask has 5 elements
mask = torch.tensor([True, False, True, True, False])

# average_grid: boolean flag
average_grid = True

# sum_vars: boolean flag
sum_vars = False

ARGS = {
    "metric_entry_vals": metric_entry_vals,
    "mask": mask,
    "average_grid": average_grid,
    "sum_vars": sum_vars,
}
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, 3) — shape (2, 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
plot_graph Ran
ran; our contract draft was wrong, not the code
joeloskarsson/neural-lam/neural_lam/create_graph.py
code served (permissive licence) · get_code("5fc397ff92181d7e")
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 torch
import torch_geometric as pyg
import numpy as np

torch.manual_seed(0)

# Create a small graph with node positions and edge indices
num_nodes = 5
num_edges = 6

# Create edge index (undirected graph)
edge_index = torch.tensor([
    [0, 1, 1, 2, 2, 3],
    [1, 0, 2, 1, 3, 2]
], dtype=torch.long)

# Create node positions (2D coordinates for plotting)
pos = torch.randn(num_nodes, 2, dtype=torch.float32)

# Create a PyG Data object
graph = pyg.data.Data(edge_index=edge_index, pos=pos)

ARGS = {
    "graph": graph,
    "title": "Test Graph"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy, torch, torch_geometric — 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
plot_graph Ran
ran on a synthesized input
joeloskarsson/neural-lam-dev/neural_lam/create_graph.py
code served (permissive licence) · get_code("a49a2231f1fecb3b")
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 torch
import torch_geometric as pyg
from torch_geometric.data import Data
import numpy as np

torch.manual_seed(0)

# Create a small graph with nodes and edges
num_nodes = 5
num_edges = 6

# Create node positions (2D coordinates)
pos = torch.randn(num_nodes, 2, dtype=torch.float32)

# Create edge index (undirected edges)
edge_index = torch.tensor([
    [0, 1, 1, 2, 2, 3],
    [1, 0, 2, 1, 3, 2]
], dtype=torch.long)

# Create a graph object with required attributes
graph = Data(pos=pos, edge_index=edge_index)

ARGS = {
    'graph': graph,
    'title': 'Test Graph Visualization'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy, torch, torch_geometric — 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
sort_nodes_internally Ran
ran; our contract draft was wrong, not the code
joeloskarsson/neural-lam-dev/neural_lam/create_graph.py
code served (permissive licence) · get_code("fd2a817d4ec92933")
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 DiGraph. 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 networkx
import torch

# Create a simple directed graph with nodes and edges
nx_graph = networkx.DiGraph()

# Add nodes with data attributes
nx_graph.add_node(3, label='node_3', value=30)
nx_graph.add_node(1, label='node_1', value=10)
nx_graph.add_node(2, label='node_2', value=20)

# Add edges with data attributes
nx_graph.add_edge(1, 2, weight=1.5)
nx_graph.add_edge(2, 3, weight=2.5)
nx_graph.add_edge(1, 3, weight=3.5)

ARGS = {
    'nx_graph': nx_graph
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
networkx, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DiGraph — a DiGraph output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DiGraph output the recorded ndim 1 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
sort_nodes_internally Ran
ran; our contract draft was wrong, not the code
joeloskarsson/neural-lam/neural_lam/create_graph.py
code served (permissive licence) · get_code("89282d125900c2ee")
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 DiGraph. 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 networkx

# Create a simple undirected graph with nodes that can be sorted
nx_graph = networkx.Graph()

# Add nodes with various labels (strings that can be sorted alphabetically)
nx_graph.add_nodes_from(['charlie', 'alice', 'bob', 'zoe'])

# Add some edges
nx_graph.add_edges_from([
    ('alice', 'bob'),
    ('bob', 'charlie'),
    ('charlie', 'zoe')
])

ARGS = {
    'nx_graph': nx_graph
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
networkx — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DiGraph — a DiGraph output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DiGraph 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
wmse Ran
ran; our contract draft was wrong, not the code
joeloskarsson/neural-lam/neural_lam/metrics.py
code served (permissive licence) · get_code("8b48ffe8dfc072b8")
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 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 torch

torch.manual_seed(0)

# Define tensor dimensions
batch_size = 2
num_grid_nodes = 5
num_variables = 3

# Create prediction tensor: shape (batch_size, num_grid_nodes, num_variables)
pred = torch.randn(batch_size, num_grid_nodes, num_variables)

# Create target tensor: same shape as pred
target = torch.randn(batch_size, num_grid_nodes, num_variables)

# Create predicted std tensor: same shape as pred
pred_std = torch.abs(torch.randn(batch_size, num_grid_nodes, num_variables)) + 0.1

# Create boolean mask over grid nodes: shape (num_grid_nodes,)
mask = torch.tensor([True, False, True, True, False])

ARGS = {
    "pred": pred,
    "target": target,
    "pred_std": pred_std,
    "mask": mask,
    "average_grid": True,
    "sum_vars": 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 Tensor, 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
wmse Ran
ran on a synthesized input
joeloskarsson/neural-lam-dev/neural_lam/metrics.py
code served (permissive licence) · get_code("e743fc22fb615b21")
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 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 torch

torch.manual_seed(0)

# Define tensor shapes according to docstring:
# (...,) = batch dimensions
# N = grid dimension (second to last)
# d_state = variable dimension (last)

# Using distinct sizes for clarity:
# batch dimensions: 2
# N (grid): 5
# d_state (variables): 3

batch_size = 2
N = 5
d_state = 3

# pred: (..., N, d_state)
pred = torch.randn(batch_size, N, d_state)

# target: (..., N, d_state)
target = torch.randn(batch_size, N, d_state)

# pred_std: (..., N, d_state) or (d_state,)
# Using (..., N, d_state) form
pred_std = torch.abs(torch.randn(batch_size, N, d_state)) + 0.1  # ensure positive

# mask: (N,), boolean mask
mask = torch.tensor([True, True, False, True, True])

ARGS = {
    'pred': pred,
    'target': target,
    'pred_std': pred_std,
    'mask': mask,
    'average_grid': True,
    'sum_vars': 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 Tensor, 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
get_gnn_class Not yet run
unverified (harvested, no recorded run)
joeloskarsson/neural-lam/neural_lam/gnn_layers.py
code served (permissive licence) · get_code("626aa0ff377e2739")
How it ran

no invocation recorded.

get_metric Not yet run
unverified (harvested, no recorded run)
joeloskarsson/neural-lam/neural_lam/metrics.py
code served (permissive licence) · get_code("754c85f8fa7a35e7")
How it ran

no invocation recorded.

get_metric Not yet run
unverified (harvested, no recorded run)
joeloskarsson/neural-lam-dev/neural_lam/metrics.py
code served (permissive licence) · get_code("5e4a4870f3bf627d")
How it ran

no invocation recorded.

make_gnn_seq Not yet run
unverified (harvested, no recorded run)
joeloskarsson/neural-lam/neural_lam/utils/networks.py
code served (permissive licence) · get_code("4e11dd84ee3be9ee")
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, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 later) and 13 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
from_networkx_with_start_index 1 other paper
1 later
2410.11807 (2024-10)
Regional Ocean Forecasting with Hierarchical Graph Neural Networks
sort_nodes_internally 1 other paper
1 later
2410.11807 (2024-10)
Regional Ocean Forecasting with Hierarchical Graph Neural Networks

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

In recent years, machine learning has established itself as a powerful tool for high-resolution weather forecasting. While most current machine learning models focus on deterministic forecasts, accurately capturing the uncertainty in the chaotic weather system calls for probabilistic modeling. We propose a probabilistic weather forecasting model called Graph-EFM, combining a flexible latent-variable formulation with the successful graph-based forecasting framework. The use of a hierarchical graph construction allows for efficient sampling of spatially coherent forecasts. Requiring only a single forward pass per time step, Graph-EFM allows for fast generation of arbitrarily large ensembles. We experiment with the model on both global and limited area forecasting. Ensemble forecasts from Graph-EFM achieve equivalent or lower errors than comparable deterministic models, with the added benefit of accurately capturing forecast uncertainty.

For agents

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

get_harvested_code_for_paper("2406.04759")
get_code_for_paper("2406.04759")
have("2406.04759")

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

curl https://syntology.ai/api/ran/2406.04759.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.04759.svg)](https://syntology.ai/paper/2406.04759)

Connect an agent — have() is free.