SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.02054 · 2024

Towards Neural Scaling Laws on Graphs

arXiv · PDF · Open in the Atlas

Code that ran

5 of 7 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; 2 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 2 failures of our instrument; 2 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2402.02054.json · badge: /api/ran/2402.02054.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
liu-jingzhe/graph-scaling-laws canonical (classifier label)
classifier basis: census
5 ran · 2 no recorded run
FunctionStatusWhere it lives
add_zeros Ran
ran; our contract draft was wrong, not the code
liu-jingzhe/graph-scaling-laws/main_graph.py
code served (permissive licence) · get_code("a8175dac03aad20d")
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

torch.manual_seed(0)

class Data:
    def __init__(self, num_nodes):
        self.num_nodes = num_nodes
        self.x = torch.randn(num_nodes, 4)

data = Data(num_nodes=2)

ARGS = {
    "data": data
}
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 Data — a Data output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Data 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
cls_func Ran
ran, honoured the contract we drafted
liu-jingzhe/graph-scaling-laws/curve_draw.py
code served (permissive licence) · get_code("e9632709104bde12")
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, 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
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# - x: used in division (1 / x), so needs to be numeric (tensor or scalar)
# - a: used as exponent (q ** a), so needs to be numeric
# - b: used in division (-b / q ** a), so needs to be numeric
# - c: used in addition (1 / x + c), so needs to be numeric
# - d: used in addition, so needs to be numeric

# All parameters are used in arithmetic operations, so we can use scalars or tensors
# Using small tensors for variety

x = torch.tensor([[2.0, 3.0], [4.0, 5.0]])  # shape (2, 2)
a = torch.tensor(2.0)  # scalar tensor
b = torch.tensor([[1.0, 2.0], [3.0, 4.0]])  # shape (2, 2)
c = torch.tensor(0.5)  # scalar tensor
d = torch.tensor([[0.1, 0.2], [0.3, 0.4]])  # shape (2, 2)

ARGS = {
    'x': x,
    'a': a,
    'b': b,
    'c': c,
    'd': d
}
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 Tensor, ndim 2, dtype family float, shape (2, 2) — shape (2, 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
reg_func Ran
ran, violated the contract we drafted
liu-jingzhe/graph-scaling-laws/curve_draw.py
code served (permissive licence) · get_code("515ef82cb44f9ec4")
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
import numpy as np

torch.manual_seed(0)

# Analyze the function:
# reg_func(x, a, b, c):
#   q = x
#   return b / q ** a + c
# 
# The function takes x, a, b, c and performs:
# - exponentiation: q ** a
# - division: b / (result)
# - addition: (result) + c
#
# All parameters appear to be numeric values or tensors that support these operations.
# Let's use small tensors/scalars.

x = torch.tensor([2.0, 3.0])  # shape (2,)
a = torch.tensor(2.0)          # scalar
b = torch.tensor(4.0)          # scalar
c = torch.tensor(1.0)          # scalar

ARGS = {
    "x": x,
    "a": a,
    "b": b,
    "c": c
}
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 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
test Ran
ran, honoured the contract we drafted
liu-jingzhe/graph-scaling-laws/main_node.py
code served (permissive licence) · get_code("8e6629dff4ea0a71")
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) 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.nn as nn

torch.manual_seed(0)

# Create a minimal model that takes x and adj_t and outputs logits
class SimpleGNN(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, x, adj_t):
        return self.linear(x)

# Create model
model = SimpleGNN(input_dim=4, output_dim=3)
model.eval()

# Create minimal data object with required attributes
class Data:
    def __init__(self):
        self.x = torch.randn(6, 4)  # 6 nodes, 4 features
        self.adj_t = torch.randn(6, 6)  # adjacency matrix (not used in forward but passed)
        self.y = torch.tensor([[0], [1], [2], [0], [1], [2]])  # 6 labels

data = Data()

# Create split indices
split_idx = {
    'train': torch.tensor([0, 1]),
    'valid': torch.tensor([2, 3]),
    'test': torch.tensor([4, 5])
}

# Create minimal evaluator that computes accuracy
class SimpleEvaluator:
    def eval(self, input_dict):
        y_true = input_dict['y_true']
        y_pred = input_dict['y_pred']
        acc = (y_true == y_pred).float().mean().item()
        return {'acc': acc}

evaluator = SimpleEvaluator()

ARGS = {
    'model': model,
    'data': data,
    'split_idx': split_idx,
    'evaluator': evaluator
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — 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
train Ran
ran; our contract draft was wrong, not the code
liu-jingzhe/graph-scaling-laws/main_node.py
code served (permissive licence) · get_code("82f54a95d4951b19")
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) and returned float, ndim 0, dtype family float, 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 torch.nn as nn
import torch.nn.functional as F
from torch.optim import Adam

torch.manual_seed(0)

# Build a minimal model that accepts (x, adj_t) and returns logits
class SimpleGNNModel(nn.Module):
    def __init__(self, input_dim, hidden_dim, output_dim):
        super().__init__()
        self.lin1 = nn.Linear(input_dim, hidden_dim)
        self.lin2 = nn.Linear(hidden_dim, output_dim)
    
    def forward(self, x, adj_t):
        # x: [num_nodes, input_dim]
        # adj_t: adjacency matrix
        # Simple MLP-based model
        x = F.relu(self.lin1(x))
        x = self.lin2(x)
        return F.log_softmax(x, dim=1)

# Create a minimal data object with required attributes
class Data:
    def __init__(self):
        self.x = torch.randn(5, 4)  # 5 nodes, 4 features
        self.adj_t = torch.eye(5)   # Simple identity adjacency matrix
        self.y = torch.tensor([[0], [1], [0], [2], [1]])  # 5 labels, 3 classes

# Instantiate components
model = SimpleGNNModel(input_dim=4, hidden_dim=8, output_dim=3)
data = Data()
train_idx = torch.tensor([0, 1, 2])  # Train on first 3 nodes
optimizer = Adam(model.parameters(), lr=0.01)

ARGS = {
    "model": model,
    "data": data,
    "train_idx": train_idx,
    "optimizer": optimizer
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, 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
eval Not yet run
unverified (harvested, no recorded run)
liu-jingzhe/graph-scaling-laws/main_graph.py
code served (permissive licence) · get_code("5ee2de00d5a3ab15")
How it ran

no invocation recorded.

train Not yet run
unverified (harvested, no recorded run)
liu-jingzhe/graph-scaling-laws/main_graph.py
code served (permissive licence) · get_code("4c7246bc9ee8803d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 2 later) and 4 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
test 5 other papers
2 dated earlier, 3 later
2111.00064 (2021-11)
Node Feature Extraction by Self-Supervised Multi-scale Neighborhood Prediction
2208.04609 (2022-08)
E2EG: End-to-End Node Classification Using Graph Topology and Text-based Node Attributes
2405.14303 (2024-05)
Similarity-Navigated Conformal Prediction for Graph Neural Networks
and 2 more not named here
train 4 other papers
2 dated earlier, 2 later
2111.00064 (2021-11)
Node Feature Extraction by Self-Supervised Multi-scale Neighborhood Prediction
2208.04609 (2022-08)
E2EG: End-to-End Node Classification Using Graph Topology and Text-based Node Attributes
2405.14303 (2024-05)
Similarity-Navigated Conformal Prediction for Graph Neural Networks
and 1 more not named here
add_zeros 1 other paper
1 dated earlier
2010.09891 (2020-10)
Robust Optimization as Data Augmentation for Large-scale Graphs

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

Deep graph models (e.g., graph neural networks and graph transformers) have become important techniques for leveraging knowledge across various types of graphs. Yet, the neural scaling laws on graphs, i.e., how the performance of deep graph models changes with model and dataset sizes, have not been systematically investigated, casting doubts on the feasibility of achieving large graph models. To fill this gap, we benchmark many graph datasets from different tasks and make an attempt to establish the neural scaling laws on graphs from both model and data perspectives. The model size we investigated is up to 100 million parameters, and the dataset size investigated is up to 50 million samples. We first verify the validity of such laws on graphs, establishing proper formulations to describe the scaling behaviors. For model scaling, we identify that despite the parameter numbers, the model depth also plays an important role in affecting the model scaling behaviors, which differs from observations in other domains such as CV and NLP. For data scaling, we suggest that the number of graphs can not effectively measure the graph data volume in scaling law since the sizes of different graphs are highly irregular. Instead, we reform the data scaling law with the number of nodes or edges as the metric to address the irregular graph sizes. We further demonstrate that the reformed law offers a unified view of the data scaling behaviors for various fundamental graph tasks including node classification, link prediction, and graph classification. This work provides valuable insights into neural scaling laws on graphs, which can serve as an important tool for collecting new graph data and developing large graph models.

For agents

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

get_harvested_code_for_paper("2402.02054")
get_code_for_paper("2402.02054")
have("2402.02054")

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

curl https://syntology.ai/api/ran/2402.02054.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.02054.svg)](https://syntology.ai/paper/2402.02054)

Connect an agent — have() is free.