SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.01899 · NeurIPS · 2025

Cross-Domain Graph Data Scaling: A Showcase with Diffusion Models

A Research, J Morgan, Jiliang Tang, Haitao Mao, Yuying Xie, Saumitra Mishra, Ivan Brugere, Wenzhuo Tang, Danial Dervovic

arXiv · PDF · Open in the Atlas

Code that ran

6 of 10 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 0 have no contract verdict; 6 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/2406.01899.json · badge: /api/ran/2406.01899.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
wenzhuotang/uniaug canonical (classifier label)
classifier basis: census
6 ran · 4 no recorded run
FunctionStatusWhere it lives
creat_pooling_layers Ran
ran; our contract draft was wrong, not the code
wenzhuotang/uniaug/models/archived.py
code served (permissive licence) · get_code("e271b69faea43984")
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 list, ndim 1, dtype family float; no output shape is stated. 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 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
from collections.abc import Iterable

torch.manual_seed(0)

# Test case 1: Single pooling layer with basic config
config_single = {
    'pool': 'sum',
    'subset': False,
    'k': 1,
    'subset_type': 'top'
}

# Test case 2: Multiple pooling layers with list arguments
config_multiple = {
    'pool': ['sum', 'mean', 'max'],
    'subset': [False, True, False],
    'k': [1, 2, 1],
    'subset_type': ['top', 'random', 'bottom']
}

# Test case 3: Mixed iterable and non-iterable arguments
config_mixed = {
    'pool': ['sum', 'mean'],
    'subset': False,
    'k': [1, 2],
    'subset_type': 'top'
}

# Test case 4: Single element lists
config_single_list = {
    'pool': ['mul'],
    'subset': [True],
    'k': [3],
    'subset_type': ['random']
}

ARGS = {
    'config': config_single
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torch_geometric — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list 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
create_activation Ran
ran; our contract draft was wrong, not the code
wenzhuotang/uniaug/diffusion/layers.py
code served (permissive licence) · get_code("dd2c145b850aaac3")
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 ReLU. 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)

# The function takes a single parameter 'name' which is a string or None
# It creates and returns different activation functions based on the name

ARGS = {
    'name': 'relu'
}
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 ReLU — a ReLU output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ReLU 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
create_norm Ran
ran; our contract draft was wrong, not the code
wenzhuotang/uniaug/diffusion/layers.py
code served (permissive licence) · get_code("1dcf0e58150c4828")
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 Identity. 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)

# Analyze function usage:
# - name: string or None, used for string comparisons and startswith()
# - n: integer, passed to LayerNorm, BatchNorm1d, GroupNorm
# - h: integer with default 16, passed to GroupNorm as num_groups

# Valid test cases for the function
ARGS = {
    "name": None,
    "n": 8,
    "h": 16
}
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 Identity — a Identity output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Identity 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
global_mul_pool Ran
ran; our contract draft was wrong, not the code
wenzhuotang/uniaug/models/archived.py
code served (permissive licence) · get_code("8a883a62d06acc31")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from torch_geometric.utils import scatter

torch.manual_seed(0)

# Create feature tensor x with shape (num_nodes, feature_dim)
# Using distinct dimensions: 6 nodes, 4 features
x = torch.randn(6, 4)

# Create batch tensor that assigns each node to a graph in the batch
# We have 2 graphs: nodes 0-2 go to graph 0, nodes 3-5 go to graph 1
batch = torch.tensor([0, 0, 0, 1, 1, 1])

# size parameter: number of graphs in the batch (optional)
size = 2

ARGS = {
    "x": x,
    "batch": batch,
    "size": size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torch_geometric — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
global_mul_pool Ran
ran; our contract draft was wrong, not the code
wenzhuotang/uniaug/models/layers.py
code served (permissive licence) · get_code("0c61658e972578ed")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from torch_geometric.utils import scatter

torch.manual_seed(0)

# Create a 2D tensor (batch_size=2, num_features=4)
x = torch.randn(6, 4)

# Create batch indices indicating which graph each node belongs to
# For 2 graphs with 3 nodes each
batch = torch.tensor([0, 0, 0, 1, 1, 1], dtype=torch.long)

# Optional size parameter (number of graphs)
size = 2

ARGS = {
    'x': x,
    'batch': batch,
    'size': size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torch_geometric — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
test_edge Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
wenzhuotang/uniaug/link_pred.py
code served (permissive licence) · get_code("db940d528be63d7a")
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 (3, 1) 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

torch.manual_seed(0)

# Define a simple score function that takes two node embeddings and returns a score
class ScoreFunc(nn.Module):
    def __init__(self, embedding_dim):
        super().__init__()
        self.linear = nn.Linear(embedding_dim * 2, 1)
    
    def forward(self, src_emb, dst_emb):
        combined = torch.cat([src_emb, dst_emb], dim=-1)
        return self.linear(combined)

# Set up parameters
embedding_dim = 4
num_nodes = 6
num_edges = 3

score_func = ScoreFunc(embedding_dim)

# edge_index: [2, num_edges] tensor with source and target node indices
edge_index = torch.tensor([
    [0, 1, 2],
    [1, 2, 3]
], dtype=torch.long)

# h: [num_nodes, embedding_dim] node embeddings
h = torch.randn(num_nodes, embedding_dim)

batch_size = 2

# For non-mrr_mode test
mrr_mode = False
negative_data = None

# Alternative for mrr_mode test
# mrr_mode = True
# negative_data = torch.randint(0, num_nodes, (num_edges * 1000,), dtype=torch.long)

ARGS = {
    "score_func": score_func,
    "edge_index": edge_index,
    "h": h,
    "batch_size": batch_size,
    "mrr_mode": mrr_mode,
    "negative_data": negative_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 Tensor, ndim 2, dtype family float, shape (3, 1) — shape (3, 1) 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)
wenzhuotang/uniaug/graph_pred.py
code served (permissive licence) · get_code("b55ab9a9350a539c")
How it ran

no invocation recorded.

jaccard_index Not yet run
unverified (harvested, no recorded run)
wenzhuotang/uniaug/models/cf_utils.py
code served (permissive licence) · get_code("75d3e57ef9228c89")
How it ran

no invocation recorded.

run_graph_pred Not yet run
unverified (harvested, no recorded run)
wenzhuotang/uniaug/graph_pred.py
code served (permissive licence) · get_code("2dcf7d8c7606483e")
How it ran

no invocation recorded.

train Not yet run
unverified (harvested, no recorded run)
wenzhuotang/uniaug/graph_pred.py
code served (permissive licence) · get_code("cf44ccd02553d4ce")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 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.

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

Models for natural language and images benefit from data scaling behavior: the more data fed into the model, the better they perform. This 'better with more' phenomenon enables the effectiveness of large-scale pre-training on vast amounts of data. However, current graph pre-training methods struggle to scale up data due to heterogeneity across graphs. To achieve effective data scaling, we aim to develop a general model that is able to capture diverse data patterns of graphs and can be utilized to adaptively help the downstream tasks. To this end, we propose UniAug, a universal graph structure augmentor built on a diffusion model. We first pre-train a discrete diffusion model on thousands of graphs across domains to learn the graph structural patterns. In the downstream phase, we provide adaptive enhancement by conducting graph structure augmentation with the help of the pre-trained diffusion model via guided generation. By leveraging the pre-trained diffusion model for structure augmentation, we consistently achieve performance improvements across various downstream tasks in a plug-and-play manner. To the best of our knowledge, this study represents the first demonstration of a data-scaling graph structure augmentor on graphs across domains.

For agents

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

get_harvested_code_for_paper("2406.01899")
get_code_for_paper("2406.01899")
have("2406.01899")

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

curl https://syntology.ai/api/ran/2406.01899.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.01899.svg)](https://syntology.ai/paper/2406.01899)

Connect an agent — have() is free.