SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2306.10280 · NeurIPS · 2023

OpenGSL: A Comprehensive Benchmark for Graph Structure Learning

Can Wang, Daochen Zha, Jiawei Chen, Qiaoyu Tan, Sheng Zhou, Zhiyao Zhou, Bochao Mao, Xuanyi Zhou, Yan Feng, Chun Chen

arXiv · PDF · Open in the Atlas

Code that ran

5 of 8 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 5 have no contract verdict; 0 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/2306.10280.json · badge: /api/ran/2306.10280.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
OpenGSL/OpenGSL canonical (classifier label)
classifier basis: census
5 ran · 3 no recorded run
FunctionStatusWhere it lives
connectivity_regularizer Ran
ran on a synthesized input
OpenGSL/OpenGSL/opengsl/module/regularizer.py
code served (permissive licence) · get_code("79f850f9a3a82b60")
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 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

torch.manual_seed(0)

# The function expects an adjacency matrix (adj) as a torch.tensor
# It checks if it's sparse, and if so converts to dense
# Then it computes: -1 * log(sum of each row + 1e-12).sum()
# We need a tensor that represents an adjacency matrix

# Create a small adjacency matrix (2x2 or similar)
# Making it a regular dense tensor
adj = torch.randn(2, 4)  # Small adjacency-like matrix, 2 nodes with 4 features/connections
adj = torch.abs(adj)  # Make it positive since adjacency matrices typically have non-negative values

ARGS = {
    "adj": adj
}
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 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
get_lr_schedule_by_sigmoid Ran
ran on a synthesized input
OpenGSL/OpenGSL/opengsl/module/model/gaug.py
code served (permissive licence) · get_code("d3f9fb0654550887")
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 (10,) 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)

ARGS = {
    "n_epochs": 10,
    "lr": 0.001,
    "warmup": 2
}
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 (10,) — shape (10,) 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
normalize Ran
ran on a synthesized input
OpenGSL/OpenGSL/opengsl/module/functional.py
code served (permissive licence) · get_code("c1280fb4d9372247")
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 (4, 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

torch.manual_seed(0)

# Create a small dense adjacency matrix (4x4)
dense_mx = torch.tensor([
    [0.0, 1.0, 0.0, 1.0],
    [1.0, 0.0, 1.0, 0.0],
    [0.0, 1.0, 0.0, 1.0],
    [1.0, 0.0, 1.0, 0.0]
], dtype=torch.float32)

# Create a small sparse adjacency matrix (4x4)
indices = torch.tensor([[0, 0, 1, 1, 2, 2, 3, 3],
                        [1, 3, 0, 2, 1, 3, 0, 2]], dtype=torch.long)
values = torch.ones(8, dtype=torch.float32)
sparse_mx = torch.sparse.FloatTensor(indices, values, torch.Size([4, 4]))

ARGS = {
    'mx': dense_mx,
    'style': 'symmetric',
    'add_loop': True,
    'p': None
}
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 (4, 4) — shape (4, 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
row_nomalize Ran
ran on a synthesized input
OpenGSL/OpenGSL/opengsl/module/functional.py
code served (permissive licence) · get_code("58e37c6839a135cc")
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 (4, 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)

# Create a small sparse matrix for testing
# The function expects a matrix with sum(1) method and @ operator
# We'll use a regular dense tensor as a stand-in since the function
# doesn't explicitly require a sparse matrix type
mx = torch.tensor([
    [2.0, 4.0, 0.0],
    [0.0, 3.0, 1.0],
    [1.0, 1.0, 1.0],
    [0.0, 0.0, 2.0]
], dtype=torch.float32)

ARGS = {
    "mx": mx
}
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 (4, 3) — shape (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
row_normalize_sp Ran
ran on a synthesized input
OpenGSL/OpenGSL/opengsl/module/functional.py
code served (permissive licence) · get_code("28febf2764cf766b")
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. 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)

# Create a small sparse tensor for testing
# The function expects a sparse tensor with shape (rows, cols)
# Let's create a 4x3 sparse tensor with a few non-zero values
indices = torch.tensor([[0, 1, 2, 3], [0, 1, 2, 0]])
values = torch.tensor([1.0, 2.0, 3.0, 4.0])
size = torch.Size([4, 3])

mx = torch.sparse.FloatTensor(indices, values, size)

ARGS = {
    "mx": mx
}
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, shape (4, 3) — shape (4, 3) as a later run recorded it
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_edge_pred Not yet run
unverified (harvested, no recorded run)
OpenGSL/OpenGSL/opengsl/module/model/gaug.py
code served (permissive licence) · get_code("724a84b7b0869be6")
How it ran

no invocation recorded.

get_block_matrix Not yet run
unverified (harvested, no recorded run)
OpenGSL/OpenGSL/opengsl/module/model/bmgcn.py
code served (permissive licence) · get_code("41f697e38cae0382")
How it ran

no invocation recorded.

smoothness_regularizer_direct Not yet run
unverified (harvested, no recorded run)
OpenGSL/OpenGSL/opengsl/module/regularizer.py
code served (permissive licence) · get_code("cf5df45adab673fe")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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

Graph Neural Networks (GNNs) have emerged as the de facto standard for representation learning on graphs, owing to their ability to effectively integrate graph topology and node attributes. However, the inherent suboptimal nature of node connections, resulting from the complex and contingent formation process of graphs, presents significant challenges in modeling them effectively. To tackle this issue, Graph Structure Learning (GSL), a family of data-centric learning approaches, has garnered substantial attention in recent years. The core concept behind GSL is to jointly optimize the graph structure and the corresponding GNN models. Despite the proposal of numerous GSL methods, the progress in this field remains unclear due to inconsistent experimental protocols, including variations in datasets, data processing techniques, and splitting strategies. In this paper, we introduce OpenGSL, the first comprehensive benchmark for GSL, aimed at addressing this gap. OpenGSL enables a fair comparison among state-of-the-art GSL methods by evaluating them across various popular datasets using uniform data processing and splitting strategies. Through extensive experiments, we observe that existing GSL methods do not consistently outperform vanilla GNN counterparts. We also find that there is no significant correlation between the homophily of the learned structure and task performance, challenging the common belief. Moreover, we observe that the learned graph structure demonstrates a strong generalization ability across different GNN models, despite the high computational and space consumption. We hope that our open-sourced library will facilitate rapid and equitable evaluation and inspire further innovative research in this field. The code of the benchmark can be found in https://github.com/OpenGSL/OpenGSL.

For agents

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

get_harvested_code_for_paper("2306.10280")
get_code_for_paper("2306.10280")
have("2306.10280")

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

curl https://syntology.ai/api/ran/2306.10280.json

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

[![Syntology run record](https://syntology.ai/api/ran/2306.10280.svg)](https://syntology.ai/paper/2306.10280)

Connect an agent — have() is free.