Papers › Fine-tuning Graph Neural Networks by Preserving Graph Generative Patterns

Fine-tuning Graph Neural Networks by Preserving Graph Generative Patterns

21 Dec 2023arXiv:2312.13583archive 2025-07-28

Yifei Sun, Qi Zhu, Yang Yang, Chunping Wang, Tianyu Fan, Jiajun Zhu, Lei Chen

Recently, the paradigm of pre-training and fine-tuning graph neural networks has been intensively studied and applied in a wide range of graph mining tasks. Its success is generally attributed to the structural consistency between pre-training and downstream datasets, which, however, does not hold in many real-world scenarios. Existing works have shown that the structural divergence between pre-training and downstream graphs significantly limits the transferability when using the vanilla fine-tuning strategy. This divergence leads to model overfitting on pre-training graphs and causes difficulties in capturing the structural properties of the downstream graphs. In this paper, we identify the fundamental cause of structural divergence as the discrepancy of generative patterns between the pre-training and downstream graphs. Furthermore, we propose G-Tuning to preserve the generative patterns of downstream graphs. Given a downstream graph G, the core idea is to tune the pre-trained GNN so that it can reconstruct the generative patterns of G, the graphon W. However, the exact reconstruction of a graphon is known to be computationally expensive. To overcome this challenge, we provide a theoretical analysis that establishes the existence of a set of alternative graphons called graphon bases for any given graphon. By utilizing a linear combination of these graphon bases, we can efficiently approximate W. This theoretical finding forms the basis of our proposed model, as it enables effective learning of the graphon bases and their associated coefficients. Compared with existing algorithms, G-Tuning demonstrates an average improvement of 0.5% and 2.6% on in-domain and out-of-domain transfer learning experiments, respectively.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 13 code samples harvested from 1 repository linked to this paper: 13 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 3 honoured, 2 violated, 5 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 13 samples from 1 repository, 13 ran (10 with no instrument failure; 3 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

zjunet/G-Tuning officialmentioned in papermentioned on GitHubpytorchSyntology: 13 samples, 13 ran (3 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

13 samples harvested: 13 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 3 honoured, 2 violated, 5 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · honoured contract
2ran · violated contract
3ran · our draft was wrong
5ran

Licence: 13 of the 13 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from zjunet/G-Tuning. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

zjunet/G-Tuning official repository 13 samples, 13 ran (3 instrument-failed) · 0 unverified
align_graphs GCC-based/methods/learner.py ran · our draft was wrong no licence file found · pointer only · d894f2c5cbab6c96 · report fixture: signature graphs:list/1/ndarray, padding:bool/0/bool · returned tuple · level recorded 2026-09-22
distance_matrix GCC-based/ot_distance.py ran fingerprinted no licence file found · pointer only · 8f6a0627aad6b737 · report fixture: signature pts_src:Tensor/2/float, pts_dst:Tensor/2/float, p:int/0/int · returned Tensor · level recorded 2026-09-24
distance_tensor GCC-based/ot_distance.py ran no licence file found · pointer only · 2b819d2bb1b0bcdb · report fixture: signature pts_src:Tensor/2/float, pts_dst:Tensor/2/float, p:int/0/int · returned Tensor · level recorded 2026-09-24
estimate_target_distribution GCC-based/methods/learner.py ran · honoured contract no licence file found · pointer only · 6faf9171aaf972ca · report fixture: signature probs:list/1/ndarray, dim_t:int/0/int · returned ndarray · level recorded 2026-09-22
graph_numpy2tensor GCC-based/methods/learner.py ran · violated contract fingerprinted no licence file found · pointer only · 45f3526206f9a0f4 · report fixture: signature graphs:list/1/ndarray · returned Tensor · level recorded 2026-09-22
gw_distance GCC-based/methods/simulator.py ran · violated contract no licence file found · pointer only · c0c6de6502f72024 · report fixture: signature graphon:ndarray/2/float, estimation:ndarray/2/float · returned float64 · level recorded 2026-09-24
option_update GCC-based/train_G_Tuning.py ran · our draft was wrong no licence file found · pointer only · 1df7f49664c77c5e · report fixture: signature opt:Options/0/? · returned Options · level recorded 2026-09-24
sampling_gaussian GCC-based/model.py ran fingerprinted no licence file found · pointer only · 1fa699677ecc4d72 · report fixture: signature mu:Tensor/2/float, logvar:Tensor/2/float, num_sample:int/0/int · returned Tensor · level recorded 2026-09-24
sampling_gmm GCC-based/model.py ran fingerprinted no licence file found · pointer only · 9b58c30cbb07b507 · report fixture: signature mu:Tensor/2/float, logvar:Tensor/2/float, num_sample:int/0/int · returned Tensor · level recorded 2026-09-24
simulate_graphs GCC-based/methods/simulator.py ran no licence file found · pointer only · 2ded63c00749fb47 · report fixture: signature w:ndarray/2/float, num_graphs:int/0/int, num_nodes:int/0/int, graph_size:str/0/str · returned list · level recorded 2026-09-24
sum_matrix GCC-based/ot_distance.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 08bfd3fe7445ab40 · report fixture: signature pts_src:Tensor/2/float, pts_dst:Tensor/2/float · returned Tensor · level recorded 2026-09-24
synthesize_graphon GCC-based/methods/simulator.py ran · honoured contract fingerprinted no licence file found · pointer only · 3c505c6193699765 · report fixture: signature r:int/0/int, type_idx:int/0/int · returned ndarray · level recorded 2026-09-24
test_moco GCC-based/generate.py ran · honoured contract no licence file found · pointer only · 48d3912bc3a7d2b9 · report fixture: signature train_loader:list/1/tuple, model:callable/0/?, opt:Options/0/? · returned Tensor · level recorded 2026-09-22

Tasks

Graph ClassificationGraph LearningGraph MiningGraph Neural NetworkTransfer Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Graph Classification BACE G-Tuning ROC-AUC 84.79 #1 of 2 Archive leaderboard report
Graph Classification BBBP G-Tuning ROC-AUC 72.59 #1 of 3 Archive leaderboard report
Graph Classification ENZYMES G-Tuning Accuracy (10-fold) 26.70 #54 of 54 Archive leaderboard report
Graph Classification HIV G-Tuning ROC-AUC 77.33 #3 of 3 Archive leaderboard report
Graph Classification IMDb-B G-Tuning Accuracy (10-fold) 74.30 #51 of 51 Archive leaderboard report
Graph Classification IMDb-M G-Tuning Accuracy (10-fold) 51.80 #36 of 36 Archive leaderboard report
Graph Classification MSRC-21 (per-class) G-Tuning Accuracy (10 fold) 11.01 #1 of 1 Archive leaderboard report
Graph Classification MUTAG G-Tuning Accuracy (10 fold) 86.14 #74 of 74 Archive leaderboard report
Graph Classification MUV G-Tuning ROC-AUC 75.84 #2 of 2 Archive leaderboard report
Graph Classification PROTEINS G-Tuning Accuracy (10 fold) 72.05 #102 of 103 Archive leaderboard report
Graph Classification REDDIT-12K G-Tuning Accuracy (10 fold) 42.80 #1 of 1 Archive leaderboard report
Graph Classification SIDER G-Tuning ROC-AUC 61.40 #2 of 2 Archive leaderboard report
Graph Classification Tox21 G-Tuning ROC-AUC 75.80 #2 of 3 Archive leaderboard report
Graph Classification ToxCast G-Tuning ROC-AUC 64.25 #2 of 3 Archive leaderboard report
Graph Classification clintox G-Tuning ROC-AUC 74.64 #1 of 2 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Discriminative Fine-TuningGraph Neural Network

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections