TGM: a Modular and Efficient Library for Machine Learning on Temporal Graphs Article Swipe
Jacob Chmura
,
Shenyang Huang
,
Tran Gia Bao Ngo
,
Farimah Poursafaei
,
Jure Leskovec
,
Guillaume Rabusseau
,
Matthias Fey
,
Reihaneh Rabbany
·
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2510.07586
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2510.07586
Well-designed open-source software drives progress in Machine Learning (ML) research. While static graph ML enjoys mature frameworks like PyTorch Geometric and DGL, ML for temporal graphs (TG), networks that evolve over time, lacks comparable infrastructure. Existing TG libraries are often tailored to specific architectures, hindering support for diverse models in this rapidly evolving field. Additionally, the divide between continuous- and discrete-time dynamic graph methods (CTDG and DTDG) limits direct comparisons and idea transfer. To address these gaps, we introduce Temporal Graph Modelling (TGM), a research-oriented library for ML on temporal graphs, the first to unify CTDG and DTDG approaches. TGM offers first-class support for dynamic node features, time-granularity conversions, and native handling of link-, node-, and graph-level tasks. Empirically, TGM achieves an average 7.8x speedup across multiple models, datasets, and tasks compared to the widely used DyGLib, and an average 175x speedup on graph discretization relative to available implementations. Beyond efficiency, we show in our experiments how TGM unlocks entirely new research possibilities by enabling dynamic graph property prediction and time-driven training paradigms, opening the door to questions previously impractical to study. TGM is available at https://github.com/tgm-team/tgm
Related Topics
Concepts
No concepts available.
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2510.07586
- https://arxiv.org/pdf/2510.07586
- OA Status
- green
- OpenAlex ID
- https://openalex.org/W4416395529
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4416395529Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2510.07586Digital Object Identifier
- Title
-
TGM: a Modular and Efficient Library for Machine Learning on Temporal GraphsWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-10-08Full publication date if available
- Authors
-
Jacob Chmura, Shenyang Huang, Tran Gia Bao Ngo, Farimah Poursafaei, Jure Leskovec, Guillaume Rabusseau, Matthias Fey, Reihaneh RabbanyList of authors in order
- Landing page
-
https://arxiv.org/abs/2510.07586Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2510.07586Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2510.07586Direct OA link when available
- Cited by
-
0Total citation count in OpenAlex
Full payload
| id | https://openalex.org/W4416395529 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2510.07586 |
| ids.doi | https://doi.org/10.48550/arxiv.2510.07586 |
| ids.openalex | https://openalex.org/W4416395529 |
| fwci | |
| type | preprint |
| title | TGM: a Modular and Efficient Library for Machine Learning on Temporal Graphs |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2510.07586 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | |
| locations[0].pdf_url | https://arxiv.org/pdf/2510.07586 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2510.07586 |
| locations[1].id | doi:10.48550/arxiv.2510.07586 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | cc-by |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | https://openalex.org/licenses/cc-by |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2510.07586 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5093047927 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Jacob Chmura |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Chmura, Jacob |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5103093830 |
| authorships[1].author.orcid | https://orcid.org/0000-0001-8788-8652 |
| authorships[1].author.display_name | Shenyang Huang |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Huang, Shenyang |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5111233212 |
| authorships[2].author.orcid | |
| authorships[2].author.display_name | Tran Gia Bao Ngo |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Ngo, Tran Gia Bao |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5021512115 |
| authorships[3].author.orcid | https://orcid.org/0009-0006-6785-3253 |
| authorships[3].author.display_name | Farimah Poursafaei |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Poursafaei, Farimah |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5091272738 |
| authorships[4].author.orcid | https://orcid.org/0000-0002-5411-923X |
| authorships[4].author.display_name | Jure Leskovec |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Leskovec, Jure |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5023766963 |
| authorships[5].author.orcid | https://orcid.org/0000-0001-8090-2810 |
| authorships[5].author.display_name | Guillaume Rabusseau |
| authorships[5].author_position | last |
| authorships[5].raw_author_name | Rabusseau, Guillaume |
| authorships[5].is_corresponding | False |
| authorships[6].author.id | https://openalex.org/A5002742256 |
| authorships[6].author.orcid | https://orcid.org/0000-0002-5727-0701 |
| authorships[6].author.display_name | Matthias Fey |
| authorships[6].author_position | middle |
| authorships[6].raw_author_name | Fey, Matthias |
| authorships[6].is_corresponding | False |
| authorships[7].author.id | https://openalex.org/A5043159673 |
| authorships[7].author.orcid | https://orcid.org/0000-0003-2348-0353 |
| authorships[7].author.display_name | Reihaneh Rabbany |
| authorships[7].author_position | middle |
| authorships[7].raw_author_name | Rabbany, Reihaneh |
| authorships[7].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2510.07586 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-11T00:00:00 |
| display_name | TGM: a Modular and Efficient Library for Machine Learning on Temporal Graphs |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-28T12:58:32.679985 |
| primary_topic | |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2510.07586 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2510.07586 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2510.07586 |
| primary_location.id | pmh:oai:arXiv.org:2510.07586 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | |
| primary_location.pdf_url | https://arxiv.org/pdf/2510.07586 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2510.07586 |
| publication_date | 2025-10-08 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 83 |
| abstract_inverted_index.ML | 13, 22, 87 |
| abstract_inverted_index.TG | 36 |
| abstract_inverted_index.To | 73 |
| abstract_inverted_index.an | 121, 138 |
| abstract_inverted_index.at | 185 |
| abstract_inverted_index.by | 163 |
| abstract_inverted_index.in | 5, 49, 153 |
| abstract_inverted_index.is | 183 |
| abstract_inverted_index.of | 112 |
| abstract_inverted_index.on | 88, 142 |
| abstract_inverted_index.to | 41, 93, 132, 146, 176, 180 |
| abstract_inverted_index.we | 77, 151 |
| abstract_inverted_index.TGM | 99, 119, 157, 182 |
| abstract_inverted_index.and | 20, 59, 65, 70, 96, 109, 115, 129, 137, 169 |
| abstract_inverted_index.are | 38 |
| abstract_inverted_index.for | 23, 46, 86, 103 |
| abstract_inverted_index.how | 156 |
| abstract_inverted_index.new | 160 |
| abstract_inverted_index.our | 154 |
| abstract_inverted_index.the | 55, 91, 133, 174 |
| abstract_inverted_index.(ML) | 8 |
| abstract_inverted_index.175x | 140 |
| abstract_inverted_index.7.8x | 123 |
| abstract_inverted_index.CTDG | 95 |
| abstract_inverted_index.DGL, | 21 |
| abstract_inverted_index.DTDG | 97 |
| abstract_inverted_index.door | 175 |
| abstract_inverted_index.idea | 71 |
| abstract_inverted_index.like | 17 |
| abstract_inverted_index.node | 105 |
| abstract_inverted_index.over | 30 |
| abstract_inverted_index.show | 152 |
| abstract_inverted_index.that | 28 |
| abstract_inverted_index.this | 50 |
| abstract_inverted_index.used | 135 |
| abstract_inverted_index.(CTDG | 64 |
| abstract_inverted_index.(TG), | 26 |
| abstract_inverted_index.DTDG) | 66 |
| abstract_inverted_index.Graph | 80 |
| abstract_inverted_index.While | 10 |
| abstract_inverted_index.first | 92 |
| abstract_inverted_index.gaps, | 76 |
| abstract_inverted_index.graph | 12, 62, 143, 166 |
| abstract_inverted_index.lacks | 32 |
| abstract_inverted_index.often | 39 |
| abstract_inverted_index.tasks | 130 |
| abstract_inverted_index.these | 75 |
| abstract_inverted_index.time, | 31 |
| abstract_inverted_index.unify | 94 |
| abstract_inverted_index.(TGM), | 82 |
| abstract_inverted_index.Beyond | 149 |
| abstract_inverted_index.across | 125 |
| abstract_inverted_index.direct | 68 |
| abstract_inverted_index.divide | 56 |
| abstract_inverted_index.drives | 3 |
| abstract_inverted_index.enjoys | 14 |
| abstract_inverted_index.evolve | 29 |
| abstract_inverted_index.field. | 53 |
| abstract_inverted_index.graphs | 25 |
| abstract_inverted_index.limits | 67 |
| abstract_inverted_index.link-, | 113 |
| abstract_inverted_index.mature | 15 |
| abstract_inverted_index.models | 48 |
| abstract_inverted_index.native | 110 |
| abstract_inverted_index.node-, | 114 |
| abstract_inverted_index.offers | 100 |
| abstract_inverted_index.static | 11 |
| abstract_inverted_index.study. | 181 |
| abstract_inverted_index.tasks. | 117 |
| abstract_inverted_index.widely | 134 |
| abstract_inverted_index.DyGLib, | 136 |
| abstract_inverted_index.Machine | 6 |
| abstract_inverted_index.PyTorch | 18 |
| abstract_inverted_index.address | 74 |
| abstract_inverted_index.average | 122, 139 |
| abstract_inverted_index.between | 57 |
| abstract_inverted_index.diverse | 47 |
| abstract_inverted_index.dynamic | 61, 104, 165 |
| abstract_inverted_index.graphs, | 90 |
| abstract_inverted_index.library | 85 |
| abstract_inverted_index.methods | 63 |
| abstract_inverted_index.models, | 127 |
| abstract_inverted_index.opening | 173 |
| abstract_inverted_index.rapidly | 51 |
| abstract_inverted_index.speedup | 124, 141 |
| abstract_inverted_index.support | 45, 102 |
| abstract_inverted_index.unlocks | 158 |
| abstract_inverted_index.Existing | 35 |
| abstract_inverted_index.Learning | 7 |
| abstract_inverted_index.Temporal | 79 |
| abstract_inverted_index.achieves | 120 |
| abstract_inverted_index.compared | 131 |
| abstract_inverted_index.enabling | 164 |
| abstract_inverted_index.entirely | 159 |
| abstract_inverted_index.evolving | 52 |
| abstract_inverted_index.handling | 111 |
| abstract_inverted_index.multiple | 126 |
| abstract_inverted_index.networks | 27 |
| abstract_inverted_index.progress | 4 |
| abstract_inverted_index.property | 167 |
| abstract_inverted_index.relative | 145 |
| abstract_inverted_index.research | 161 |
| abstract_inverted_index.software | 2 |
| abstract_inverted_index.specific | 42 |
| abstract_inverted_index.tailored | 40 |
| abstract_inverted_index.temporal | 24, 89 |
| abstract_inverted_index.training | 171 |
| abstract_inverted_index.Geometric | 19 |
| abstract_inverted_index.Modelling | 81 |
| abstract_inverted_index.available | 147, 184 |
| abstract_inverted_index.datasets, | 128 |
| abstract_inverted_index.features, | 106 |
| abstract_inverted_index.hindering | 44 |
| abstract_inverted_index.introduce | 78 |
| abstract_inverted_index.libraries | 37 |
| abstract_inverted_index.questions | 177 |
| abstract_inverted_index.research. | 9 |
| abstract_inverted_index.transfer. | 72 |
| abstract_inverted_index.comparable | 33 |
| abstract_inverted_index.frameworks | 16 |
| abstract_inverted_index.paradigms, | 172 |
| abstract_inverted_index.prediction | 168 |
| abstract_inverted_index.previously | 178 |
| abstract_inverted_index.approaches. | 98 |
| abstract_inverted_index.comparisons | 69 |
| abstract_inverted_index.continuous- | 58 |
| abstract_inverted_index.efficiency, | 150 |
| abstract_inverted_index.experiments | 155 |
| abstract_inverted_index.first-class | 101 |
| abstract_inverted_index.graph-level | 116 |
| abstract_inverted_index.impractical | 179 |
| abstract_inverted_index.open-source | 1 |
| abstract_inverted_index.time-driven | 170 |
| abstract_inverted_index.Empirically, | 118 |
| abstract_inverted_index.conversions, | 108 |
| abstract_inverted_index.Additionally, | 54 |
| abstract_inverted_index.Well-designed | 0 |
| abstract_inverted_index.discrete-time | 60 |
| abstract_inverted_index.possibilities | 162 |
| abstract_inverted_index.architectures, | 43 |
| abstract_inverted_index.discretization | 144 |
| abstract_inverted_index.infrastructure. | 34 |
| abstract_inverted_index.implementations. | 148 |
| abstract_inverted_index.time-granularity | 107 |
| abstract_inverted_index.research-oriented | 84 |
| abstract_inverted_index.https://github.com/tgm-team/tgm | 186 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 8 |
| citation_normalized_percentile |