Pith. sign in

REVIEW 2 cited by

Retrofitting Temporal Graph Neural Networks with Transformer

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2409.05477 v3 pith:IGMKXJEO submitted 2024-09-09 cs.LG

classification cs.LG
keywords graphtf-tgntrainingtransformertemporaltgnnsmodelingtgnn
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Temporal graph neural networks (TGNNs) outperform regular GNNs by incorporating time information into graph-based operations. However, TGNNs adopt specialized models (e.g., TGN, TGAT, and APAN ) and require tailored training frameworks (e.g., TGL and ETC). In this paper, we propose TF-TGN, which uses Transformer decoder as the backbone model for TGNN to enjoy Transformer's codebase for efficient training. In particular, Transformer achieves tremendous success for language modeling, and thus the community developed high-performance kernels (e.g., flash-attention and memory-efficient attention) and efficient distributed training schemes (e.g., PyTorch FSDP, DeepSpeed, and Megatron-LM). We observe that TGNN resembles language modeling, i.e., the message aggregation operation between chronologically occurring nodes and their temporal neighbors in TGNNs can be structured as sequence modeling. Beside this similarity, we also incorporate a series of algorithm designs including suffix infilling, temporal graph attention with self-loop, and causal masking self-attention to make TF-TGN work. During training, existing systems are slow in transforming the graph topology and conducting graph sampling. As such, we propose methods to parallelize the CSR format conversion and graph sampling. We also adapt Transformer codebase to train TF-TGN efficiently with multiple GPUs. We experiment with 9 graphs and compare with 2 state-of-the-art TGNN training frameworks. The results show that TF-TGN can accelerate training by over 2.20 while providing comparable or even superior accuracy to existing SOTA TGNNs. TF-TGN is available at https://github.com/qianghuangwhu/TF-TGN.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. When Speed meets Accuracy: an Efficient and Effective Graph Model for Temporal Link Prediction

    cs.AI 2025-07 conditional novelty 6.0 of 10

    EAGLE predicts temporal links with top-k recent neighbors plus top-k shared temporal PageRank influencers, matching or beating transformer T-GNNs while running far faster.

  2. GTIN: A Unified Framework for Joint Event and Time Prediction in Temporal Graphs

    cs.SI 2026-07 conditional novelty 4.0 of 10

    An edge-centric message-passing model jointly scores next-event location, Gaussian time, and features from a global temporal graph embedding and beats listed baselines on time RMSE.

Pith tools