SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2409.17986 · 2024

Supra-Laplacian Encoding for Transformer on Dynamic Graphs

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 6 functions out of this paper's own repositories and ran 6 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
ykrmm/slate canonical 6 of 6
FunctionStatusWhere it lives
edge_index_to_adj_matrix Ran ykrmm/slate/slate/lib/utils.py
code served (permissive licence) · get_code("24c179d1051d65ff")
feed_dict_to_device Ran ykrmm/slate/slate/lib/utils.py
code served (permissive licence) · get_code("97a9edbc9b184cc6")
g_to_device Ran ykrmm/slate/slate/lib/utils.py
code served (permissive licence) · get_code("023f8488b000df51")
graphs_to_supra Ran ykrmm/slate/slate/lib/supra.py
code served (permissive licence) · get_code("2de4e2cc94d2b971")
pad_with_last_val Ran ykrmm/slate/slate/models/egcn_h.py
code served (permissive licence) · get_code("b4a1a41169b57e33")
reindex_edge_index Ran ykrmm/slate/slate/lib/supra.py
code served (permissive licence) · get_code("2378ee541055b5be")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Fully connected Graph Transformers (GT) have rapidly become prominent in the static graph community as an alternative to Message-Passing models, which suffer from a lack of expressivity, oversquashing, and under-reaching. However, in a dynamic context, by interconnecting all nodes at multiple snapshots with self-attention, GT loose both structural and temporal information. In this work, we introduce Supra-LAplacian encoding for spatio-temporal TransformErs (SLATE), a new spatio-temporal encoding to leverage the GT architecture while keeping spatio-temporal information. Specifically, we transform Discrete Time Dynamic Graphs into multi-layer graphs and take advantage of the spectral properties of their associated supra-Laplacian matrix. Our second contribution explicitly model nodes' pairwise relationships with a cross-attention mechanism, providing an accurate edge representation for dynamic link prediction. SLATE outperforms numerous state-of-the-art methods based on Message-Passing Graph Neural Networks combined with recurrent models (e.g LSTM), and Dynamic Graph Transformers, on 9 datasets. Code is available at: github.com/ykrmm/SLATE.

For agents

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

get_harvested_code_for_paper("2409.17986")
get_code_for_paper("2409.17986")
have("2409.17986")

Connect an agent — have() is free.