Metraipolis › archive › dc_8bbb0fefd2f5
Rethinking Message Passing as Retrieval for Text-Attributed Graph Learning — arXiv:2608.26732
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.26732
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T14:28:01.360Z
- SHA-256 of the archived text
a4b36ca7fb2c5c59af3a6f6b36a3f57a0abe8a08f2b7d764178d8a1037051f07- Stable URL of this page
https://metraipolis.com/documento/dc_8bbb0fefd2f5
How to cite
Rethinking Message Passing as Retrieval for Text-Attributed Graph Learning — arXiv:2608.26732 — arXiv (en), https://arxiv.org/abs/2608.26732 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8bbb0fefd2f5 (Metraipolis archive dc_8bbb0fefd2f5, sha256 a4b36ca7fb2c5c59af3a6f6b36a3f57a0abe8a08f2b7d764178d8a1037051f07)
BibTeX
@misc{dc8bbb0fefd2f5,
title = {Rethinking Message Passing as Retrieval for Text-Attributed Graph Learning — arXiv:2608.26732},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.26732},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8bbb0fefd2f5; sha256 a4b36ca7fb2c5c59af3a6f6b36a3f57a0abe8a08f2b7d764178d8a1037051f07},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26732. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Graph neural networks (GNNs) are typically conceptualized as message-passing neural networks, yet it remains unclear why neighborhood aggregation reliably outperforms node-wise multilayer perceptrons (MLPs). Despite its empirical success, this paradigm can be computationally expensive and sensitive to imperfect graph structures. In this work, we present a retrieval-augmented view of GNNs: each layer makes predictions by applying an MLP to a node representation together with a permutation-invariant summary of retrieved graph context. Motivated by this perspective, we propose RTA, a simple MLP-based framework that replaces structural message passing with label-aware retrieval and propagation. We provide theoretical insights that (i) connect retrieval-based aggregation to softmax-attention message passing, and (ii) establish the robustness of retrieved-context supervision to mis-retrieved outliers. Experiments on multiple text-attributed graph benchmarks show that RTA matches or even outperforms strong GNN and graph LLM baselines while improving efficiency and robustness across diverse scenarios. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.AI). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.26732 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T13:16:53.658Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).