Metraipolis › archive › dc_2818142fee53

POI Recommendation with LLM-Augmented Multi-Graph Learning and Contrastive Alignment — arXiv:2608.16407

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.16407
License (as archived)
CC0-1.0
Archived
2026-10-18T19:44:24.750Z
SHA-256 of the archived text
43c15e109d0ba2794ba6aa3e39f1260ff30f58dca1d6330bc2371754e36adbba
Stable URL of this page
https://metraipolis.com/documento/dc_2818142fee53

How to cite

POI Recommendation with LLM-Augmented Multi-Graph Learning and Contrastive Alignment — arXiv:2608.16407 — arXiv (en), https://arxiv.org/abs/2608.16407 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2818142fee53 (Metraipolis archive dc_2818142fee53, sha256 43c15e109d0ba2794ba6aa3e39f1260ff30f58dca1d6330bc2371754e36adbba)
BibTeX
@misc{dc2818142fee53,
  title = {POI Recommendation with LLM-Augmented Multi-Graph Learning and Contrastive Alignment — arXiv:2608.16407},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.16407},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2818142fee53; sha256 43c15e109d0ba2794ba6aa3e39f1260ff30f58dca1d6330bc2371754e36adbba},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.16407. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Point-of-interest (POI) recommendation models based on graph neural networks achieve strong performance by propagating collaborative signals over user-item interactions, yet they struggle with the cold-start problem, where items with few or no interactions are not represented. In this paper, we propose LLM-augmented Multi-Graph Contrastive Learning (LLM-MGCL), a multi-graph neural network that uses semantic and spatial information about items to extend the LightGCN backbone with two auxiliary item-item graphs: a semantic graph constructed from sentence embeddings of LLM-generated photo summaries and keywords, and a geographic graph derived from Haversine distances between business locations. Item embeddings are propagated over all three graphs in parallel, fused additively, and aligned across views through a bidirectional InfoNCE contrastive objective that connects behavioral, semantic, and spatial representations of the same items. Experiments on the Yelp Multimodal Recommendation Dataset show that LLM-MGCL outperforms classical collaborative filtering, matrix factorization, and interaction-only graph neural network baselines. It improves Recall@20 by 52.0% and NDCG@20 by 64.8% over LightGCN while performing on par with the strongest contrastive baseline, Self-supervised Graph Learning (SGL), which is also affected by the cold-start problem. An ablation study reveals that the cross-view contrastive alignment (CA) is the primary driver of these gains, with the best performance achieved when all three graphs are combined. Our results suggest that externally grounded, LLM-derived item knowledge can effectively compensate for missing collaborative signal and mitigate the item cold-start problem in POI recommendation.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.LG).
Categorie arXiv: Information Retrieval (cs.IR) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.16407 — 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 2026-10-18T19:02:44.310Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).