POI Recommendation with LLM-Augmented Multi-Graph Learning and Contrastive Alignment

Making Collaborative Signals Count: Graph-Aware Large Language Models for Sequential Recommendation

GALLM constructs a collaborative graph over text tokens and item tokens, and models three types of relations that are transformed into lightweight learnable attention biases and incorporated into the LLM attention mechanism, enabling collaborative-aware token interactions without introducing an additional graph encoder.

DIGEST: Dynamic Graph Refinement with Dual Contrastive Semantic Transfer for Multimodal Recommendation

Multimodal recommendation benefits from leveraging rich content signals such as images and texts to alleviate interaction sparsity, yet existing graph-based approaches are still hindered by (i) noisy user—item edges that are treated as static during training and (ii) inconsistent representation spaces across interaction-driven and modality-induced graph views. To address these issues, we propose DIGEST, a multi-graph framework that propagates trainable ID embeddings on a denoised user—item graph and a fused modality-induced item—item graph, and interleaves message passing with dynamic graph refinement that iteratively reweights existing edges to suppress noisy connections. To enable reliable semantic transfer across views, DIGEST further introduces a dual contrastive alignment that (i) aligns the collaborative and semantic item views and (ii) constrains the semantic graph representations to projected multimodal features, together with a lightweight dimension decorrelation regularizer and adaptive gated fusion to reduce redundancy and stabilize multi-view learning. Extensive experiments on three Amazon benchmark datasets demonstrate that DIGEST consistently outperforms state-of-the-art multimodal recommenders, achieving up to 8.43% relative improvement on NDCG@20 and 7.66% on Recall@20 over the strongest baselines.

TRWH: A Text-Driven Random Walk Heterogeneous GNN for Semantic-Aware Sparse Recommendation

Graph Neural Networks (GNNs) and Large Language Models (LLMs) have each advanced recommendation systems by modeling structural and semantic signals, respectively. However, integrating their complementary strengths remains challenging, particularly in sparse settings where maintaining semantic precision is critical. We propose TRWH (Text-driven Random Walk Heterogeneous Graph Neural Network), a novel framework that fuses LLM-generated textual profiles with heterogeneous graph structures through strategic random walk augmentation. TRWH consists of three core components: (1) Embedding Creation, which produces user and item representations using both Word2Vec and LLM-based profiling; (2) a Heterogeneous Graph Neural Network (HeteroGNN) that propagates information across multi-relational edges; and (3) Random Walk-based Path Construction, which enriches sparse graphs with second-order user-user and item-item links. Experiments on the Amazon-2023 Fashion (2M users, 825K items) and Beauty (631K users, 112K items) datasets demonstrate that TRWH achieves substantial performance gains over state-of-the-art methods, including 80.0% RMSE and 52.6% MAE reductions on Fashion, and 25.7% and 10.8% improvements on Beauty. Notably, while random walks improve performance with traditional embeddings, they can dilute the nuanced representations learned by LLMs, underscoring the importance of adaptive integration strategies.

G²PRO: Gradient-guided Graph Prompt Optimization for LLM-based POI Recommendation

Large Language Models (LLMs) have shown strong potential for sequential reasoning, creating new opportunities for next Point-of-Interest (POI) recommendation. However, applying LLMs to POI prediction remains challenging due to the modality gap between textual semantics and continuous spatio-temporal signals. Existing rule-based prompting methods often introduce redundant context when bridging this gap. To address this issue, we propose G2PRO, a collaborative framework that combines the structural perception of Graph Neural Networks (GNNs) with the reasoning capability of LLMs. Specifically, we construct a User-Behavior Spatio-Temporal Knowledge Graph (UST-KG) to capture POI relations and transition dynamics, and train a lightweight GNN-based Prompt Selector (GPS) to select informative POI nodes for prompt construction. We further introduce a gradient-guided positive prompt labeling strategy that estimates each POI's contribution to the target prediction through gradients over prompt embeddings, turning prompt selection into an optimizable learning objective rather than a hand-crafted heuristic. Experiments on four real-world datasets show that G2PRO consistently outperforms state-of-the-art traditional and LLM-based baselines. Ablation and breakdown studies further validate the effectiveness of each component and demonstrate the benefits of structure-aware, attribution-guided prompting for LLM-based POI recommendation.

LRGCL: LLM-refined graph contrastive learning for review-based recommendation

Review-based recommendations commonly construct user-item interaction graphs in which textual review semantics serve as edge attributes. However, two practical limitations persist. First, raw reviews are inherently noisy: they often contain personal anecdotes, emotional expressions, and redundant phrasing that interfere with the preference-relevant signal on each edge. Second, many widely used graph contrastive learning methods generate positive views through random node or edge dropping, which may inadvertently discard semantically critical interactions and degrade the quality of learned representations. Although recent neighbor-aware methods have begun to address this limitation from structural or latent-semantic perspectives, review-based recommendation still lacks an edge-level semantic-preserving contrastive strategy that explicitly uses review evidence to protect key user-item interactions. To tackle these challenges, we propose LRGCL, an LLM-Refined Graph Contrastive Learning model for personalized recommendation. LRGCL first employs a large language model to refine each raw review into a concise, aspect-focused text that retains only key preference-feature information, and then encodes the refined review to obtain the edge semantic representation of the user-item bipartite graph. LRGCL further introduces a semantic-preserving contrastive objective in which key interaction edges are identified and preserved in augmented graph views, while only non-critical edges are dropped. Results across multiple datasets confirm that LRGCL achieves the lowest MSE and MAE among strong rating-based and review-based baselines, with especially clear MSE gains and smaller MAE margins on some domains. Ablation studies further verify the effectiveness of LLM-based review refinement and graph contrastive learning.

DCGCE: dual-channel graph embedding for multimodal recommendation with a causal perspective