CaST-POI: Candidate-Conditioned Spatiotemporal Modeling for Next POI Recommendation

• URL: http://arxiv.org/abs/2604.20845v1
• Date: Tue, 10 Feb 2026 07:10:18 GMT
• Title: CaST-POI: Candidate-Conditioned Spatiotemporal Modeling for Next POI Recommendation
• Abstract summary: Next Point-of-Interest (POI) recommendation plays a crucial role in location-based services by predicting users' future mobility patterns.<n>Existing methods typically compute a single user representation from historical trajectories and use it to score all candidate POIs uniformly.<n>We propose CaST-POI, a candidate-conditioned model for next POI recommendation.
• Score: 18.263381277536357
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Next Point-of-Interest (POI) recommendation plays a crucial role in location-based services by predicting users' future mobility patterns. Existing methods typically compute a single user representation from historical trajectories and use it to score all candidate POIs uniformly. However, this candidate-agnostic paradigm overlooks that the relevance of historical visits inherently depends on which candidate is being evaluated. In this paper, we propose CaST-POI, a candidate-conditioned spatiotemporal model for next POI recommendation. Our key insight is that the same user history should be interpreted differently when evaluating different candidate POIs. CaST-POI employs a candidate-conditioned sequence reader that uses candidates as queries to dynamically attend to user history. In addition, we introduce candidate-relative temporal and spatial biases to capture fine-grained mobility patterns based on the relationships between historical visits and each candidate POI. Extensive experiments on three benchmark datasets demonstrate that CaST-POI consistently outperforms state-of-the-art methods, yielding substantial improvements across multiple evaluation metrics, with particularly strong advantages under large candidate pools. Code is available at https://github.com/YuZhenyuLindy/CaST-POI.git.

Related papers

• Which Histories Matter for Time Series Forecasting? Learning Predictive Relevance with Future Supervision [0.0]
  A normalized-pattern retriever learns a listwise future-compatibility target while keeping inference-time scoring strictly past-only.<n>Our method retains similarity-based candidate generation but reranks its candidates by a more predictive relevance criterion.<n>Across six benchmarks, the reranker improves Pattern retrieval while revealing candidate-global, query-specific, and mixed relevance regimes.
  arXiv  Detail & Related papers  (2026-08-24T13:15:29Z)
• HARP: Hierarchical Adaptive Ranking with Preference-Adaptive Fusion for Query-Based CVE Prioritization [84.9738509922482]
  We propose a graph-grounded multi-view framework that ranks candidates from a natural-language query together with a support bank of historical labeled examples.<n>HarP retrieves evidence from a vulnerability knowledge graph, scores candidates with policy-conditioned global, enterprise, and user views, and fits view-fusion weights from sampled supports.
  arXiv  Detail & Related papers  (2026-08-19T20:29:00Z)
• Think2Go: Generative Next POI Recommendation with LLM Reasoning [57.37706163990998]
  Next Point-of-Interest (POI) recommendation task focuses on mining user behavioral preference patterns from historical check-ins to provide personalized suggestions for the next destination.<n>Existing methods rely on shallow contextual information and handcrafted feature interactions to predict the next POI.<n>We propose Think2Go, a novel generative next POI recommendation framework, which enhances the model's comprehension of SID representations and explores diverse spatial-temporal patterns via test-time computational scaling.
  arXiv  Detail & Related papers  (2026-07-31T03:52:20Z)
• GenCI: Generative Modeling of User Interest Shift via Cohort-based Intent Learning for CTR Prediction [84.0125708499372]
  We propose a generative user intent framework to model user preferences for click-through rate (CTR) prediction.<n>The framework first employs a generative model, trained with a next-item prediction objective, to proactively produce candidate interest cohorts.<n>A hierarchical candidate-aware network then injects this rich contextual signal into the ranking stage, refining them with cross-attention to align with both user history and the target item.
  arXiv  Detail & Related papers  (2026-01-26T08:15:04Z)
• Next Point-of-interest (POI) Recommendation Model Based on Multi-modal Spatio-temporal Context Feature Embedding [29.460520413964243]
  This paper introduces user-term preference information and keytemporal context information, and proposes a POI recommendation model based on multimodal context feature embedding.<n>The results show that the POI prediction model combining multiple types of features has higher prediction accuracy than existing SOTA models and methods.
  arXiv  Detail & Related papers  (2025-08-29T14:33:56Z)
• Enhancing POI Recommendation through Global Graph Disentanglement with POI Weighted Module [4.736059095502583]
  Next point of interest (POI) recommendation primarily predicts future activities based on users' past check-in data and current status.<n>This paper proposes a novel next POI recommendation framework called Graph Disentangler with POI Weighted Module (GDPW)
  arXiv  Detail & Related papers  (2025-07-19T13:16:44Z)
• Bridging SFT and DPO for Diffusion Model Alignment with Self-Sampling Preference Optimization [67.8738082040299]
  Self-Sampling Preference Optimization (SSPO) is a new alignment method for post-training reinforcement learning.<n>SSPO eliminates the need for paired data and reward models while retaining the training stability of SFT.<n>SSPO surpasses all previous approaches on the text-to-image benchmarks and demonstrates outstanding performance on the text-to-video benchmarks.
  arXiv  Detail & Related papers  (2024-10-07T17:56:53Z)
• SA-LSPL:Sequence-Aware Long- and Short- Term Preference Learning for next POI recommendation [19.40796508546581]
  Point of Interest (POI) recommendation aims to recommend the POI for users at a specific time. We propose a novel approach called Sequence-Aware Long- and Short-Term Preference Learning (SA-LSPL) for next-POI recommendation.
  arXiv  Detail & Related papers  (2024-03-30T13:40:25Z)
• Bayes-enhanced Multi-view Attention Networks for Robust POI Recommendation [81.4999547454189]
  Existing works assume the available POI check-ins reported by users are the ground-truth depiction of user behaviors. In real application scenarios, the check-in data can be rather unreliable due to both subjective and objective causes. We propose a Bayes-enhanced Multi-view Attention Network to address the uncertainty factors of the user check-ins.
  arXiv  Detail & Related papers  (2023-11-01T12:47:38Z)
• Recommender Systems with Generative Retrieval [58.454606442670034]
  We propose a novel generative retrieval approach, where the retrieval model autoregressively decodes the identifiers of the target candidates. To that end, we create semantically meaningful of codewords to serve as a Semantic ID for each item. We show that recommender systems trained with the proposed paradigm significantly outperform the current SOTA models on various datasets.
  arXiv  Detail & Related papers  (2023-05-08T21:48:17Z)
• Self-supervised Graph-based Point-of-interest Recommendation [66.58064122520747]
  Next Point-of-Interest (POI) recommendation has become a prominent component in location-based e-commerce. We propose a Self-supervised Graph-enhanced POI Recommender (S2GRec) for next POI recommendation. In particular, we devise a novel Graph-enhanced Self-attentive layer to incorporate the collaborative signals from both global transition graph and local trajectory graphs.
  arXiv  Detail & Related papers  (2022-10-22T17:29:34Z)
• Exploiting Bi-directional Global Transition Patterns and Personal Preferences for Missing POI Category Identification [37.025295828186955]
  We propose a novel neural network approach to identify the missing POI categories. Specifically, we design an attention matching cell to model how well the check-in category information matches their non-personal transition patterns and personal preferences. Our model can be naturally extended to address next POI category recommendation and prediction tasks with competitive performance.
  arXiv  Detail & Related papers  (2021-12-31T04:15:37Z)
• Modelling of Bi-directional Spatio-Temporal Dependence and Users' Dynamic Preferences for Missing POI Check-in Identification [38.51964956686177]
  We develop a model, named Bi-STDDP, which can integrate bi-directional-temporal dependence and users' dynamic preferences. Results demonstrate significant improvements of our model compared with state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-12-31T03:54:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.