Abstract:Next Point-of-Interest (POI) recommendation is a critical task in location-based services that aim to provide personalized suggestions for the user’s next destination. Previous works on POI recommendation have laid focus on modeling the user’s spatial preference. However, existing works that leverage spatial information are only based on the aggregation of users’ previous visited positions, which discourages the model from recommending POIs in novel areas. This trait of position-based methods will harm the model’s performance in many situations. Additionally, incorporating sequential information into the user’s spatial preference remains a challenge. In this paper, we propose Diff-POI: a Diffusion-based model that samples the user’s spatial preference for the next POI recommendation. Inspired by the wide application of diffusion algorithm in sampling from distributions, Diff-POI encodes the user’s visiting sequence and spatial character with two tailor-designed graph encoding modules, followed by a diffusion-based sampling strategy to explore the user’s spatial visiting trends. We leverage the diffusion process and its reversed form to sample from the posterior distribution and optimized the corresponding score function. We design a joint training and inference framework to optimize and evaluate the proposed Diff-POI. Extensive experiments on four real-world POI recommendation datasets demonstrate the superiority of our Diff-POI over state-of-the-art baseline methods. Further ablation and parameter studies on Diff-POI reveal the functionality and effectiveness of the proposed diffusion-based sampling strategy for addressing the limitations of existing methods.

Positional Diffusion: Ordering Unordered Sets with Diffusion Probabilistic Models

LVDiffusor: Distilling Functional Rearrangement Priors from Large Models into Diffusor

PuzzleFusion: Unleashing the Power of Diffusion Models for Spatial Puzzle Solving

Beyond Autoregression: Discrete Diffusion for Complex Reasoning and Planning

LayoutDiffusion: Improving Graphic Layout Generation by Discrete Diffusion Probabilistic Models

Diffusion of Thoughts: Chain-of-Thought Reasoning in Diffusion Language Models

Non-rigid Relative Placement through 3D Dense Diffusion

SymmetricDiffusers: Learning Discrete Diffusion on Finite Symmetric Groups

Think While You Generate: Discrete Diffusion with Planned Denoising

Directed Diffusion: Direct Control of Object Placement through Attention Guidance

Diffusion Model is an Effective Planner and Data Synthesizer for Multi-Task Reinforcement Learning

Diffusion as Reasoning: Enhancing Object Goal Navigation with LLM-Biased Diffusion Model

DiffAssemble: A Unified Graph-Diffusion Model for 2D and 3D Reassembly

DiffPop: Plausibility-Guided Object Placement Diffusion for Image Composition

A Diffusion model for POI recommendation

Where to Diffuse, How to Diffuse, and How to Get Back: Automated Learning for Multivariate Diffusions

Pard: Permutation-Invariant Autoregressive Diffusion for Graph Generation

AffordDP: Generalizable Diffusion Policy with Transferable Affordance

ActionDiffusion: An Action-aware Diffusion Model for Procedure Planning in Instructional Videos

Diffused Task-Agnostic Milestone Planner