Skip to content
#diffusion models Book Open access

Give the Long-tail More SPACE: Promoting Provider Fairness in Next POI Recommendation

Aug 2026 · Proceedings of the 20th ACM Conference on Recommender Systems · pp. 281-290 · 0 citations · 49 references
Computer Science

TL;DR

SPACE (Supply- and Physics-Aware Conditional Embedding generation), a model-agnostic framework that improves long-tail POI exposure via virtual user generation under explicit feasibility and supply control, is proposed.

Abstract

Next point-of-interest (POI) recommendation predicts users’ future destinations from historical mobility sequences and has become a key component of location-based services. However, mainstream models often concentrate exposure on a small set of popular POIs, leaving long-tail merchants systematically under-exposed. While provider fairness has recently attracted increasing attention, directly applying existing provider-fairness techniques to POI recommendation is problematic: (i) users face execution constraints; and (ii) POIs face resource supply constraints. To address this, we propose SPACE (Supply- and Physics-Aware Conditional Embedding generation), a model-agnostic framework that improves long-tail POI exposure via virtual user generation under explicit feasibility and supply control. SPACE consists of three stages: (1) community inference to capture heterogeneous user execution constraints; (2) unbalanced optimal-transport allocation to decide how many virtual users each tail POI should receive from which communities under POI-specific supply budgets; and (3) constraint-guided latent diffusion to generate POI-conditional, community-consistent virtual user embeddings. The generated user–POI pairs can be seamlessly used to train existing recommenders without modifying their architectures. Extensive experiments on three real-world datasets demonstrate that SPACE substantially improves provider fairness while maintaining—and often improving—recommendation accuracy across multiple backbone models. Our code is publicly available at https://github.com/Anniran1/SPACE-main.

Read PDF

Similar papers

Preprint Aug 2026

Toward Equitable Low-Carbon Mobility: Fairness-Aware Demand Prediction for Expanding Bike-Sharing Systems

Bike-sharing systems are an important component of low-carbon urban mobility, but continued expansion creates challenges in both cold-start prediction and equitable resource allocation. Newly deployed stations lack historical ridership records, causing a mismatch between training and inference for graph-based models on...

Man Luo, Yi-Xuan Zhao · 0 citations
Sep 2026

Cost-based distributionally robust points-of-interest recommendation with demand uncertainty

A novel distributionally robust model for the Operator-oriented Point-of-interest Recommendation Problem (OO-POIRP) with demand uncertainty is proposed, aiming to assist operators in minimizing cooperation cost, ticket booking cost, promotion cost, and transportation cost.

Ze-Xin Li, Zun-Hao Luo, Qiu-Lin Li et al. · 0 citations

UniPolicy: Unified Objective-Specific Policies for Generative Search Advertising

Search advertising connects user intent with commercial content and plays a critical role in platform monetization. Recent systems typically align pretrained generative models with a single business reward, such as eCPM, or use naive reward fusion for preliminary multi-objective alignment. However, an ideal search adve...

Kun Yao, Yu-Hang Zhou, Yi-Chi Zhang et al. · 0 citations
#artificial intelligence Preprint Aug 2026

An Empirical Evaluation of Cross-City POI Recommendation on a Large-Scale Benchmark

The need for task-specific designs that support cross-city preference transfer, semantic grounding, and scalable reasoning over unseen destination inventories is highlighted, with results highlighting the need for task-specific designs that support cross-city preference transfer, semantic grounding, and scalable reason...

Pei-Bo Li, Yang Song, Hao Xue et al. · 0 citations
Book Open access Aug 2026

Is the Last Check-In All You Need? Next POI Recommendation: Recall and Rerank

K1-POI is proposed, a concise recall-and-rerank framework that formulates sparse next-POI prediction as candidate generation and calibration and an efficient Spatio-Temporal Prior Reranker (ST-Reranker) that calibrates the ordering within the top-K candidates using additive, model-agnostic spatio-temporal priors comput...

Zhengjia Xu, Ding Lyu, Zitai Qiu et al. · 0 citations
Open access Sep 2026

A Multi-Objective Optimization Framework for Intelligent Volunteer Matching

Volunteer matching is not well described by conventional one-sided recommendation because effective allocation depends on reciprocal suitability, fairness, interpretability, and sustained participation. Existing studies usually address these objectives separately, and public benchmarks rarely support their joint evalua...

Wen-Na Liu, Xin Zhang, Bai-Yi Liu et al. · 0 citations

Related blog posts

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.