A Plug-and-Play Spatially-Constrained Representation Enhancement Framework for Local-Life Recommendation

TL;DR

This work proposes ReST, a Plug-And-Play Spatially-Constrained Representation Enhancement Framework for Long-Tail Local-Life Recommendation, and proposes a novel Spatially-Constrained ID Representation Enhancement Network (SIDENet) based on contrastive learning, which incorporates two efficient strategies.

Abstract

Local-life recommendation have witnessed rapid growth, providing users with convenient access to daily essentials. However, this domain faces two key challenges: (1) spatial constraints, driven by the requirements of the local-life scenario, where items are usually shown only to users within a limited geographic area, indirectly reducing their exposure probability; and (2) long-tail sparsity, where few popular items dominate user interactions, while many high-quality long-tail items are largely overlooked due to imbalanced interaction opportunities. Existing methods typically adopt a user-centric perspective, such as modeling spatial user preferences or enhancing long-tail representations with collaborative filtering signals. However, we argue that an item-centric perspective is more suitable for this domain, focusing on enhancing long-tail items representation that align with the spatially-constrained characteristics of local lifestyle services. To tackle this issue, we propose ReST, a Plug-And-Play Spatially-Constrained Representation Enhancement Framework for Long-Tail Local-Life Recommendation. Specifically, we first introduce a Meta ID Warm-up Network, which initializes fundamental ID representations by injecting their basic attribute-level semantic information. Subsequently, we propose a novel Spatially-Constrained ID Representation Enhancement Network (SIDENet) based on contrastive learning, which incorporates two efficient strategies: a spatially-constrained hard sampling strategy and a dynamic representation alignment strategy. This design adaptively identifies weak ID representations based on their attribute-level information during training. It additionally enhances them by capturing latent item relationships within the spatially-constrained characteristics of local lifestyle services, while preserving compatibility with popular items.

Figures (5)

• Figure 1: Illustration of local lifestyle recommendation: spatial constraints limit user-accessible candidates, leading to inaccurate predictions for disadvantaged items and causing the serious Matthew Effect.
• Figure 2: This figure illustrates the overall structure and key components of the ReST model. Specifically:(a) shows the architecture of ReST, where the Basic Recommendation Tower can be any recommendation model. In this paper, we adopt DIN as the ranking backbone. (b) depicts the core concept of spatially-constrained sampling, where prior knowledge is leveraged to retrieve semantically similar items, and a spatially-constrained hard sampling strategy is applied. For instance, “Burger Restaurants” located in Beijing and Shanghai should not be considered similar due to differences in their target user location and competing item environments. (c) presents the dynamic representation alignment strategy, which adaptively determines the enhancement strength for different ID representations.
• Figure 3: Ablation study on ReST.
• Figure 4: Performance comparison with different numbers of hard negative samples.
• Figure 5: Performance improvement in orders and GMV across item popularity: long-tail items show greater relative gains due to naturally lower baseline GMV.