Point-of-Interest (POI) recommendation plays a vital role in various location-aware services. It has been observed that POI recommendation is driven by both sequential and geographical influences. However, since there is no annotated label of the dominant influence during recommendation, existing methods tend to entangle these two influences, which may lead to sub-optimal recommendation performance and poor interpretability. In this paper, we address the above challenge by proposing DisenPOI, a novel Disentangled dual-graph framework for POI recommendation, which jointly utilizes sequential and geographical relationships on two separate graphs and disentangles the two influences with self-supervision. The key novelty of our model compared with existing approaches is to extract disentangled representations of both sequential and geographical influences with contrastive learning. To be specific, we construct a geographical graph and a sequential graph based on the check-in sequence of a user. We tailor their propagation schemes to become sequence-/geo-aware to better capture the corresponding influences. Preference proxies are extracted from check-in sequence as pseudo labels for the two influences, which supervise the disentanglement via a contrastive loss. Extensive experiments on three datasets demonstrate the superiority of the proposed model.
Click-Through Rate (CTR) prediction, estimating the probability of a user clicking on an item, is essential in industrial applications, such as online advertising. Many works focus on user behavior modeling to improve CTR prediction performance. However, most of those methods only model users' positive interests from users' click items while ignoring the context information, which is the display items around the clicks, resulting in inferior performance. In this paper, we highlight the importance of context information on user behavior modeling and propose a novel model named Deep Context Interest Network (DCIN), which integrally models the click and its display context to learn users' context-aware interests. DCIN consists of three key modules: 1) Position-aware Context Aggregation Module (PCAM), which performs aggregation of display items with an attention mechanism; 2) Feedback-Context Fusion Module (FCFM), which fuses the representation of clicks and display contexts through non-linear feature interaction; 3) Interest Matching Module (IMM), which activates interests related with the target item. Moreover, we provide our hands-on solution to implement DCIN on large-scale industrial systems. The significant improvements in both offline and online evaluations demonstrate the superiority of our proposed DCIN method. Notably, DCIN has been deployed on our online advertising system serving the main traffic, which brings 1.5% CTR and 1.5% RPM lift.
As an emerging privacy-preserving approach to leveraging cross-platform user interactions, vertical federated learning (VFL) has been increasingly applied in recommender systems. However, vanilla VFL is only applicable to overlapped users, ignoring potential universal interest patterns hidden among non-overlapped users and suffers from limited user group benefits, which hinders its application in real-world recommenders. In this paper, we extend the traditional vertical federated recommendation problem (VFR) to a more realistic Fully-Vertical federated recommendation setting (Fully-VFR) which aims to utilize all available data and serve full user groups. To tackle challenges in implementing Fully-VFR, we propose a Retrieval-enhanced Vertical Federated recommender (ReFer), a groundbreaking initiative that explores retrieval-enhanced machine learning approaches in VFL. Specifically, we establish a general "retrieval-and-utilization" algorithm to enhance the quality of representations across all parties. We design a flexible federated retrieval augmentation (RA) mechanism for VFL: (i) Cross-RA to complement field missing and (ii) Local-RA to promote mutual understanding between user groups. We conduct extensive experiments on both public and industry datasets. Results on both sequential and non-sequential CTR prediction tasks demonstrate that our method achieves significant performance improvements over baselines and is beneficial for all user groups.
Despite the recent significant advancements in poster layout generation, existing works are mainly unaware of the given design elements (i.e., text, logo, and underlay), which leads to undesirable layouts or visual artifacts. The visual artifacts we refer to include (1) improper sizes, e.g., placing a short piece of text into a large textbox or long texts into small text boxes, and (2) image distortion, e.g., the stretched logo in Fig. 1. To advance research in this field, we propose a new design-element aware poster layout generation task, which require the generated layouts to not only have harmonic relationships but also fit well with the design elements. To address this task, we propose Design Element aware Transformer (DET), an encoder-decoder based transformer network, to generate reasonable layouts that fit not only the background images but also the design elements. The encoder extracts a fine-grained multi-scale representation from the background image and its saliency map. The decoder receives the background features and produces layouts conditioned on the content and desired sizes of the design elements. Adopting the multi-scale representation and the deformable attention in both the encoder and decoder enables our method to accurately understand/generate the spatial relationships between the background objects and design elements. We adapted three public poster layout generation datasets to fit our task and conducted experiments on them. In the meantime, we propose a new evaluation metric called AspDiff to measure whether the generated layout matches the given design elements. Quantitative and qualitative evaluation on three datasets demonstrates that DET yields superior results compared to other layout generation methods. Our code and datasets will be released.