Large language models (LLMs) have exhibited remarkable capabilities and achieved significant breakthroughs across various domains, leading to their widespread adoption in recent years. Building on this progress, we investigate their potential in the realm of local life services. In this study, we establish a comprehensive benchmark and systematically evaluate the performance of diverse LLMs across a wide range of tasks relevant to local life services. To further enhance their effectiveness, we explore two key approaches: model fine-tuning and agent-based workflows. Our findings reveal that even a relatively compact 7B model can attain performance levels comparable to a much larger 72B model, effectively balancing inference cost and model capability. This optimization greatly enhances the feasibility and efficiency of deploying LLMs in real-world online services, making them more practical and accessible for local life applications. Available resources are at https://github.com/tsinghua-fib-lab/LocalEval.

Recommender systems that learn from implicit feedback often use large volumes of a single type of implicit user feedback, such as clicks, to enhance the prediction of sparse target behavior such as purchases. Using multiple types of implicit user feedback for such target behavior prediction purposes is still an open question. Existing studies that attempted to learn from multiple types of user behavior often fail to: (i) learn universal and accurate user preferences from different behavioral data distributions, and (ii) overcome the noise and bias in observed implicit user feedback.

Learning reinforcement learning (RL)-based recommenders from historical user-item interaction sequences is vital to generate high-reward recommendations and improve long-term cumulative benefits. However, existing RL recommendation methods encounter difficulties (i) to estimate the value functions for states which are not contained in the offline training data, and (ii) to learn effective state representations from user implicit feedback due to the lack of contrastive signals.

Modern recommender systems are trained to predict users’ potential future interactions from users’ historical behavior data. During the interaction process, despite the data coming from the user side, recommender systems also generate exposure data to provide users with personalized recommendation slates. Compared with the sparse user behavior data, the system exposure data are much larger in volume since only very few exposed items would be clicked by the user. In addition, user historical behavior data are privacy sensitive and commonly protected with careful access authorization. However, the large volume of recommender exposure data generated by the service provider itself usually receives less attention and could be accessed within a relatively larger scope of various information seekers or even potential adversaries. In this article, we investigate the problem of user behavior data leakage in the field of recommender systems. We show that the privacy-sensitive user past behavior data can be inferred through the modeling of system exposure. In other words, one can infer which items the user has clicked just from the observation of current system exposure for this user . Given the fact that system exposure data could be widely accessed from a relatively larger scope, we believe that user past behavior privacy has a high risk of leakage in recommender systems. More precisely, we conduct an attack model whose input is the current recommended item slate (i.e., system exposure) for the user while the output is the user’s historical behavior. Specifically, we exploit an encoder-decoder structure to construct the attack model and apply different encoding and decoding strategies to verify attack performance. Experimental results on two real-world datasets indicate a great danger of user behavior data leakage. To address the risk, we propose a two-stage privacy-protection mechanism that first selects a subset of items from the exposure slate and then replaces the selected items with uniform or popularity-based exposure. Experimental evaluation reveals a trade-off effect between the recommendation accuracy and the privacy disclosure risk, which is an interesting and important topic for privacy concerns in recommender systems.

In online information systems, users make decisions based on factors of several specific aspects, such as brand, price, etc. Existing recommendation engines ignore the explicit modeling of these factors, leading to sub-optimal recommendation performance. In this paper, we focus on the real-world scenario where these factors can be explicitly captured (the users are exposed with decision factor-based persuasion texts, i.e., persuasion factors). Although it allows us for explicit modeling of user-decision process, there are critical challenges including the persuasion factor's representation learning and effect estimation, along with the data-sparsity problem. To address them, in this work, we present our POEM (short for Persuasion factOr Effect Modeling) system. We first propose the persuasion-factor graph convolutional layers for encoding and learning representations from the persuasion-aware interaction data. Then we develop a prediction layer that fully considers the user sensitivity to the persuasion factors. Finally, to address the data-sparsity issue, we propose a counterfactual learning-based data augmentation method to enhance the supervision signal. Real-world experiments demonstrate the effectiveness of our proposed framework of modeling the effect of persuasion factors.

Optimal store placement aims to identify the optimal location for a new brick-and-mortar store that can maximize its sale by analyzing and mining users’ preferences from large-scale urban data. In recent years, the expansion of chain enterprises in new cities brings some challenges because of two aspects: (1) data scarcity in new cities, so most existing models tend to not work (i.e., overfitting), because the superior performance of these works is conditioned on large-scale training samples; (2) data distribution discrepancy among different cities, so knowledge learned from other cities cannot be utilized directly in new cities. In this article, we propose a task-adaptative model-agnostic meta-learning framework, namely, MetaStore, to tackle these two challenges and improve the prediction performance in new cities with insufficient data for optimal store placement, by transferring prior knowledge learned from multiple data-rich cities. Specifically, we develop a task-adaptative meta-learning algorithm to learn city-specific prior initializations from multiple cities, which is capable of handling the multimodal data distribution and accelerating the adaptation in new cities compared to other methods. In addition, we design an effective learning strategy for MetaStore to promote faster convergence and optimization by sampling high-quality data for each training batch in view of noisy data in practical applications. The extensive experimental results demonstrate that our proposed method leads to state-of-the-art performance compared with various baselines.