In large-scale e-commerce platforms, Conversion Rate (CVR) prediction is crucial for recommender system, yet existing approach face a fundamental granularity mismatch: models operate at the item level while users purchase at the fine-grained Stock Keeping Unit (SKU) level. This mismatch causes loss of fine-grained user intent signals. Moreover, it also introduces price inconsistency bias due to the gap between static exposure prices and actual transaction prices. While direct SKU-level modeling would resolve these issues, it is impractical for industrial deployment due to the extreme data sparsity and prohibitive inference costs.

Click-Through Rate (CTR) prediction models enable users to discover matched items in recommender systems. Industrial-scale models typically adopt a unified embedding approach for both hot and cold items. However, existing embedding-based models exhibit limitations in representation learning for cold items due to sparse historical user interactions. In this paper, we propose a Meta-Adaptive Network for Effective Cold-Start Recommendation (MANE). Inspired by meta-learning, we develop a lightweight plug-and-play meta-learner that generates enhanced representations to model full-lifecycle representations for cold items. Our meta-network dynamically adjusts the contribution of generalized features in final representations as item exposure increases, enabling adaptive balancing between generalization and specificity for cold items. In addition, certain high-potential items in cold-start scenarios face challenges in effective exposure due to limited interaction signals. Therefore, we further propose a novel representation learning method that incorporates a warm-aware contrastive loss, which aligns the representations of cold items with those of hot items exhibiting high multimodal similarity. Experimental results on the Taobao production dataset and online A/B testing validate the effectiveness of our method, achieving 4.34% item page views (IPV) and 2.84% CTR improvement.

Maintaining a healthy ecosystem in billion-scale online platforms is challenging, as users naturally gravitate toward popular items, leaving cold and less-explored items behind. This ''rich-get-richer'' phenomenon hinders the growth of potentially valuable cold items and harms the platform's ecosystem. Existing cold-start models primarily focus on improving initial recommendation performance for cold items but fail to address users' natural preference for popular content. In this paper, we introduce AliBoost, Alibaba's ecological boosting framework, designed to complement user-oriented natural recommendations and foster a healthier ecosystem. AliBoost incorporates a tiered boosting structure and boosting principles to ensure high-potential items quickly gain exposure while minimizing disruption to low-potential items. To achieve this, we propose the Stacking Fine-Tuning Cold Predictor to enhance the foundation CTR model's performance on cold items for accurate CTR and potential prediction. AliBoost then employs an Item-oriented Bidding Boosting mechanism to deliver cold items to the most suitable users while balancing boosting speed with user-personalized preferences. Over the past six months, AliBoost has been deployed across Alibaba's mainstream platforms, successfully cold-starting over a billion new items and increasing both clicks and GMV of cold items by over 60% within 180 days. Extensive online analysis and A/B testing demonstrate the effectiveness of AliBoost in addressing ecological challenges, offering new insights into the design of billion-scale recommender systems.

Online Food Ordering Service (OFOS) is a popular location-based service that helps people order what they want. Compared with traditional e-commerce recommendation systems, users’ interests may be diverse under different spatiotemporal contexts, leading to various spatiotemporal data distributions, which increases the difficulty of model learning. However, numerous current works simply mix all samples to train a set of model parameters, which makes it challenging to capture the diversity in different spatiotemporal contexts. Therefore, we address this challenge by proposing a Bottom-up Adaptive Spatiotemporal Model(BASM) to adaptively fit the spatiotemporal data distribution, further improving the fitting capability of the model. Specifically, a spatiotemporal-aware embedding layer performs weight adaptation on field granularity in feature embedding to achieve the purpose of dynamically perceiving spatiotemporal contexts. Meanwhile, we propose a spatiotemporal semantic transformation layer to explicitly convert the concatenated input of the raw semantic to the spatiotemporal semantic, which can further enhance the semantic representation under different spatiotemporal contexts. Furthermore, we introduce a novel spatiotemporal adaptive bias tower to capture diverse spatiotemporal bias, reducing the difficulty of modeling spatiotemporal distinction. To further verify the effectiveness of BASM, we propose two new metrics, Time-period-wise AUC (TAUC) and City-wise AUC (CAUC). Extensive offline evaluations on public and industrial datasets are conducted to demonstrate the effectiveness of our proposed model. The online A/B experiment also further illustrates the practicability of the model online service. This proposed method has now been implemented on Ele.me, a major online food ordering platform in China, serving more than 100 million online users.