HiGR: Efficient Generative Slate Recommendation via Hierarchical Planning and Multi-Objective Preference Alignment
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910030930903040 |
|---|---|
| author | Pang, Yunsheng Liu, Zijian Li, Yudong Zhu, Shaojie Luo, Zijian Yu, Chenyun Wu, Sikai Shen, Shichen Xu, Cong Wang, Bin Jiang, Kai Yu, Hongyong Zhuo, Chengxiang Li, Zang |
| author_facet | Pang, Yunsheng Liu, Zijian Li, Yudong Zhu, Shaojie Luo, Zijian Yu, Chenyun Wu, Sikai Shen, Shichen Xu, Cong Wang, Bin Jiang, Kai Yu, Hongyong Zhuo, Chengxiang Li, Zang |
| contents | Slate recommendation, which presents users with a ranked item list in a single display, is ubiquitous across mainstream online platforms. Recent advances in generative models have shown significant potential for this task via autoregressive modeling of discrete semantic ID sequences. However, existing methods suffer from three key limitations: entangled item tokenization, inefficient sequential decoding, and the absence of holistic slate planning. These issues often result in substantial inference overhead and inadequate alignment with diverse user preferences and practical business requirements, hindering the industrial deployment of generative slate recommendation systems. In this paper, we propose HiGR, an efficient generative slate recommendation framework that integrates hierarchical planning with listwise preference alignment. First, we design an auto-encoder incorporating residual quantization and contrastive constraints, which tokenizes items into semantically structured IDs to enable controllable generation. Second, HiGR decouples the generation process into two stages: a list-level planning stage to capture global slate intent, and an item-level decoding stage to select specific items, effectively reducing the search space and enabling efficient generation. Third, we introduce a multi-objective and listwise preference alignment mechanism that enhances slate quality by leveraging implicit user feedback. Extensive experiments have validated the effectiveness of our HiGR method. Notably, it outperforms state-of-the-art baselines by over 10\% in offline recommendation quality while achieving a $5\times$ inference speedup. Furthermore, we have deployed HiGR on a commercial platform under Tencent (serving hundreds of millions of users), and online A/B tests show that it increases average watch time and average video plays by 1.22\% and 1.73\%, respectively. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_24787 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | HiGR: Efficient Generative Slate Recommendation via Hierarchical Planning and Multi-Objective Preference Alignment Pang, Yunsheng Liu, Zijian Li, Yudong Zhu, Shaojie Luo, Zijian Yu, Chenyun Wu, Sikai Shen, Shichen Xu, Cong Wang, Bin Jiang, Kai Yu, Hongyong Zhuo, Chengxiang Li, Zang Information Retrieval Artificial Intelligence Slate recommendation, which presents users with a ranked item list in a single display, is ubiquitous across mainstream online platforms. Recent advances in generative models have shown significant potential for this task via autoregressive modeling of discrete semantic ID sequences. However, existing methods suffer from three key limitations: entangled item tokenization, inefficient sequential decoding, and the absence of holistic slate planning. These issues often result in substantial inference overhead and inadequate alignment with diverse user preferences and practical business requirements, hindering the industrial deployment of generative slate recommendation systems. In this paper, we propose HiGR, an efficient generative slate recommendation framework that integrates hierarchical planning with listwise preference alignment. First, we design an auto-encoder incorporating residual quantization and contrastive constraints, which tokenizes items into semantically structured IDs to enable controllable generation. Second, HiGR decouples the generation process into two stages: a list-level planning stage to capture global slate intent, and an item-level decoding stage to select specific items, effectively reducing the search space and enabling efficient generation. Third, we introduce a multi-objective and listwise preference alignment mechanism that enhances slate quality by leveraging implicit user feedback. Extensive experiments have validated the effectiveness of our HiGR method. Notably, it outperforms state-of-the-art baselines by over 10\% in offline recommendation quality while achieving a $5\times$ inference speedup. Furthermore, we have deployed HiGR on a commercial platform under Tencent (serving hundreds of millions of users), and online A/B tests show that it increases average watch time and average video plays by 1.22\% and 1.73\%, respectively. |
| title | HiGR: Efficient Generative Slate Recommendation via Hierarchical Planning and Multi-Objective Preference Alignment |
| topic | Information Retrieval Artificial Intelligence |
| url | https://arxiv.org/abs/2512.24787 |