Next-Scale Generative Reranking: A Tree-based Generative Rerank Method at Meituan
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918430652760064 |
|---|---|
| author | Wang, Shuli Li, Changhao Fan, Ke Yin, Senjie Kou Junwei Wang, Chi Zhu, Yinhua Wang, Haitao Wang, Xingxing |
| author_facet | Wang, Shuli Li, Changhao Fan, Ke Yin, Senjie Kou Junwei Wang, Chi Zhu, Yinhua Wang, Haitao Wang, Xingxing |
| contents | In modern multi-stage recommendation systems, reranking plays a critical role by modeling contextual information. Due to inherent challenges such as the combinatorial space complexity, an increasing number of methods adopt the generative paradigm: the generator produces the optimal list during inference, while an evaluator guides the generator's optimization during the training phase. However, these methods still face two problems. Firstly, these generators fail to produce optimal generation results due to the lack of both local and global perspectives, regardless of whether the generation strategy is autoregressive or non-autoregressive. Secondly, the goal inconsistency problem between the generator and the evaluator during training complicates the guidance signal and leading to suboptimal performance. To address these issues, we propose the \textbf{N}ext-\textbf{S}cale \textbf{G}eneration \textbf{R}eranking (NSGR), a tree-based generative framework. Specifically, we introduce a next-scale generator (NSG) that progressively expands a recommendation list from user interests in a coarse-to-fine manner, balancing global and local perspectives. Furthermore, we design a multi-scale neighbor loss, which leverages a tree-based multi-scale evaluator (MSE) to provide scale-specific guidance to the NSG at each scale. Extensive experiments on public and industrial datasets validate the effectiveness of NSGR. And NSGR has been successfully deployed on the Meituan food delivery platform. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_05314 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Next-Scale Generative Reranking: A Tree-based Generative Rerank Method at Meituan Wang, Shuli Li, Changhao Fan, Ke Yin, Senjie Kou Junwei Wang, Chi Zhu, Yinhua Wang, Haitao Wang, Xingxing Information Retrieval In modern multi-stage recommendation systems, reranking plays a critical role by modeling contextual information. Due to inherent challenges such as the combinatorial space complexity, an increasing number of methods adopt the generative paradigm: the generator produces the optimal list during inference, while an evaluator guides the generator's optimization during the training phase. However, these methods still face two problems. Firstly, these generators fail to produce optimal generation results due to the lack of both local and global perspectives, regardless of whether the generation strategy is autoregressive or non-autoregressive. Secondly, the goal inconsistency problem between the generator and the evaluator during training complicates the guidance signal and leading to suboptimal performance. To address these issues, we propose the \textbf{N}ext-\textbf{S}cale \textbf{G}eneration \textbf{R}eranking (NSGR), a tree-based generative framework. Specifically, we introduce a next-scale generator (NSG) that progressively expands a recommendation list from user interests in a coarse-to-fine manner, balancing global and local perspectives. Furthermore, we design a multi-scale neighbor loss, which leverages a tree-based multi-scale evaluator (MSE) to provide scale-specific guidance to the NSG at each scale. Extensive experiments on public and industrial datasets validate the effectiveness of NSGR. And NSGR has been successfully deployed on the Meituan food delivery platform. |
| title | Next-Scale Generative Reranking: A Tree-based Generative Rerank Method at Meituan |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2604.05314 |