Search papers, labs, and topics across Lattice.
This paper addresses the critical issue of topology distortion in item tokenization within generative recommendation systems, which disrupts the intrinsic adjacency relationships of items in the semantic embedding space after quantization. To mitigate this problem, the authors introduce Topology-Aware Tokenization (TopoTok), a multi-level distillation framework that progressively recovers item relational structures at varying granularities. Extensive experiments reveal that TopoTok significantly enhances recommendation accuracy, achieving up to 9.42% improvement in Recall@5 compared to existing tokenization methods.
TopoTok not only preserves item relationships during tokenization but also boosts generative recommendation accuracy by over 9%鈥攁 game changer for recommendation systems.
Generative recommendation reformulates sequential recommendation as an autoregressive generation task, yet a critical issue in this paradigm remains overlooked: topology distortion in item tokenization. In particular, we observe that the intrinsic adjacency relationships of items in the pretrained semantic embedding space are significantly disrupted after quantization. This topology distortion misleads the model's perception of item similarity, ultimately bottlenecking the accuracy of generative recommendations. To address this issue, we propose Topology-Aware Tokenization (TopoTok), an item tokenization framework that preserves item relational structure throughout the quantization hierarchy. Different from the prior monolithic supervision in tokenization, TopoTok introduces a multi-level distillation scheme to progressively recover the topology from coarse to fine granularity: 1) Inter-Group Distillation to capture global cluster-wise relations; 2) Intra-Group Distillation to refine local structures within semantic clusters; and 3) Inter-Item Distillation to enforce fine-grained alignment at the individual item level. Extensive experiments on three benchmark datasets demonstrate that TopoTok effectively alleviates topology distortion and consistently outperforms state-of-the-art tokenizers, achieving significant performance gains of up to 9.42% in Recall@5.
