Modeling Behavioral Intensity and Transitions for Generative Recommendation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918470027837440 |
|---|---|
| author | Yang, Wenxuan Xu, Xiaoyang Zhang, Hanyu Xu, Zhexuan Xiong, Wanqiang Chen, Zhaoqun |
| author_facet | Yang, Wenxuan Xu, Xiaoyang Zhang, Hanyu Xu, Zhexuan Xiong, Wanqiang Chen, Zhaoqun |
| contents | Multi-behavior recommendation aims to predict user conversions by modeling various interaction types that carry distinct intent signals. Recently, generative sequence modeling methods have emerged as an important paradigm for multi-behavior recommendation by achieving flexible sequence generation. However, existing generative methods typically treat behaviors as auxiliary token features and feed them into unified attention mechanisms. These models implicitly assume uniform activation of dependencies among historical behaviors, thereby failing to discern differences in intensity or capture transition patterns. To address these limitations, we propose BITRec, a novel generative multi-behavior recommendation framework that introduces structured behavioral modeling through selective dependency activation. BITRec incorporates (i) Hierarchical Behavior Aggregation (HBA), which explicitly models behavioral intensity differences through separated exploration and commitment pathways, and (ii) Transition Relation Encoding (TRE), which encodes transition structures through explicit learnable relation matrices. Experiments on four large-scale datasets (RetailRocket, Taobao, Tmall, Insurance Dataset) with millions of interactions achieve consistent improvements of 15-23% across multiple metrics, with peak gains of 22.79% MRR on Tmall and 17.83% HR@10, 17.55% NDCG@10 on Taobao. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_24472 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Modeling Behavioral Intensity and Transitions for Generative Recommendation Yang, Wenxuan Xu, Xiaoyang Zhang, Hanyu Xu, Zhexuan Xiong, Wanqiang Chen, Zhaoqun Information Retrieval Artificial Intelligence Machine Learning Multi-behavior recommendation aims to predict user conversions by modeling various interaction types that carry distinct intent signals. Recently, generative sequence modeling methods have emerged as an important paradigm for multi-behavior recommendation by achieving flexible sequence generation. However, existing generative methods typically treat behaviors as auxiliary token features and feed them into unified attention mechanisms. These models implicitly assume uniform activation of dependencies among historical behaviors, thereby failing to discern differences in intensity or capture transition patterns. To address these limitations, we propose BITRec, a novel generative multi-behavior recommendation framework that introduces structured behavioral modeling through selective dependency activation. BITRec incorporates (i) Hierarchical Behavior Aggregation (HBA), which explicitly models behavioral intensity differences through separated exploration and commitment pathways, and (ii) Transition Relation Encoding (TRE), which encodes transition structures through explicit learnable relation matrices. Experiments on four large-scale datasets (RetailRocket, Taobao, Tmall, Insurance Dataset) with millions of interactions achieve consistent improvements of 15-23% across multiple metrics, with peak gains of 22.79% MRR on Tmall and 17.83% HR@10, 17.55% NDCG@10 on Taobao. |
| title | Modeling Behavioral Intensity and Transitions for Generative Recommendation |
| topic | Information Retrieval Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2604.24472 |