OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914563989962752 |
|---|---|
| author | Chen, Ben Wang, Siyuan Ma, Yufei Liang, Zihan Zhang, Xuxin Lv, Yue Yang, Ying Dai, Huangyu Mao, Lingtao Zhao, Tong Qian, Zhipeng Sun, Xinyu Zhai, Zhixin Zhao, Yang Liu, Bochao Lv, Jingshan Liang, Xiao Kong, Hui Chen, Jing Li, Han Lei, Chenyi Ou, Wenwu Gai, Kun |
| author_facet | Chen, Ben Wang, Siyuan Ma, Yufei Liang, Zihan Zhang, Xuxin Lv, Yue Yang, Ying Dai, Huangyu Mao, Lingtao Zhao, Tong Qian, Zhipeng Sun, Xinyu Zhai, Zhixin Zhao, Yang Liu, Bochao Lv, Jingshan Liang, Xiao Kong, Hui Chen, Jing Li, Han Lei, Chenyi Ou, Wenwu Gai, Kun |
| contents | Generative Retrieval (GR) has emerged as a promising paradigm for modern search systems. Compared to multi-stage cascaded architecture, it offers advantages such as end-to-end joint optimization and high computational efficiency. OneSearch, as a representative industrial-scale deployed generative search framework, has brought significant commercial and operational benefits. However, its inadequate understanding of complex queries, inefficient exploitation of latent user intents, and overfitting to narrow historical preferences have limited its further performance improvement. To address these challenges, we propose OneSearch-V2, a latent reasoning enhanced self-distillation generative search framework. It contains three key innovations: (1) a thought-augmented complex query understanding module, which enables deep query understanding and overcomes the shallow semantic matching limitations of direct inference; (2) a reasoning-internalized self-distillation training pipeline, which uncovers users' potential yet precise e-commerce intentions beyond log-fitting through implicit in-context learning; (3) a behavior preference alignment optimization system, which mitigates reward hacking arising from the single conversion metric, and addresses personal preference via direct user feedback. Extensive offline evaluations demonstrate OneSearch-V2's strong query recognition and user profiling capabilities. Online A/B tests further validate its business effectiveness, yielding +3.98\% item CTR, +2.07\% buyer volume, and +2.11\% order volume. Manual evaluation further confirms gains in search experience quality, with +1.37\% in page good rate and +1.65\% in query-item relevance. More importantly, OneSearch-V2 effectively mitigates common search system issues such as information bubbles and long-tail sparsity, without incurring additional inference costs or serving latency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_24422 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework Chen, Ben Wang, Siyuan Ma, Yufei Liang, Zihan Zhang, Xuxin Lv, Yue Yang, Ying Dai, Huangyu Mao, Lingtao Zhao, Tong Qian, Zhipeng Sun, Xinyu Zhai, Zhixin Zhao, Yang Liu, Bochao Lv, Jingshan Liang, Xiao Kong, Hui Chen, Jing Li, Han Lei, Chenyi Ou, Wenwu Gai, Kun Information Retrieval Artificial Intelligence Computation and Language Generative Retrieval (GR) has emerged as a promising paradigm for modern search systems. Compared to multi-stage cascaded architecture, it offers advantages such as end-to-end joint optimization and high computational efficiency. OneSearch, as a representative industrial-scale deployed generative search framework, has brought significant commercial and operational benefits. However, its inadequate understanding of complex queries, inefficient exploitation of latent user intents, and overfitting to narrow historical preferences have limited its further performance improvement. To address these challenges, we propose OneSearch-V2, a latent reasoning enhanced self-distillation generative search framework. It contains three key innovations: (1) a thought-augmented complex query understanding module, which enables deep query understanding and overcomes the shallow semantic matching limitations of direct inference; (2) a reasoning-internalized self-distillation training pipeline, which uncovers users' potential yet precise e-commerce intentions beyond log-fitting through implicit in-context learning; (3) a behavior preference alignment optimization system, which mitigates reward hacking arising from the single conversion metric, and addresses personal preference via direct user feedback. Extensive offline evaluations demonstrate OneSearch-V2's strong query recognition and user profiling capabilities. Online A/B tests further validate its business effectiveness, yielding +3.98\% item CTR, +2.07\% buyer volume, and +2.11\% order volume. Manual evaluation further confirms gains in search experience quality, with +1.37\% in page good rate and +1.65\% in query-item relevance. More importantly, OneSearch-V2 effectively mitigates common search system issues such as information bubbles and long-tail sparsity, without incurring additional inference costs or serving latency. |
| title | OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework |
| topic | Information Retrieval Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2603.24422 |