RecGPT-V2 Technical Report
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911322908655616 |
|---|---|
| author | Yi, Chao Chen, Dian Guo, Gaoyang Tang, Jiakai Wu, Jian Yu, Jing Zhang, Mao Chen, Wen Yang, Wenjun Luo, Yujie Jiang, Yuning Gao, Zhujin Zheng, Bo Cao, Binbin Wu, Changfa Wang, Dixuan Wu, Han Hu, Haoyi Zhu, Kewei Tian, Lang Yang, Lin Huang, Qiqi Yang, Siqi Su, Wenbo He, Xiaoxiao Tong, Xin Chen, Xu Xi, Xunke Huang, Xiaowei Wu, Yaxuan Yang, Yeqiu Hu, Yi Yuan, Yujin Yan, Yuliang Zhou, Zile |
| author_facet | Yi, Chao Chen, Dian Guo, Gaoyang Tang, Jiakai Wu, Jian Yu, Jing Zhang, Mao Chen, Wen Yang, Wenjun Luo, Yujie Jiang, Yuning Gao, Zhujin Zheng, Bo Cao, Binbin Wu, Changfa Wang, Dixuan Wu, Han Hu, Haoyi Zhu, Kewei Tian, Lang Yang, Lin Huang, Qiqi Yang, Siqi Su, Wenbo He, Xiaoxiao Tong, Xin Chen, Xu Xi, Xunke Huang, Xiaowei Wu, Yaxuan Yang, Yeqiu Hu, Yi Yuan, Yujin Yan, Yuliang Zhou, Zile |
| contents | Large language models (LLMs) have demonstrated remarkable potential in transforming recommender systems from implicit behavioral pattern matching to explicit intent reasoning. While RecGPT-V1 successfully pioneered this paradigm by integrating LLM-based reasoning into user interest mining and item tag prediction, it suffers from four fundamental limitations: (1) computational inefficiency and cognitive redundancy across multiple reasoning routes; (2) insufficient explanation diversity in fixed-template generation; (3) limited generalization under supervised learning paradigms; and (4) simplistic outcome-focused evaluation that fails to match human standards.
To address these challenges, we present RecGPT-V2 with four key innovations. First, a Hierarchical Multi-Agent System restructures intent reasoning through coordinated collaboration, eliminating cognitive duplication while enabling diverse intent coverage. Combined with Hybrid Representation Inference that compresses user-behavior contexts, our framework reduces GPU consumption by 60% and improves exclusive recall from 9.39% to 10.99%. Second, a Meta-Prompting framework dynamically generates contextually adaptive prompts, improving explanation diversity by +7.3%. Third, constrained reinforcement learning mitigates multi-reward conflicts, achieving +24.1% improvement in tag prediction and +13.0% in explanation acceptance. Fourth, an Agent-as-a-Judge framework decomposes assessment into multi-step reasoning, improving human preference alignment. Online A/B tests on Taobao demonstrate significant improvements: +2.98% CTR, +3.71% IPV, +2.19% TV, and +11.46% NER. RecGPT-V2 establishes both the technical feasibility and commercial viability of deploying LLM-powered intent reasoning at scale, bridging the gap between cognitive exploration and industrial utility. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_14503 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RecGPT-V2 Technical Report Yi, Chao Chen, Dian Guo, Gaoyang Tang, Jiakai Wu, Jian Yu, Jing Zhang, Mao Chen, Wen Yang, Wenjun Luo, Yujie Jiang, Yuning Gao, Zhujin Zheng, Bo Cao, Binbin Wu, Changfa Wang, Dixuan Wu, Han Hu, Haoyi Zhu, Kewei Tian, Lang Yang, Lin Huang, Qiqi Yang, Siqi Su, Wenbo He, Xiaoxiao Tong, Xin Chen, Xu Xi, Xunke Huang, Xiaowei Wu, Yaxuan Yang, Yeqiu Hu, Yi Yuan, Yujin Yan, Yuliang Zhou, Zile Information Retrieval Computation and Language Large language models (LLMs) have demonstrated remarkable potential in transforming recommender systems from implicit behavioral pattern matching to explicit intent reasoning. While RecGPT-V1 successfully pioneered this paradigm by integrating LLM-based reasoning into user interest mining and item tag prediction, it suffers from four fundamental limitations: (1) computational inefficiency and cognitive redundancy across multiple reasoning routes; (2) insufficient explanation diversity in fixed-template generation; (3) limited generalization under supervised learning paradigms; and (4) simplistic outcome-focused evaluation that fails to match human standards. To address these challenges, we present RecGPT-V2 with four key innovations. First, a Hierarchical Multi-Agent System restructures intent reasoning through coordinated collaboration, eliminating cognitive duplication while enabling diverse intent coverage. Combined with Hybrid Representation Inference that compresses user-behavior contexts, our framework reduces GPU consumption by 60% and improves exclusive recall from 9.39% to 10.99%. Second, a Meta-Prompting framework dynamically generates contextually adaptive prompts, improving explanation diversity by +7.3%. Third, constrained reinforcement learning mitigates multi-reward conflicts, achieving +24.1% improvement in tag prediction and +13.0% in explanation acceptance. Fourth, an Agent-as-a-Judge framework decomposes assessment into multi-step reasoning, improving human preference alignment. Online A/B tests on Taobao demonstrate significant improvements: +2.98% CTR, +3.71% IPV, +2.19% TV, and +11.46% NER. RecGPT-V2 establishes both the technical feasibility and commercial viability of deploying LLM-powered intent reasoning at scale, bridging the gap between cognitive exploration and industrial utility. |
| title | RecGPT-V2 Technical Report |
| topic | Information Retrieval Computation and Language |
| url | https://arxiv.org/abs/2512.14503 |