Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2510.13291 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914254243758080 |
|---|---|
| author | Cheng, Xuxin Zeng, Ke Cao, Zhiquan Dai, Linyi Gao, Wenxuan Han, Fei Jian, Ai Hong, Feng Hu, Wenxing Huang, Zihe Kong, Dejian Leng, Jia Liao, Zhuoyuan Liu, Pei Lin, Jiaye Ma, Xing Ruan, Jingqing Song, Jiaxing Tan, Xiaoyu Xiao, Ruixuan Yu, Wenhui Zhan, Wenyu Zhang, Haoxing Zhou, Chao Zhou, Hao Zheng, Shaodong Chen, Ruinian Chen, Siyuan Chen, Ziyang Dong, Yiwen Fan, Yaoyou Fang, Yangyi Gan, Yang Guo, Shiguang He, Qi Hu, Chaowen Li, Binghui Li, Dailin Li, Xiangyu Li, Yan Liu, Chengjian Liu, Xiangfeng Lv, Jiahui Ma, Qiao Pan, Jiang Qin, Cong Sun, Chenxing Sun, Wen Wang, Zhonghui Wuerkaixi, Abudukelimu Yang, Xin Yuan, Fangyi Zhu, Yawen Zhai, Tianyi Zhang, Jie Zhang, Runlai Xu, Yao Zhao, Yiran Wang, Yifan Cai, Xunliang Hu, Yangen Liu, Cao Pan, Lu Wang, Xiaoli Xiao, Bo Yao, Wenyuan Zhou, Qianlin Zhu, Benchang |
| author_facet | Cheng, Xuxin Zeng, Ke Cao, Zhiquan Dai, Linyi Gao, Wenxuan Han, Fei Jian, Ai Hong, Feng Hu, Wenxing Huang, Zihe Kong, Dejian Leng, Jia Liao, Zhuoyuan Liu, Pei Lin, Jiaye Ma, Xing Ruan, Jingqing Song, Jiaxing Tan, Xiaoyu Xiao, Ruixuan Yu, Wenhui Zhan, Wenyu Zhang, Haoxing Zhou, Chao Zhou, Hao Zheng, Shaodong Chen, Ruinian Chen, Siyuan Chen, Ziyang Dong, Yiwen Fan, Yaoyou Fang, Yangyi Gan, Yang Guo, Shiguang He, Qi Hu, Chaowen Li, Binghui Li, Dailin Li, Xiangyu Li, Yan Liu, Chengjian Liu, Xiangfeng Lv, Jiahui Ma, Qiao Pan, Jiang Qin, Cong Sun, Chenxing Sun, Wen Wang, Zhonghui Wuerkaixi, Abudukelimu Yang, Xin Yuan, Fangyi Zhu, Yawen Zhai, Tianyi Zhang, Jie Zhang, Runlai Xu, Yao Zhao, Yiran Wang, Yifan Cai, Xunliang Hu, Yangen Liu, Cao Pan, Lu Wang, Xiaoli Xiao, Bo Yao, Wenyuan Zhou, Qianlin Zhu, Benchang |
| contents | Enhancing customer experience is essential for business success, particularly as service demands grow in scale and complexity. Generative artificial intelligence and Large Language Models (LLMs) have empowered intelligent interaction systems to deliver efficient, personalized, and 24/7 support. In practice, intelligent interaction systems encounter several challenges: (1) Constructing high-quality data for cold-start training is difficult, hindering self-evolution and raising labor costs. (2) Multi-turn dialogue performance remains suboptimal due to inadequate intent understanding, rule compliance, and solution extraction. (3) Frequent evolution of business rules affects system operability and transferability, constraining low-cost expansion and adaptability. (4) Reliance on a single LLM is insufficient in complex scenarios, where the absence of multi-agent frameworks and effective collaboration undermines process completeness and service quality. (5) The open-domain nature of multi-turn dialogues, lacking unified golden answers, hampers quantitative evaluation and continuous optimization. To address these challenges, we introduce WOWService, an intelligent interaction system tailored for industrial applications. With the integration of LLMs and multi-agent architectures, WOWService enables autonomous task management and collaborative problem-solving. Specifically, WOWService focuses on core modules including data construction, general capability enhancement, business scenario adaptation, multi-agent coordination, and automated evaluation. Currently, WOWService is deployed on the Meituan App, achieving significant gains in key metrics, e.g., User Satisfaction Metric 1 (USM 1) -27.53% and User Satisfaction Metric 2 (USM 2) +25.51%, demonstrating its effectiveness in capturing user needs and advancing personalized service. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_13291 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Higher Satisfaction, Lower Cost: A Technical Report on How LLMs Revolutionize Meituan's Intelligent Interaction Systems Cheng, Xuxin Zeng, Ke Cao, Zhiquan Dai, Linyi Gao, Wenxuan Han, Fei Jian, Ai Hong, Feng Hu, Wenxing Huang, Zihe Kong, Dejian Leng, Jia Liao, Zhuoyuan Liu, Pei Lin, Jiaye Ma, Xing Ruan, Jingqing Song, Jiaxing Tan, Xiaoyu Xiao, Ruixuan Yu, Wenhui Zhan, Wenyu Zhang, Haoxing Zhou, Chao Zhou, Hao Zheng, Shaodong Chen, Ruinian Chen, Siyuan Chen, Ziyang Dong, Yiwen Fan, Yaoyou Fang, Yangyi Gan, Yang Guo, Shiguang He, Qi Hu, Chaowen Li, Binghui Li, Dailin Li, Xiangyu Li, Yan Liu, Chengjian Liu, Xiangfeng Lv, Jiahui Ma, Qiao Pan, Jiang Qin, Cong Sun, Chenxing Sun, Wen Wang, Zhonghui Wuerkaixi, Abudukelimu Yang, Xin Yuan, Fangyi Zhu, Yawen Zhai, Tianyi Zhang, Jie Zhang, Runlai Xu, Yao Zhao, Yiran Wang, Yifan Cai, Xunliang Hu, Yangen Liu, Cao Pan, Lu Wang, Xiaoli Xiao, Bo Yao, Wenyuan Zhou, Qianlin Zhu, Benchang Computation and Language Artificial Intelligence Enhancing customer experience is essential for business success, particularly as service demands grow in scale and complexity. Generative artificial intelligence and Large Language Models (LLMs) have empowered intelligent interaction systems to deliver efficient, personalized, and 24/7 support. In practice, intelligent interaction systems encounter several challenges: (1) Constructing high-quality data for cold-start training is difficult, hindering self-evolution and raising labor costs. (2) Multi-turn dialogue performance remains suboptimal due to inadequate intent understanding, rule compliance, and solution extraction. (3) Frequent evolution of business rules affects system operability and transferability, constraining low-cost expansion and adaptability. (4) Reliance on a single LLM is insufficient in complex scenarios, where the absence of multi-agent frameworks and effective collaboration undermines process completeness and service quality. (5) The open-domain nature of multi-turn dialogues, lacking unified golden answers, hampers quantitative evaluation and continuous optimization. To address these challenges, we introduce WOWService, an intelligent interaction system tailored for industrial applications. With the integration of LLMs and multi-agent architectures, WOWService enables autonomous task management and collaborative problem-solving. Specifically, WOWService focuses on core modules including data construction, general capability enhancement, business scenario adaptation, multi-agent coordination, and automated evaluation. Currently, WOWService is deployed on the Meituan App, achieving significant gains in key metrics, e.g., User Satisfaction Metric 1 (USM 1) -27.53% and User Satisfaction Metric 2 (USM 2) +25.51%, demonstrating its effectiveness in capturing user needs and advancing personalized service. |
| title | Higher Satisfaction, Lower Cost: A Technical Report on How LLMs Revolutionize Meituan's Intelligent Interaction Systems |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2510.13291 |