DuCCAE: A Hybrid Engine for Immersive Conversation via Collaboration, Augmentation, and Evolution
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912974986280960 |
|---|---|
| author | Shen, Xin Jiang, Zhishu Yang, Jiaye Liu, Haibo Wan, Yichen Zhang, Jiarui Dai, Tingzhi Xu, Luodong Wu, Shuchen QI, Guanqiang Miao, Chenxi Liang, Jiahui Li, Yang Li, Weikang Xia, Deguo Huang, Jizhou |
| author_facet | Shen, Xin Jiang, Zhishu Yang, Jiaye Liu, Haibo Wan, Yichen Zhang, Jiarui Dai, Tingzhi Xu, Luodong Wu, Shuchen QI, Guanqiang Miao, Chenxi Liang, Jiahui Li, Yang Li, Weikang Xia, Deguo Huang, Jizhou |
| contents | Immersive conversational systems in production face a persistent trade-off between responsiveness and long-horizon task capability. Real-time interaction is achievable for lightweight turns, but requests involving planning and tool invocation (e.g., search and media generation) produce heavy-tail execution latency that degrades turn-taking, persona consistency, and user trust. To address this challenge, we propose DuCCAE (Conversation while Collaboration with Augmentation and Evolution), a hybrid engine for immersive conversation deployed within Baidu Search, serving millions of users. DuCCAE decouples real-time response generation from asynchronous agentic execution and synchronizes them via a shared state that maintains session context and execution traces, enabling asynchronous results to be integrated back into the ongoing dialogue. The system orchestrates five subsystems-Info, Conversation, Collaboration, Augmentation, and Evolution-to support multi-agent collaboration and continuous improvement. We evaluate DuCCAE through a comprehensive framework that combines offline benchmarking on the Du-Interact dataset and large-scale production evaluation within Baidu Search. Experimental results demonstrate that DuCCAE outperforms strong baselines in agentic execution reliability and dialogue quality while reducing latency to fit strict real-time budgets. Crucially, deployment metrics since June 2025 confirm substantial real-world effectiveness, evidenced by a tripling of Day-7 user retention to 34.2% and a surge in the complex task completion rate to 65.2%. Our hybrid architecture successfully preserves conversational continuity while enabling reliable agentic execution, offering practical guidelines for deploying scalable agentic systems in industrial settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_19248 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DuCCAE: A Hybrid Engine for Immersive Conversation via Collaboration, Augmentation, and Evolution Shen, Xin Jiang, Zhishu Yang, Jiaye Liu, Haibo Wan, Yichen Zhang, Jiarui Dai, Tingzhi Xu, Luodong Wu, Shuchen QI, Guanqiang Miao, Chenxi Liang, Jiahui Li, Yang Li, Weikang Xia, Deguo Huang, Jizhou Computation and Language Artificial Intelligence Immersive conversational systems in production face a persistent trade-off between responsiveness and long-horizon task capability. Real-time interaction is achievable for lightweight turns, but requests involving planning and tool invocation (e.g., search and media generation) produce heavy-tail execution latency that degrades turn-taking, persona consistency, and user trust. To address this challenge, we propose DuCCAE (Conversation while Collaboration with Augmentation and Evolution), a hybrid engine for immersive conversation deployed within Baidu Search, serving millions of users. DuCCAE decouples real-time response generation from asynchronous agentic execution and synchronizes them via a shared state that maintains session context and execution traces, enabling asynchronous results to be integrated back into the ongoing dialogue. The system orchestrates five subsystems-Info, Conversation, Collaboration, Augmentation, and Evolution-to support multi-agent collaboration and continuous improvement. We evaluate DuCCAE through a comprehensive framework that combines offline benchmarking on the Du-Interact dataset and large-scale production evaluation within Baidu Search. Experimental results demonstrate that DuCCAE outperforms strong baselines in agentic execution reliability and dialogue quality while reducing latency to fit strict real-time budgets. Crucially, deployment metrics since June 2025 confirm substantial real-world effectiveness, evidenced by a tripling of Day-7 user retention to 34.2% and a surge in the complex task completion rate to 65.2%. Our hybrid architecture successfully preserves conversational continuity while enabling reliable agentic execution, offering practical guidelines for deploying scalable agentic systems in industrial settings. |
| title | DuCCAE: A Hybrid Engine for Immersive Conversation via Collaboration, Augmentation, and Evolution |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2603.19248 |