Paper
Reinforcing Real-world Service Agents: Balancing Utility and Cost in Task-oriented Dialogue
Authors
Ning Gao, Wei Zhang, Yuqin Dai, Ling Shi, Ziyin Wang, Yujie Wang, Wei He, Jinpeng Wang, Chaozheng Wang
Abstract
The rapid evolution of Large Language Models (LLMs) has accelerated the transition from conversational chatbots to general agents. However, effectively balancing empathetic communication with budget-aware decision-making remains an open challenge. Since existing methods fail to capture these complex strategic trade-offs, we propose InteractCS-RL, a framework that reframes task-oriented dialogue as a multi-granularity reinforcement learning process. Specifically, we first establish a User-centric Interaction Framework to provide a high-fidelity training gym, enabling agents to dynamically explore diverse strategies with persona-driven users. Then, we introduce Cost-aware Multi-turn Policy Optimization (CMPO) with a hybrid advantage estimation strategy. By integrating generative process credits and employing a PID-Lagrangian cost controller, CMPO effectively guides the policy to explore Pareto boundary between user reward and global cost constraints. Extensive experiments on customized real business scenarios demonstrate that InteractCS-RL significantly outperform other baselines across three evaluation dimensions. Further evaluation on tool-agent-user interaction benchmarks verify InteractCS-RL robustness across diverse domains.
Metadata
Related papers
Vibe Coding XR: Accelerating AI + XR Prototyping with XR Blocks and Gemini
Ruofei Du, Benjamin Hersh, David Li, Nels Numan, Xun Qian, Yanhe Chen, Zhongy... • 2026-03-25
Comparing Developer and LLM Biases in Code Evaluation
Aditya Mittal, Ryan Shar, Zichu Wu, Shyam Agarwal, Tongshuang Wu, Chris Donah... • 2026-03-25
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
Biplab Pal, Santanu Bhattacharya • 2026-03-25
Retrieval Improvements Do Not Guarantee Better Answers: A Study of RAG for AI Policy QA
Saahil Mathur, Ryan David Rittner, Vedant Ajit Thakur, Daniel Stuart Schiff, ... • 2026-03-25
MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
Zhuo Li, Yupeng Zhang, Pengyu Cheng, Jiajun Song, Mengyu Zhou, Hao Li, Shujie... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2602.22697v1</id>\n <title>Reinforcing Real-world Service Agents: Balancing Utility and Cost in Task-oriented Dialogue</title>\n <updated>2026-02-26T07:19:57Z</updated>\n <link href='https://arxiv.org/abs/2602.22697v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2602.22697v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>The rapid evolution of Large Language Models (LLMs) has accelerated the transition from conversational chatbots to general agents. However, effectively balancing empathetic communication with budget-aware decision-making remains an open challenge. Since existing methods fail to capture these complex strategic trade-offs, we propose InteractCS-RL, a framework that reframes task-oriented dialogue as a multi-granularity reinforcement learning process. Specifically, we first establish a User-centric Interaction Framework to provide a high-fidelity training gym, enabling agents to dynamically explore diverse strategies with persona-driven users. Then, we introduce Cost-aware Multi-turn Policy Optimization (CMPO) with a hybrid advantage estimation strategy. By integrating generative process credits and employing a PID-Lagrangian cost controller, CMPO effectively guides the policy to explore Pareto boundary between user reward and global cost constraints. Extensive experiments on customized real business scenarios demonstrate that InteractCS-RL significantly outperform other baselines across three evaluation dimensions. Further evaluation on tool-agent-user interaction benchmarks verify InteractCS-RL robustness across diverse domains.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n <published>2026-02-26T07:19:57Z</published>\n <arxiv:comment>35 pages, 8 tables, 3 figures</arxiv:comment>\n <arxiv:primary_category term='cs.CL'/>\n <author>\n <name>Ning Gao</name>\n </author>\n <author>\n <name>Wei Zhang</name>\n </author>\n <author>\n <name>Yuqin Dai</name>\n </author>\n <author>\n <name>Ling Shi</name>\n </author>\n <author>\n <name>Ziyin Wang</name>\n </author>\n <author>\n <name>Yujie Wang</name>\n </author>\n <author>\n <name>Wei He</name>\n </author>\n <author>\n <name>Jinpeng Wang</name>\n </author>\n <author>\n <name>Chaozheng Wang</name>\n </author>\n </entry>"
}