SEAD: Self-Evolving Agent for Multi-Turn Service Dialogue
This addresses the challenge of noisy, low-quality data in service dialogues for AI systems, representing a strong domain-specific advancement.
The paper tackles the problem of suboptimal performance in service dialogues by proposing SEAD, a framework that enables agents to learn effective strategies without large-scale human annotations, resulting in a 17.6% improvement in task completion rate and 11.1% improvement in dialogue efficiency.
Large Language Models have demonstrated remarkable capabilities in open-domain dialogues. However, current methods exhibit suboptimal performance in service dialogues, as they rely on noisy, low-quality human conversation data. This limitation arises from data scarcity and the difficulty of simulating authentic, goal-oriented user behaviors. To address these issues, we propose SEAD (Self-Evolving Agent for Service Dialogue), a framework that enables agents to learn effective strategies without large-scale human annotations. SEAD decouples user modeling into two components: a Profile Controller that generates diverse user states to manage training curriculum, and a User Role-play Model that focuses on realistic role-playing. This design ensures the environment provides adaptive training scenarios rather than acting as an unfair adversary. Experiments demonstrate that SEAD significantly outperforms Open-source Foundation Models and Closed-source Commercial Models, improving task completion rate by 17.6% and dialogue efficiency by 11.1%. Code is available at: https://github.com/Da1yuqin/SEAD.