DarwinTOD: LLM-driven Lifelong Self-evolution for Task-oriented Dialog Systems

TL;DR

DarwinTOD is a novel framework that enables task-oriented dialog systems to autonomously evolve and improve over time through a dual-loop process combining online interactions and offline evolution, without human intervention.

Contribution

It introduces a unified lifelong self-evolving dialog framework integrating evolutionary computation and LLM-driven self-improvement, enabling continuous strategy refinement from scratch.

Findings

DarwinTOD outperforms previous state-of-the-art methods.

The system demonstrates continuous performance improvements during evolution.

It operates effectively without task-specific fine-tuning.

Abstract

Traditional task-oriented dialog systems are unable to evolve from ongoing interactions or adapt to new domains after deployment, that is a critical limitation in real-world dynamic environments. Continual learning approaches depend on episodic retraining with human curated data, failing to achieve autonomy lifelong improvement. While evolutionary computation and LLM driven self improvement offer promising mechanisms for dialog optimization, they lack a unified framework for holistic, iterative strategy refinement. To bridge this gap, we propose DarwinTOD, a lifelong self evolving dialog framework that systematically integrates these two paradigms, enabling continuous strategy optimization from a zero-shot base without task specific fine-tuning. DarwinTOD maintains an Evolvable Strategy Bank and operates through a dual-loop process: online multi-agent dialog execution with peer critique, and offline structured evolutionary operations that refine the strategy bank using accumulated feedback. This closed-loop design enables autonomous continuous improvement without human intervention. Extensive experiments show that DarwinTOD surpasses previous state-of-the-art methods and exhibits continuous performance gains throughout evolution. Our work provides a novel framework for building dialog systems with lifelong self evolution capabilities.