New GODR framework solves multi-goal conversational AI continuity

A new paper from Mariano Garralda-Barrio introduces the Goal-Oriented Dialogue Runtime (GODR), a design pattern aimed at solving a critical gap in production LLM workflows: maintaining conversational continuity when users juggle multiple interdependent goals. Current graph and multi-agent orchestration frameworks work well for simple guided processes, but they break down when objectives can be suspended, resumed, revised, or invalidated by actions in other goals. GODR addresses this by treating goals, task frames, lifecycle states, invalidation rules, and resumption contracts as first-class runtime objects, while delegating bounded execution to existing graph runtimes, agents, tools, or APIs.

The paper formalizes the problem, proposes specific runtime objects and architecture-selection criteria, and outlines an evaluation agenda for future empirical validation. GODR is not meant to replace simple workflow graphs—it targets complex, multi-domain, interruptible conversations where objective continuity cannot be recovered from agent identity, chat history, or execution-graph position alone. With 21 pages, 7 figures, and 10 tables, the work spans software engineering, AI, computation and language, and multiagent systems. It represents a conceptual blueprint for building more resilient conversational AI systems that can handle real-world task switching and goal interdependencies.