Behavior Trees Enable Structured Programming of Language Model Agents
This work addresses the problem of ensuring reliable operation of language-model agents for developers and researchers, though it appears incremental as it builds on existing behavior tree concepts.
The paper tackles the brittleness of language models in larger systems by proposing behavior trees as a unifying framework, introducing Dendron, a Python library for programming language model agents, and demonstrating its application in three case studies including a chat agent, an infrastructure inspection agent, and a safety-constrained agent.
Language models trained on internet-scale data sets have shown an impressive ability to solve problems in Natural Language Processing and Computer Vision. However, experience is showing that these models are frequently brittle in unexpected ways, and require significant scaffolding to ensure that they operate correctly in the larger systems that comprise "language-model agents." In this paper, we argue that behavior trees provide a unifying framework for combining language models with classical AI and traditional programming. We introduce Dendron, a Python library for programming language model agents using behavior trees. We demonstrate the approach embodied by Dendron in three case studies: building a chat agent, a camera-based infrastructure inspection agent for use on a mobile robot or vehicle, and an agent that has been built to satisfy safety constraints that it did not receive through instruction tuning or RLHF.