BTGenBot-2: Efficient Behavior Tree Generation with Small Language Models

TL;DR

BTGenBot-2 is a lightweight, open-source small language model that efficiently generates executable behavior trees from natural language, enabling real-world robotic task planning with high success rates and faster inference.

Contribution

The paper introduces BTGenBot-2, the first open-source small language model capable of zero-shot behavior tree generation for robotics, with a new standardized benchmark for evaluation.

Findings

Outperforms larger models in success rates

Achieves up to 16x faster inference

Supports zero-shot and error recovery in behavior tree generation

Abstract

Recent advances in robot learning increasingly rely on LLM-based task planning, leveraging their ability to bridge natural language with executable actions. While prior works showcased great performances, the widespread adoption of these models in robotics has been challenging as 1) existing methods are often closed-source or computationally intensive, neglecting the actual deployment on real-world physical systems, and 2) there is no universally accepted, plug-and-play representation for robotic task generation. Addressing these challenges, we propose BTGenBot-2, a 1B-parameter open-source small language model that directly converts natural language task descriptions and a list of robot action primitives into executable behavior trees in XML. Unlike prior approaches, BTGenBot-2 enables zero-shot BT generation, error recovery at inference and runtime, while remaining lightweight enough for resource-constrained robots. We further introduce the first standardized benchmark for LLM-based BT generation, covering 52 navigation and manipulation tasks in NVIDIA Isaac Sim. Extensive evaluations demonstrate that BTGenBot-2 consistently outperforms GPT-5, Claude Opus 4.1, and larger open-source models across both functional and non-functional metrics, achieving average success rates of 90.38% in zero-shot and 98.07% in one-shot, while delivering up to 16x faster inference compared to the previous BTGenBot.