DeCo: Task Decomposition and Skill Composition for Zero-Shot Generalization in Long-Horizon 3D Manipulation

Generalizing language-conditioned multi-task imitation learning (IL) models to novel long-horizon 3D manipulation tasks is challenging. To address this, we propose <bold>DeCo</bold> (<italic>Task</italic> <bold>De</bold> <italic>composition and Skill</italic> <bold>Co</bold> <italic>mposition</italic>), a model-agnostic framework that enhances zero-shot generalization to compositional long-horizon manipulation tasks. DeCo decomposes IL demonstrations into modular atomic tasks based on gripper-object interactions, creating a dataset that enables models to learn reusable skills. At inference, DeCo uses a vision-language model (VLM) to parse high-level instructions, retrieve relevant skills, and dynamically schedule their execution. A spatially-aware skill-chaining module ensures smooth, collision-free transitions between skills. We introduce <monospace>DeCoBench</monospace>, a benchmark designed to evaluate compositional generalization in long-horizon manipulation tasks. DeCo improves the success rate of three IL models—RVT-2, 3DDA, and ARP—by <bold>66.67%, 21.53%, and 57.92%</bold>, respectively, on 12 novel tasks. In real-world experiments, the DeCo-enhanced model, trained on only 6 atomic tasks, completes 9 novel tasks in zero-shot, with a <bold>53.33%</bold> improvement over the baseline model.