Plan Verification for LLM-Based Embodied Task Completion Agents
This provides a scalable method for improving training data quality for imitation learning in embodied AI, though it appears incremental as it builds on existing LLM capabilities.
The paper tackles the problem of noisy LLM-based task plans in embodied AI by proposing an iterative verification framework where a Judge LLM critiques action sequences and a Planner LLM applies revisions, achieving up to 90% recall and 100% precision on the TEACh dataset with 96.5% of sequences converging in at most three iterations.
Large language model (LLM) based task plans and corresponding human demonstrations for embodied AI may be noisy, with unnecessary actions, redundant navigation, and logical errors that reduce policy quality. We propose an iterative verification framework in which a Judge LLM critiques action sequences and a Planner LLM applies the revisions, yielding progressively cleaner and more spatially coherent trajectories. Unlike rule-based approaches, our method relies on natural language prompting, enabling broad generalization across error types including irrelevant actions, contradictions, and missing steps. On a set of manually annotated actions from the TEACh embodied AI dataset, our framework achieves up to 90% recall and 100% precision across four state-of-the-art LLMs (GPT o4-mini, DeepSeek-R1, Gemini 2.5, LLaMA 4 Scout). The refinement loop converges quickly, with 96.5% of sequences requiring at most three iterations, while improving both temporal efficiency and spatial action organization. Crucially, the method preserves human error-recovery patterns rather than collapsing them, supporting future work on robust corrective behavior. By establishing plan verification as a reliable LLM capability for spatial planning and action refinement, we provide a scalable path to higher-quality training data for imitation learning in embodied AI.