Search papers, labs, and topics across Lattice.
This paper introduces PACE, a novel framework that enhances embodied planning by interleaving reasoning and execution, addressing the critical bottleneck of inference delays in existing models. By employing an Interleaved Think-Act architecture and a Dynamic Budget Allocator, PACE achieves a remarkable 67% improvement in success rates on the Robotouille benchmark while accelerating thinking time by 6.9 times compared to traditional methods. The findings indicate that optimizing cognitive effort allocation can significantly enhance both the quality of planning and time efficiency in latency-sensitive environments.
Interleaving reasoning and execution allows PACE to hide 66.8% of thinking time within action execution, drastically improving planning efficiency.
Reasoning-enhanced large language models have achieved remarkable improvements in planning tasks, yet their deployment in embodied systems remains impractical due to prohibitive inference delays-often exceeding minutes per planning instance. The fundamental bottleneck stems from the serial nature of existing paradigms: models must complete all reasoning before any action execution, leaving execution time windows entirely unexploited. We introduce PACE (Planning with Adaptive Cognitive Effort), a framework that enables interleaved reasoning and execution through two key innovations: an Interleaved Think-Act architecture that pipelines cognitive processing with action execution, and a Dynamic Budget Allocator that adapts reasoning token budgets to available execution time windows. On the Robotouille benchmark using Qwen3-8B-AWQ, PACE achieves a 10% success rate-representing a 67% improvement over the ReAct+Think baseline-while delivering 6.9 times acceleration in thinking time compared to unconstrained reasoning. The framework hides 66.8% of thinking time within execution windows, demonstrating that strategic cognitive effort allocation can simultaneously improve both planning quality and time efficiency. These results provide evidence that time-aware architectural innovations enable reasoning models to operate in latency-sensitive embodied domains where they were previously impractical.
