Coordinating heterogeneous robot teams from free-form natural-language instructions is hard. Language-only planners struggle with long-horizon coordination and hallucination, while purely formal methods require closed-world models. We present FLEET, a hybrid decentralized framework that turns language into optimized multi-robot schedules. An LLM front-end produces (i) a task graph with durations and precedence and (ii) a capability-aware robot--task fitness matrix; a formal back-end solves a makespan-minimization problem while the underlying robots execute their free-form subtasks with agentic closed-loop control. Across multiple free-form language-guided autonomy coordination benchmarks, FLEET improves success over state of the art generative planners on two-agent teams across heterogeneous tasks. Ablations show that mixed integer linear programming (MILP) primarily improves temporal structure, while LLM-derived fitness is decisive for capability-coupled tasks; together they deliver the highest overall performance. We demonstrate the translation to real world challenges with hardware trials using a pair of quadruped robots with disjoint capabilities.

In this paper, we study the role of machine-learned predictions in offline NP-hard problems.

Learn-to-Defer is a paradigm that enables learning algorithms to work not in isolation but as a team with human experts.

We theoretically establish that our algorithms almost surely converge to locally optimal policies of our safe optimization framework.

For instance, the US Federal Reserve publishes synthetic market stress scenarios given by the constrained time series for financial institutions to assess their performance in hypothetical recessions.

The first and third equalities are straightforward. We restate a result extracted from the monograph by Luenberger [1969]. It relies on the dual functional φ, whose expression we recall below. Theorem 2 (stated as Theorem 1 in Section 8.6, page 224 in Luenberger, 1969) . " is required to apply the theorem.