Search papers, labs, and topics across Lattice.
This paper conducts a comprehensive survey of the functional roles of language in embodied agents, identifying five distinct roles: Specification, Embodied Representation, Action Orchestration, Grounding Regulation, and Execution Coupling. By auditing the existing literature, the authors reveal a significant disconnect between the functional claims made about language and the empirical evidence supporting these claims, highlighting that many linguistic intermediates may be misapplied or ineffective. The study emphasizes the need for a more rigorous evaluation of grounding claims, advocating for a role-based comparison that allows for a nuanced understanding of language's contributions across different agent architectures.
Language's role in embodied agents is often overstated, with many claims lacking robust empirical support, revealing a critical gap in our understanding of its contributions.
Foundation models place language throughout embodied agents, but its presence does not show what it contributes or how well that contribution is grounded. This survey separates these two questions. We define five non-exclusive functional roles for language: Specification, Embodied Representation, Action Orchestration, Grounding Regulation, and Execution Coupling. For each role, we trace the path from linguistic content to its embodied consumer and identify the observations or interventions that can test the claimed responsibility. Applying this framework to the reviewed literature reveals a recurring gap between functional use and evidential support. Interpretable or revised linguistic intermediates may be incorrect, go unused, or fail to affect later behavior. Even when actions are directly conditioned on language, system-level success does not by itself isolate language's contribution. We therefore evaluate grounding claim by claim, asking whether the reported evidence supports the specific responsibility assigned to language. Using role claims rather than architectures as the unit of comparison allows us to compare modular and end-to-end embodied agents without extending conclusions beyond the reported evidence.
