NeSyPr: Neurosymbolic Proceduralization For Efficient Embodied Reasoning

TL;DR

NeSyPr introduces a neurosymbolic proceduralization framework that enables efficient, structured reasoning in language model agents for embodied tasks, reducing reliance on external symbolic tools and improving performance in resource-constrained environments.

Contribution

The paper proposes a novel neurosymbolic proceduralization method that compiles symbolic knowledge into procedural representations for efficient LM-based embodied reasoning.

Findings

NeSyPr achieves efficient reasoning on benchmarks like PDDLGym, VirtualHome, and ALFWorld.

It enables large-scale reasoning with smaller language models.

The framework reduces dependency on external symbolic planners during inference.

Abstract

We address the challenge of adopting language models (LMs) for embodied tasks in dynamic environments, where online access to large-scale inference engines or symbolic planners is constrained due to latency, connectivity, and resource limitations. To this end, we present NeSyPr, a novel embodied reasoning framework that compiles knowledge via neurosymbolic proceduralization, thereby equipping LM-based agents with structured, adaptive, and timely reasoning capabilities. In NeSyPr, task-specific plans are first explicitly generated by a symbolic tool leveraging its declarative knowledge. These plans are then transformed into composable procedural representations that encode the plans' implicit production rules, enabling the resulting composed procedures to be seamlessly integrated into the LM's inference process. This neurosymbolic proceduralization abstracts and generalizes multi-step symbolic structured path-finding and reasoning into single-step LM inference, akin to human knowledge compilation. It supports efficient test-time inference without relying on external symbolic guidance, making it well suited for deployment in latency-sensitive and resource-constrained physical systems. We evaluate NeSyPr on the embodied benchmarks PDDLGym, VirtualHome, and ALFWorld, demonstrating its efficient reasoning capabilities over large-scale reasoning models and a symbolic planner, while using more compact LMs.