Assured Autonomy with Neuro-Symbolic Perception
R. Spencer Hallyburton, Miroslav Pajic · 2025
Read Assured Autonomy with Neuro-Symbolic Perception by R. Spencer Hallyburton, Miroslav Pajic (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-21322.txt.
Description
Spencer Hallyburton and Miroslav Pajic argue that autonomous systems need perception with built-in reasoning, in a 2025 paper. Modern AI models are highly accurate pattern matchers. Yet they offer few safety guarantees. That worries designers of safety-critical machines. The authors propose adding symbolic structure to perception. Their paradigm joins object detection with scene graph building. A scene graph records objects and their relations.
Foundation models supply background knowledge offline. Specialist algorithms then build graphs in real time. The result is a structured picture of the scene. Such structure supports checks on what the system believes. Tests use physics simulators and real datasets. The authors show graphs bridge raw sensing and high-level reasoning. They frame the work as a base for resilient, context-aware autonomy. Trust, they say, must be designed in, not hoped for. The paper is a position and framework piece.
Catalog record
| CREATOR | R. Spencer Hallyburton, Miroslav Pajic |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-21322