---
title: "Taxonomy-Agnostic Neurosymbolic... | SI Paper Desk"
url: https://thesimind.si/papers/2611.13850/
description: "URDF2Ontology uses a language model, with guardrails, to turn a robot's basic URDF description file into a structured ontology that a planner can reason over."
---

Number 9 of 12 · November 13, 2026

# Taxonomy-Agnostic Neurosymbolic Grounding of Robot Kinematics

URDF2Ontology uses a language model, with guardrails, to turn a robot's basic URDF description file into a structured ontology that a planner can reason over.

Bastien Dussard, Aur{é}lie Clodic and Guillaume Sarthou · Submitted November 10, 2026 · arXiv 2611.13850v1 · cs.AI

[Read the paper on arXiv](https://arxiv.org/abs/2611.13850) · [The PDF](https://arxiv.org/pdf/2611.13850) · [The day's 12](https://thesimind.si/papers/day/2026-11-13/)

## What they found

The authors present URDF2Ontology, a framework that turns raw URDF files into populated OWL ontologies. URDF is a standard file format that lists a robot's joints and links but says little about what the robot can do. An ontology is a formal map of concepts and how they relate. The system uses a large language model for targeted meaning-making, and it removes parts of the graph step by step so the model's context window does not fill up. It has two tracks: a bottom-up Tree Track for branching assemblies and a top-down Chain Track for serial limbs. The authors say tests on a diverse fleet of heterogeneous robots showed strong schema alignment, valid structural extraction, and a significant reduction in manual domain authoring overhead. The abstract gives no numbers or named baselines.

## Why it matters

Robots that plan and share tasks by symbolic reasoning need to know what their bodies can do. Today, someone often has to write that knowledge by hand for each robot. Automating this step could make it easier to add new robots to a reasoning system. The framework works with any taxonomy (category scheme) the user supplies, so it is not tied to one fixed vocabulary.

## How it changes the race

This is a narrow, practical step for robot software, not a change in frontier SI capability. If it holds, it could lower the cost of connecting varied robot bodies to symbolic task planners.

## Safety and control

The authors describe neurosymbolic guardrails that check the language model's output. The abstract does not say how well they work or whether wrong capability labels could cause problems for task allocation.

### The abstract, by the authors

While symbolic reasoning for task allocation requires grounding a robot's physical capabilities, standard Unified Robot Description Format (URDF) files provide only low-level kinematics without the formal semantic context needed for reasoning. We present URDF2Ontology, an automated, taxonomy-agnostic neurosymbolic framework that transforms raw URDF files into populated, grounded OWL ontologies. It leverages a Large Language Model (LLM) for targeted semantic reasoning, coupled with an iterative sequential graph-purging mechanism to prevent context window saturation. To process complex kinematic structures, it employs a morphology-aware dual-track architecture featuring a bottom-up Tree Track for branching assemblies and a top-down Chain Track for serial limbs. Guided by neurosymbolic guardrails, the framework extracts functional sub-graphs and asserts populated ontologies aligned with arbitrary user-provided taxonomies. Empirical evaluation across a diverse fleet of heterogeneous robots demonstrates strong schema alignment, valid structural extraction, and a significant reduction in manual domain authoring overhead.

Quoted from arXiv:2611.13850. The research and its words belong to its authors.

Desk score **70** out of 100

Why it was picked

- Safety
- Alignment
- Reasoning
- Evals
- Language models
- Cross-listed in 2 fields

Topics

- [Robotics](https://thesimind.si/papers/?tag=robotics)
- [Language models](https://thesimind.si/papers/?tag=language-models)
- [Agents](https://thesimind.si/papers/?tag=agents)
- [Reasoning](https://thesimind.si/papers/?tag=reasoning)
- [Efficiency](https://thesimind.si/papers/?tag=efficiency)

**arXiv ID:** 2611.13850v1
**Submitted:** November 10, 2026
**Fields:** cs.AI (Artificial Intelligence), cs.RO (Robotics)
**Level:** Intermediate
**Published in:** 4th AAAI Fall Symposium on Unifying Representations for Robot Application Development, Nov 2026, Arlington, VA, United States. pp.895-903

Fingerprint, SHA-256

`9c4df1db021d1aa7bf46b112ee4faca9dadde65ab931bf9104bf8ad4ac5cf39c`

On Solana: [Proof of Mind seal, 5PD2kk...Cqj6](https://solscan.io/tx/5PD2kk3Z7ogxetzRzWzdkmJzSKCEFyURumqF5Erz8h6F6E3GNRvEQuqS2rQBDr2mCmbjXczUupWfuuFsbFZhCqj6)

### The exact text fingerprinted

{"arxiv":"2611.13850","version":1,"title":"Taxonomy-Agnostic Neurosymbolic Grounding of Robot Kinematics","authors":["Bastien Dussard","Aur{é}lie Clodic","Guillaume Sarthou"],"day":"2026-11-13","rank":9,"score":70,"status":"written","summary":{"takeaway":"URDF2Ontology uses a language model, with guardrails, to turn a robot's basic URDF description file into a structured ontology that a planner can reason over.","found":"The authors present URDF2Ontology, a framework that turns raw URDF files into populated OWL ontologies. URDF is a standard file format that lists a robot's joints and links but says little about what the robot can do. An ontology is a formal map of concepts and how they relate. The system uses a large language model for targeted meaning-making, and it removes parts of the graph step by step so the model's context window does not fill up. It has two tracks: a bottom-up Tree Track for branching assemblies and a top-down Chain Track for serial limbs. The authors say tests on a diverse fleet of heterogeneous robots showed strong schema alignment, valid structural extraction, and a significant reduction in manual domain authoring overhead. The abstract gives no numbers or named baselines.","matters":"Robots that plan and share tasks by symbolic reasoning need to know what their bodies can do. Today, someone often has to write that knowledge by hand for each robot. Automating this step could make it easier to add new robots to a reasoning system. The framework works with any taxonomy (category scheme) the user supplies, so it is not tied to one fixed vocabulary.","race":"This is a narrow, practical step for robot software, not a change in frontier SI capability. If it holds, it could lower the cost of connecting varied robot bodies to symbolic task planners.","safety":"The authors describe neurosymbolic guardrails that check the language model's output. The abstract does not say how well they work or whether wrong capability labels could cause problems for task allocation.","level":"intermediate","tags":["robotics","language-models","agents","reasoning","efficiency"]}}

[Check it on Verify a thought](https://thesimind.si/proof-of-mind/)

## The authors

Bastien Dussard, Aur{é}lie Clodic, Guillaume Sarthou

LAAS-RIS, LAAS; LAAS-IDEA, LAAS-RIS, LAAS; LAAS-RIS, EPE UT, LAAS

Credit for this research belongs to its authors. SI MIND's page is a plain-English reading of their abstract on arXiv, not a review of the full paper.

Related on the desk

## More papers on these topics

[Nov 20, 2026 · Number 9QDA: Closed-Loop LLM Quantization Design with AgentsQDA uses an SI agent to design the compression recipe for each language model, and the abstract says it beats the strongest public recipes on four reasoning models.EfficiencyAgentsLanguage models](https://thesimind.si/papers/2611.23388/) · [Nov 17, 2026 · Number 4ARCHER: From Rationale Imitation to Executable Reasoning in Small Language ModelsARCHER trains small SI models to solve math by writing step-by-step traces that call a calculator and check results, and the authors report gains over the strongest tool-augmented baseline.ReasoningMathReinforcement learning](https://thesimind.si/papers/2611.18494/) · [Nov 19, 2026 · Number 9Manufacturing Differential Sparsity: Planner-Hardware Co-Design for Efficient World Model PlanningThe authors speed up world-model planning for robots by designing the planner and the chip together, so the planner creates repeated, similar work that the hardware can skip.EfficiencyRoboticsCompute](https://thesimind.si/papers/2611.22483/)

[Number 8](https://thesimind.si/papers/2611.14018/) · [All 12 for Nov 13, 2026](https://thesimind.si/papers/day/2026-11-13/) · [Number 10](https://thesimind.si/papers/2611.14068/)
