Neuro-Symbolic Framework for Tacit Knowledge Extraction in Manufacturing
A new research paper introduces a neuro-symbolic framework combining Logic-Augmented Generation and Active Inference to extract tacit knowledge from procedural domains. Published on arXiv (2605.07639), the study addresses the challenge of capturing implicit assumptions, contextual constraints, and embodied skills that are rarely documented. The approach is evaluated in a manufacturing case study using assembly-like repair procedures, aiming to transform tacit knowledge into machine-interpretable Knowledge Graphs. The framework integrates symbolic reasoning with generative models to formalize process-centric expertise for querying, validation, and reuse.
Key facts
- Paper published on arXiv with ID 2605.07639
- Focuses on tacit knowledge extraction in procedural domains
- Combines Logic-Augmented Generation and Active Inference
- Uses ontology-grounded Knowledge Graph construction
- Evaluated in manufacturing case study with assembly-like repair procedures
- Addresses challenge of capturing implicit and embodied skills
- Aims to create machine-interpretable representations
- Neuro-symbolic approach integrating symbolic reasoning and generative models
Entities
Institutions
- arXiv