Measuring Task-Level Technological Exposure: A Language Model Approach

This repository contains the public dataset for "Measuring Task-Level Technological Exposure: A Language Model Approach". See github.com/armouton/patents-tasks for related code and documentation, and click here for a current draft of the paper. Data are drawn from ONET (versions 6.0-30.0), the Bureau of Labor Statistics' OEWS (annual releases, 2001-2025), and USPTO patent application publications (2001-2025). Models are trained versions of Sentence Transformers' mpnet-base-v2. Note: this is a stub! Current data version is preliminary, and code/documentation are not yet available. Please check back in January 2026.