Skip to main content
Search

# Semantic Centroid Vector Drift

1 related articles

Subscribe to updates
  1. arXiv

    HSTA tracks technology diffusion across 30,000 arXiv preprints and USPTO patents, finding the highest semantic drift in Large Language Models (0.332) while paper-volume velocity fails to Granger-cause frontier compute surges

    The study introduces Hyperspherical Semantic Trajectory Analysis (HSTA), an unsupervised pipeline that encodes 20,000 arXiv preprints and 10,000 USPTO patent abstracts with Sentence-BERT, projects them onto a unit hypersphere, clusters them into eight sub-topics with Spherical K-Means alongside UMAP reduction, and defines two metrics, Semantic Centroid Vector Drift and Commercialization Offset, then links them to Epoch AI compute data through Vector Autoregressive Granger tests, finding that Large Language Models (drift 0.332) and Artificial Intelligence Systems (0.234) evolve fastest semantically while quarterly paper volume velocity alone does not Granger-cause frontier training compute surges at conventional significance levels.

You’re all caught up