NLP

More Words, Less Thinking: The Paradoxical Homogenization of AI-Assisted Writing

AI makes writing more eloquent but less original. A Georgetown study of 373,000 college admissions essays found that after ChatGPT became widely available, essays used richer vocabulary but expressed narrower, more similar ideas. The author calls this paradoxical homogenization: the illusion of originality while our thoughts quietly converge.

Building RAG systems over technical documentation

Technical teams sit on large document sets: manuals, SOPs, API specs, incident runbooks. Retrieval-augmented generation (RAG) lets a language model answer questions only with evidence drawn from those documents—reducing hallucinations compared with raw prompting.

🚦 Quick Guide: How Model Context Protocol (MCP) Works

What is the Model Context Protocol (MCP)? The Model Context Protocol (MCP) is an open specification for secure, streamable communication and tooling integration between Language Model clients and servers. It enables language models and their helper agents to share context and resources efficiently—across different platforms, tools, and workflows.

📊 Build your own unigram frequency corpus with Google data

Interested in linguistic corpora and natural language processing? This post describes how to build your own English word-frequency (unigram) corpus using the Create_google_unigram_frecuency_corpus project.

🔍 Extracting Important Words Using Weirdness

Learn how to identify and extract rare or meaningful words from a corpus by leveraging the concept of “weirdness.” This approach enriches language processing applications, enabling you to highlight unique vocabulary and gain new linguistic insights.