What is mattwood.fyi?

A live, structured feed of riffs and links from Matt Wood, Chief AI & Technology Officer at AWS — readable by people and queryable by agents as a typed knowledge graph.

Is there an API and is it free?

Yes. A public, read-only query API is available with no authentication and no API keys. See https://mattwood.fyi/openapi.json and https://mattwood.fyi/docs.

How should an AI agent use this site?

Fetch https://mattwood.fyi/feed.json for the full structured feed, or query the API for keyword/semantic search and the connection graph. See https://mattwood.fyi/agents/ for when-to-use guidance.

Evolution of Text Classification: From Bag-of-Words to Modern APIs

A comprehensive guide tracing the history of text classification techniques from traditional bag-of-words and logistic regression through RNNs, CNNs, and Transformers, examining how modern models like Jev balance generality, speed, and cost for classification tasks.

magazine.sebastianraschka.com →

Connections

Develops into

Contrastive Language Model (CLM) The new item's arc from bag-of-words through Transformers traces the lineage that leads directly to contrastive language models, making CLM a natural endpoint of the evolutionary trajectory described

Supports

Engineering a Calibrated Classifier Model The historical survey of classification techniques provides foundational context for calibration engineering, as understanding model evolution helps explain why calibration remains necessary even with modern transformer-based classifiers
LLM Classification Is Feature Engineering Both examine text classification with LLMs, but the new item provides the historical evolution context (BoW → Transformers) that directly explains why LLM classification is now treated as feature engineering rather than end-to-end ML
PLUM: Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations The survey's coverage of how pre-trained language models transformed NLP classification provides direct historical grounding for understanding why PLUM adapts pre-trained LMs for recommendation—a downstream application of the same classification advances

Related to

Semantic ID-based Generative Retrieval for Podcast Discovery at Spotify Both deal with semantic representation for retrieval/classification tasks; the evolutionary history of text classification techniques is foundational to understanding semantic ID-based generative retrieval approaches at Spotify

Type:	link
First published:	Wednesday, 30 September 2026 at 09:12 UTC
Location:	https://magazine.sebastianraschka.com/p/classifier-history-and-jev
URL:	https://mattwood.fyi/i/45bb2d96

I'm Matt Wood, and this is For Your Information. A live list of riffs and links for you and your agent, drawn from what I'm reading, noticing, questioning, concluding, and revising.

Connections