liveBook · Manning

Co-occurrence

As explained in Machines that Think

The appearance of two or more events, features, or elements together, without necessarily implying causation.

Co-occurrence is the repeated appearance of elements together. For language models, it provides statistical associations: patterns in what appears together, not an account of why those patterns occur. section 4.5.4 section 1.5.6

In language, these patterns can organize words into related representations. This yields relational semantics: words such as “king,” “queen,” “throne,” and “monarch” cluster through related contexts, without requiring direct contact with what they describe. section 1.5.6

Broader inputs can strengthen associations across text, images, audio, and video, but do not by themselves establish causation. A model may associate smoke with fire without representing that fire causes smoke; multimodal conditioning narrows the gap between language and perception, not perception and understanding. section 4.5.4 section 4.8

How Machines that Think covers it

How deep it goes
Connects frequency-based associations to relational semantics and multimodal conditioning, while stopping short of causal explanation or interaction-based learning.
Examples
Uses smoke with fire, sirens with emergencies, facial expressions with emotions, and related words clustering through linguistic contexts.
What you can do after
Evaluate claims of grounding by distinguishing recurring associations from causal explanation and direct interaction with the world.
How it connects
Links co-occurrence to statistical learning, relational semantics, multimodal models, and the distinction between correlation and causation.

Also explained in

Compare how 4 books explain co-occurrence →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Machines that Think. The section links point to the parts of the book it draws on.