The Counterfactual
Subscribe
Sign in
Home
Archive
About
Why benchmarks don't resolve disagreements
In which I argue it mostly comes down to uncertainty about robust generalization of behavior.
Apr 3
•
Sean Trott
7
9
Latest
Top
Discussions
Language models and the intentional stance
When are words like "belief" and "desire" useful, and when do we risk mistaking the map for the territory?
Mar 18
•
Sean Trott
16
16
1
In what sense—if any—are LLMs "model organisms" for the study of human cognition?
When researchers suggest using LLMs as "in silico" models, what exactly might that mean?
Mar 10
•
Sean Trott
5
4
1
Situation models: a cognitively plausible abstraction
An old idea from Cognitive Science can lend theoretical purchase to debates about the capacities of LLMs.
Feb 18
•
Sean Trott
6
3
The necessity (and complexity) of fixed points in measurement
Hasok Chang's "Inventing Temperature" offers an instructive example of epistemic iteration in practice.
Feb 11
•
Sean Trott
5
1
Healthy minds and sick souls
William James on two kinds of religious temperament.
Jan 31
•
Sean Trott
8
6
1
Calibrating expectations
The "habitability problem" never really went away.
Jan 17
•
Sean Trott
5
5
Evaluating empirical claims: practical advice
A conceptual framework for identifying the type and validity of empirical claims.
Dec 30, 2025
•
Sean Trott
9
1
See all
The Counterfactual
Cognitive Science, statistics, AI.
Subscribe
Recommendations
By the Numbers
Hannah Ritchie
Understanding AI
Timothy B. Lee
The Discourse Lounge
Darrell Owens
Align
Pam
The Counterfactual
Subscribe
About
Archive
Recommendations
Sitemap
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
