factual

LLMs show position bias humans don't

Unlike humans, who across roughly 40 million judgments show no significant position bias (about 50.5% left vs 49.5% right), large language models display strong position bias of 10-25%, frequently changing which essay they prefer when the same two pieces are presented in reversed left-right order, typically favoring the left.

factualpending

Speaker

Daisy Christodoulou

Evidence Quote

“If the LLM sees it in order of A on the left, B on the right, it would pick A. And, if those are reversed, it might pick B.”

Source

How Better Feedback Can Revolutionize Education (with Daisy Christodoulou)— EconTalk
Created: 6/13/2026, 12:13:24 AM

Your notes

Loading notes...