Science Briefing · Jul 6, 2026
On the Referential Capacity of Language Models: An Internalist Rejoinder to Mandelkern & Linzen
0Sign in to vote or save
The paper qualifies Mandelkern and Linzen’s claim that words generated by language models refer to entities in the world, arguing it holds only for a narrow class of expressions rather than as a gene…
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.