Instruction tuning modulates discourse biases in language models.

Florian Kankowski, Torgrim Solstad, Sina Zarrie, Oliver Bott

Browse the full CogSci paper archive.