Columbia | Zuckerman Institute
•82% Informative
A new study shows that large language models remain vulnerable to mistaking nonsense for natural language.
Researchers at Columbia 's Zuckerman Institute challenged nine different language models with hundreds of pairs of sentences.
All the models made mistakes, sometimes choosing sentences that sound like nonsense to a human ear.
"Every model exhibited blind spots, labeling some sentences as meaningful".
VR Score
90
Informative language
94
Neutral language
51
Article tone
informal
Language
English
Language complexity
57
Offensive language
possibly offensive
Hate speech
not hateful
Attention-grabbing headline
not detected
Known propaganda techniques
not detected
Time-value
long-living
External references
1
Source diversity
1
Affiliate links
no affiliate links