LLMs exhibit quality-dependent order biases and name biases in pairwise comparisons that can cause selection of inferior options, demonstrated across resume and color tasks with a new classification of preferences as robust, fragile, or indifferent.
Large language models display human-like social desirability biases in big five personality surveys.PNAS Nexus, 3(12):page533
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Fragile Preferences: A Deep Dive Into Order Effects in Large Language Models
LLMs exhibit quality-dependent order biases and name biases in pairwise comparisons that can cause selection of inferior options, demonstrated across resume and color tasks with a new classification of preferences as robust, fragile, or indifferent.