Injecting explicit suggestions or biased recollections into prompts reduces LLM accuracy on multiple-choice QA tasks, and attention weights shift toward the suggested answer.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Investigating the Effects of Cognitive Biases in Prompts on Large Language Model Outputs
Injecting explicit suggestions or biased recollections into prompts reduces LLM accuracy on multiple-choice QA tasks, and attention weights shift toward the suggested answer.