LLM-based simulations of opinion dynamics can approximate a classical model, but they are so sensitive to minor prompt variations that their scientific reliability is questionable.
Emergence of scaling in random networks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CY 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Sense and Sensitivity: Evaluating the simulation of social dynamics via Large Language Models
LLM-based simulations of opinion dynamics can approximate a classical model, but they are so sensitive to minor prompt variations that their scientific reliability is questionable.