LLM action selection approximates but does not reliably preserve a reference first-order Markov policy in OSN simulations and runs several hundred times slower.
arXiv preprint arXiv:2512.22082 (2025)
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Off-the-shelf LLMs underproduce hate speech and show model-specific biases relative to real Spanish news audience reactions, with fine-tuning yielding uneven improvements across models.
Simulations in synthetic social networks show amplification maximizes exposure, counter-messaging most effectively shifts beliefs, and narrative release requires larger attacker footprints.
citing papers explorer
-
Should LLM Agents Decide in Social Simulations? Comparing Finite-State and LLM-Based Decision Policies
LLM action selection approximates but does not reliably preserve a reference first-order Markov policy in OSN simulations and runs several hundred times slower.
-
Execution and assessment of agentic influence operations in simulated social networks
Simulations in synthetic social networks show amplification maximizes exposure, counter-messaging most effectively shifts beliefs, and narrative release requires larger attacker footprints.