Local Predictability and Collective Fidelity in LLM-Agent Societies
Quick summary
arXiv:2609.35813v1 Announce Type: cross Abstract: Compact surrogates could reduce the cost of simulating large language model societies, but must reproduce collective behavior. We compare individual predictions and collective forecasts using 9,455 published trajectories and new experiments on opinion dynamics. Neighbor information improves individual prediction in all 16 public-data settings and pooled collective forecasts on held-out questions, although collective gains depend on transfer conditions. Tests on 24 new statements do not confirm earlier contrasting history effects in forecasts fr
Key takeaways
- arXiv:2609.35813v1 Announce Type: cross Abstract: Compact surrogates could reduce the cost of simulating large language model societies, but must reproduce collective behavior.
- We compare individual predictions and collective forecasts using 9,455 published trajectories and new experiments on opinion dynamics.
- Neighbor information improves individual prediction in all 16 public-data settings and pooled collective forecasts on held-out questions, although collective gains depend on transfer conditions.
Why it matters
“Local Predictability and Collective Fidelity in LLM-Agent Societies” should be evaluated beyond branding and benchmark scores. Its practical importance will emerge in task accuracy, latency, unit cost, safety and integration with real workflows.
Member comments