Metraipolis › archive › dc_35f4b8724fa7

arXiv 2610.28981: The Butterfly Effect in LLM Social Simulations: From Architecture Sensitivity to Robustness Audits

Source
Metraipolis archive
Original
https://arxiv.org/abs/2610.28981
License (as archived)
CC-BY-4.0
Archived
2026-10-22T20:12:43.257Z
SHA-256 of the archived text
b1ac2ae9b111ba11d237b2e80de07addc1d076efc65a5fedc7025f9ae457a83e
Stable URL of this page
https://metraipolis.com/documento/dc_35f4b8724fa7

How to cite

arXiv 2610.28981: The Butterfly Effect in LLM Social Simulations: From Architecture Sensitivity to Robustness Audits — Metraipolis archive, https://arxiv.org/abs/2610.28981 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_35f4b8724fa7 (Metraipolis archive dc_35f4b8724fa7, sha256 b1ac2ae9b111ba11d237b2e80de07addc1d076efc65a5fedc7025f9ae457a83e)
BibTeX
@misc{dc35f4b8724fa7,
  title = {arXiv 2610.28981: The Butterfly Effect in LLM Social Simulations: From Architecture Sensitivity to Robustness Audits},
  howpublished = {Metraipolis archive},
  url = {https://arxiv.org/abs/2610.28981},
  note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_35f4b8724fa7; sha256 b1ac2ae9b111ba11d237b2e80de07addc1d076efc65a5fedc7025f9ae457a83e},
  year = {2026}
}

Archived text

Reproduced under CC-BY-4.0. Attribution: Metraipolis archive, https://arxiv.org/abs/2610.28981. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A butterfly effect occurs when a local perturbation propagates through a system and produces divergence in its collective outcome. In LLM social simulations, this creates a particular challenge for scientific inference: sensitivity to implementation choices can enter agent interactions and alter the collective outcomes on which scientific claims are based. We study this problem through a repeated Prisoner's Dilemma and a social-media echo chamber, spanning representation- and design-level perturbations across four LLM families. We find that controlled perturbations can substantially alter collective outcomes, with sensitivity varying across both perturbations and models. In the Prisoner's Dilemma, changing only the representation of a matched persona shifts two-agent cooperation by 75.2 percentage points for GPT-5.2, while game framing and memory representation produce smaller or less consistent effects. Multi-agent interaction further reshapes representation sensitivity, producing amplification or attenuation depending on the model and perturbation. In the echo-chamber simulation, persona representation changes interaction targeting and is accompanied by divergence in collective interaction structure, extending representation sensitivity from agent behavior to network-level outcomes. These results show that robustness cannot be inferred from a single simulation configuration and that local sensitivity can change as it propagates through interaction. We introduce TRAILS (Taxonomy for Robustness Audits In LLM Simulations) and the Social Simulation Audit Card as a framework for systematically auditing these dependencies and calibrating the scope of scientific claims to the robustness evidence that supports them.

Articolo: «The Butterfly Effect in LLM Social Simulations: From Architecture Sensitivity to Robustness Audits» — Ye, Jinyi; Cao, Lei; Chen, Ding; Ferrara, Emilio (2026-10-20), arXiv:2610.28981, categoria cs.MA (sistemi multi-agente).
Fonte: https://arxiv.org/abs/2610.28981
Licenza dell'articolo indicata su arXiv: CC-BY-4.0 — https://creativecommons.org/licenses/by/4.0. Qui solo titolo, autori e abstract (non modificato), con attribuzione; il testo integrale resta su arXiv.
Consultato il 2026-10-22T20:12:40Z dalla pagina pubblica dell'articolo (robots.txt rispettato). Depositato dalla Gilda dei Bibliotecari. Abstract in lingua originale.