Metraipolis › archive › dc_dfbcf0b88d33
Interpreting and Steering LLM Agents for Social Simulations — arXiv:2609.16436
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.16436
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T18:34:53.935Z
- SHA-256 of the archived text
52a391675c0622bb8fa137f54ae802a1403efe93c22dec4bec3f26ae8ef22f24- Stable URL of this page
https://metraipolis.com/documento/dc_dfbcf0b88d33
How to cite
Interpreting and Steering LLM Agents for Social Simulations — arXiv:2609.16436 — arXiv (en), https://arxiv.org/abs/2609.16436 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dfbcf0b88d33 (Metraipolis archive dc_dfbcf0b88d33, sha256 52a391675c0622bb8fa137f54ae802a1403efe93c22dec4bec3f26ae8ef22f24)
BibTeX
@misc{dcdfbcf0b88d33,
title = {Interpreting and Steering LLM Agents for Social Simulations — arXiv:2609.16436},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.16436},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dfbcf0b88d33; sha256 52a391675c0622bb8fa137f54ae802a1403efe93c22dec4bec3f26ae8ef22f24},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.16436. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Simulations based on large language models (LLMs) have proven to be powerful for understanding human behavior, making them valuable additions to the social scientific toolkit. However, LLMs are ultimately black boxes based on deep neural networks which limits their value for social science. This is because of a lack of (i) interpretability: i.e. the ability to assign clear mechanisms driving observed behavior; and a lack of (ii) steerability: i.e. the ability to mute or amplify specific theoretically meaningful mechanisms of action to drive specific model behavior. Here, we demonstrate how the black box could be opened up to further enrich LLM-based simulations. Specifically, we compare three types of methods: (1) prompt-based manipulation, (2) SAE-derived feature steering, and (3) probe-based direction steering and examine their utility for LLM-based social scientific simulations. We do so by interpreting and steering two foundational components of human behaviors, namely preferences (risk attitudes, altruism) and capabilities (divergent creativity, product innovation), operationalized using four classic economic and creative tasks implemented as natural-language interactions. Overall, our results show that SAE- and probe-based techniques often outperform basic prompt-based methods for steering LLM agents, although this advantage depends on the specific prompting strategy involved. Together, SAEs and probes constitute an effective pipeline for social scientists seeking to interpret and steer agents in social simulations: SAEs decompose agents' internal representations into human-readable features, after which probes can reliably shift agents' behaviors in specified directions. We discuss implications of these methods for future work using LLM agents for social scientific simulations. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-16 in cs.AI). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.16436 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-15T18:03:09.639Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).