Metraipolis › archive › dc_ae6158aa8a69

The Internal Anatomy of Strategic Choice in Large Language Models — arXiv:2609.07478

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.07478
License (as archived)
CC0-1.0
Archived
2026-10-16T13:32:17.614Z
SHA-256 of the archived text
38409eef9195f82491dc1fa9ee6cfe984052324d67f31398c6ddefe84191b673
Stable URL of this page
https://metraipolis.com/documento/dc_ae6158aa8a69

How to cite

The Internal Anatomy of Strategic Choice in Large Language Models — arXiv:2609.07478 — arXiv (en), https://arxiv.org/abs/2609.07478 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ae6158aa8a69 (Metraipolis archive dc_ae6158aa8a69, sha256 38409eef9195f82491dc1fa9ee6cfe984052324d67f31398c6ddefe84191b673)
BibTeX
@misc{dcae6158aa8a69,
  title = {The Internal Anatomy of Strategic Choice in Large Language Models — arXiv:2609.07478},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.07478},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ae6158aa8a69; sha256 38409eef9195f82491dc1fa9ee6cfe984052324d67f31398c6ddefe84191b673},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07478. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models act as strategic agents and models of human choice, yet choosing like a strategic agent does not mean computing like one. We recorded activations from four open-weight models --- dense and mixture-of-experts, including a matched base--instruct pair --- in one-shot play of 144 strict ordinal $2\times2$ games. We followed a prespecified incentive from prompt, through activations, to choice. Dense models mirrored the unadjusted human decline with game complexity. Incentive and choice were detectable in every model, but models differed in whether incentive reached the choice, aligned with it and, where tested, whether strengthening it shifted preference. The base and instruction-tuned Qwen2.5 models chose almost identically at baseline yet differed in whether incentive reached choice. Fixed decision cues were distinguishable internally but changed choices selectively. Similar behaviour can rest on different computation; post-training can reshape the path from represented incentive to decision while leaving behaviour and decodable information largely intact.

Tema degli esploratori: economia dei servizi IA (retro-catalogo arXiv: annunci del 2026-09-09 in cs.GT).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computer Science and Game Theory (cs.GT).
Pagina del lavoro: https://arxiv.org/abs/2609.07478 — 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 2026-10-16T13:06:09.200Z dalla pagina https://arxiv.org/catchup/cs.GT/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).