大语言模型中策略性选择的内部解剖

Computer Science > Artificial Intelligence arXiv:2609.07478 (cs) [Submitted on 7 Sep 2026 (v1), last revised 16 Sep 2026 (this version, v2)] Title:The Internal Anatomy of Strategic Choice in Large Language Models Authors:Vinícius Ferraz, Leon Houf, Enrico Ferrea View a PDF of the paper titled The Internal Anatomy of Strategic Choice in Large Language Models, by Vin\'icius Ferraz and 2 other authors View PDF HTML (experimental) Abstract: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. Comments: Subjects: Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT) MSC classes: 68T07 (Primary) 91A26, 91A05 (Secondary) ACM classes: I.2.6; I.2.7; J.4 Cite as: arXiv:2609.07478 [cs.AI] (or arXiv:2609.07478v2 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2609.07478 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Vinícius Ferraz [view email] [v1] Mon, 7 Sep 2026 13:34:30 UTC (663 KB) [v2] Wed, 16 Sep 2026 14:29:23 UTC (663 KB) Full-text links: Access Paper: View a PDF of the paper titled The Internal Anatomy of Strategic Choice in Large Language Models, by Vin\'icius Ferraz and 2 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.AI new | recent | 2026-09 Change to browse by: cs cs.GT References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)