Dalarna University's logo and link to the university's website

du.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • chicago-author-date
  • chicago-note-bibliography
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Språkmodeller i samspel: En studie av strategiskt beslutsfattande i ett tärningsbaserat spel
Dalarna University, School of Information and Engineering.
Dalarna University, School of Information and Engineering.
2025 (Swedish)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Language models in interaction : A study of strategic decision-making in a dice-based game (English)
Abstract [sv]

Denna studie undersöker hur agenter styrda av stora språkmodeller (LLM:er) beter sig i ett strukturerat tärningsspel med olika målstrukturer. Tre GPT-4o mini-agenter spelade flera omgångar med antingen individuell eller kollektiv målsättning, och fick diskutera sina drag i naturligt språk före varje beslut. Genom kvantitativ analys av poängfördelning, strategifördelning och poängspridning, samt kvalitativ analys av agenternas kommunikation, identifierades skillnader i samordning, strategi och rättvisa mellan målsättningarna. Resultaten visar att agenterna koordinerar sig bättre i det kollektiva läget, men att detta inte alltid leder till mer jämna utfall. Språklig interaktion används strategiskt, men utan stabil rollfördelning. Studien visar att LLM-agenter kan uppvisa viss emergent samverkan även utan central styrning, men att förmågan till långsiktig kollektiv strategi är begränsad. Resultaten bidrar till förståelsen av hur LLM-baserade multi-agent-system fungerar i semi-kooperativa miljöer, och pekar på behovet av framtida forskning kring arkitekturstöd för samordning.

Abstract [en]

This study investigates how agents powered by large language models (LLMs) behave in a structured dice game under different goal structures. Three GPT-4o mini agents played multiple rounds with either individual or collective objectives and were allowed to discuss their moves using natural language before making decisions. Through quantitative analysis of score distribution, strategy use, and point disparity, as well as qualitative analysis of agent communication, we identified differences in coordination, strategy, and fairness between the goal types. The results show that agents exhibit better coordination under collective conditions, but this does not necessarily lead to more equal outcomes. Language is used strategically to avoid conflict, but stable role allocation is lacking. The study shows that LLM agents can demonstrate emergent collaboration without centralized control, but their capacity for long-term collective strategy remains limited. These findings contribute to understanding how LLM-based multi-agent systems operate in semi-cooperative environments and highlight the need for architectural support to enhance coordination.

Place, publisher, year, edition, pages
2025.
Keywords [en]
Large Language Models (LLM), Multi-agent systems, Strategic decision-making, Linguistic coordination, Goal structure, Emergent collaboration, Role allocation, Qualitative analysis, Decentralized coordination, Score disparity
Keywords [sv]
Large Language Models (LLM), Multi-agent-system, Strategiskt beslutsfattande, Språklig koordinering, Målstruktur, Emergent samverkan, Rollfördelning, Kvalitativ analys, Koordination utan central styrning, Poängspridning
National Category
Information Systems
Identifiers
URN: urn:nbn:se:du-51070OAI: oai:DiVA.org:du-51070DiVA, id: diva2:1988613
Subject / course
Informatics
Available from: 2025-08-12 Created: 2025-08-12 Last updated: 2025-10-09

Open Access in DiVA

fulltext(1024 kB)73 downloads
File information
File name FULLTEXT01.pdfFile size 1024 kBChecksum SHA-512
84334ae5d735b113b66accbad6204262f3a6752328a7706c703105f454e3ff185d5d33d3921255a34ccf07ca31b530004fc0a46276a56c8efb81ad8bcf250b69
Type fulltextMimetype application/pdf

By organisation
School of Information and Engineering
Information Systems

Search outside of DiVA

GoogleGoogle Scholar
Total: 74 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 168 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • chicago-author-date
  • chicago-note-bibliography
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf