More at Stake: How Payoff and Language Shape LLM Agent Strategies in Cooperation Dilemmas
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912851716734976 |
|---|---|
| author | Huynh, Trung-Kiet Duy-Minh, Dao-Sy Cao, Thanh-Bang Le, Phong-Hao Nguyen, Hong-Dan Quy, Nguyen Lam Phu Nguyen-Vo, Minh-Luan Pham, Hong-Phat Hoa, Pham Phu Than, Thien-Kim Tran, Chi-Nguyen Tran, Huy Tran-Le, Gia-Thoai Buscemi, Alessio Trang, Le Hong Han, The Anh |
| author_facet | Huynh, Trung-Kiet Duy-Minh, Dao-Sy Cao, Thanh-Bang Le, Phong-Hao Nguyen, Hong-Dan Quy, Nguyen Lam Phu Nguyen-Vo, Minh-Luan Pham, Hong-Phat Hoa, Pham Phu Than, Thien-Kim Tran, Chi-Nguyen Tran, Huy Tran-Le, Gia-Thoai Buscemi, Alessio Trang, Le Hong Han, The Anh |
| contents | As LLMs increasingly act as autonomous agents in interactive and multi-agent settings, understanding their strategic behavior is critical for safety, coordination, and AI-driven social and economic systems. We investigate how payoff magnitude and linguistic context shape LLM strategies in repeated social dilemmas, using a payoff-scaled Prisoner's Dilemma to isolate sensitivity to incentive strength. Across models and languages, we observe consistent behavioral patterns, including incentive-sensitive conditional strategies and cross-linguistic divergence. To interpret these dynamics, we train supervised classifiers on canonical repeated-game strategies and apply them to LLM decisions, revealing systematic, model- and language-dependent behavioral intentions, with linguistic framing sometimes matching or exceeding architectural effects. Our results provide a unified framework for auditing LLMs as strategic agents and highlight cooperation biases with direct implications for AI governance and multi-agent system design. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_19082 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | More at Stake: How Payoff and Language Shape LLM Agent Strategies in Cooperation Dilemmas Huynh, Trung-Kiet Duy-Minh, Dao-Sy Cao, Thanh-Bang Le, Phong-Hao Nguyen, Hong-Dan Quy, Nguyen Lam Phu Nguyen-Vo, Minh-Luan Pham, Hong-Phat Hoa, Pham Phu Than, Thien-Kim Tran, Chi-Nguyen Tran, Huy Tran-Le, Gia-Thoai Buscemi, Alessio Trang, Le Hong Han, The Anh Artificial Intelligence Computation and Language Computer Science and Game Theory Machine Learning Multiagent Systems 91A26, 68T05 I.2.11; I.2.6 As LLMs increasingly act as autonomous agents in interactive and multi-agent settings, understanding their strategic behavior is critical for safety, coordination, and AI-driven social and economic systems. We investigate how payoff magnitude and linguistic context shape LLM strategies in repeated social dilemmas, using a payoff-scaled Prisoner's Dilemma to isolate sensitivity to incentive strength. Across models and languages, we observe consistent behavioral patterns, including incentive-sensitive conditional strategies and cross-linguistic divergence. To interpret these dynamics, we train supervised classifiers on canonical repeated-game strategies and apply them to LLM decisions, revealing systematic, model- and language-dependent behavioral intentions, with linguistic framing sometimes matching or exceeding architectural effects. Our results provide a unified framework for auditing LLMs as strategic agents and highlight cooperation biases with direct implications for AI governance and multi-agent system design. |
| title | More at Stake: How Payoff and Language Shape LLM Agent Strategies in Cooperation Dilemmas |
| topic | Artificial Intelligence Computation and Language Computer Science and Game Theory Machine Learning Multiagent Systems 91A26, 68T05 I.2.11; I.2.6 |
| url | https://arxiv.org/abs/2601.19082 |