In a public goods game with costly sanctioning, reasoning-focused LLMs (o1, o3-mini) cooperated less and free-rode more than traditional LLMs such as Llama-3.3 and GPT-4o.
Whether it be control of their own destiny or control of others’ behavior
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Corrupted by Reasoning: Reasoning Language Models Become Free-Riders in Public Goods Games
In a public goods game with costly sanctioning, reasoning-focused LLMs (o1, o3-mini) cooperated less and free-rode more than traditional LLMs such as Llama-3.3 and GPT-4o.