Search papers, labs, and topics across Lattice.
This study investigates the impact of explicit relational priors on the behavior of large language model-based multi-agent systems (LLM-MAS) during commons-governance simulations and debate tasks. By integrating natural-language representations of relational semantics into agent prompts, the research reveals that increased relational positivity enhances coordination and agreement among agents, although it does not consistently improve accuracy in objective tasks. The findings suggest that while relational priors can facilitate behavioral alignment in certain contexts, their application should be cautious and context-dependent, as they may not universally benefit all scenarios.
Explicit relational semantics can boost agent coordination in multi-agent systems, but may compromise accuracy when truth is paramount.
Large language model-based multi-agent systems (LLM-MAS) are designed through roles, debate protocols, and aggregation rules. These choices create implicit social expectations: agents may be expected to trust, challenge, defer to, or collaborate with peers. We study the effects of making inter-agent relation semantics explicit. We use a minimal signed-network formulation of relational priors and inject natural-language renderings into agent system prompts while holding the task protocol fixed. Across a commons-governance simulation and multi-agent debate, relational priors primarily act as convergence pressure: increasing relational positivity tends to make agents coordinate or agree more readily. This pressure can help when utility rewards behavioral alignment, as in sustainable resource governance and subjective consensus. It does not, however, reliably improve accuracy. In objective QA debates, higher positivity can increase agreement even when correctness-conditioned agreement does not improve and may decline in some settings. Effects vary by model backbone, relation type, and topology; explicit neutrality is not equivalent to omitting relational framing. We argue that relational priors should not be a default add-on for LLM-MAS. Their safer use is diagnostic and task-specific: compare against a no-prior baseline, monitor correctness-conditioned metrics when truth matters, and omit the relational layer when validation does not justify it.
Mila