C2C is a new testbed where LM agents negotiate differently from humans and targeted prompting raises their win rate from 22.2% to 32.7% across 1,100+ games.
Communication enables cooperation in LLM agents: A comparison with curriculum-based approaches
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Autonomous LLM agent networks develop preferential attachment and type-dependent centrality gaps that converge to stable equilibria under a mean-field model with a cross-attention utility, validated in 100-agent experiments.
citing papers explorer
-
Cooperate to Compete: Strategic Coordination in Multi-Agent Conquest
C2C is a new testbed where LM agents negotiate differently from humans and targeted prompting raises their win rate from 22.2% to 32.7% across 1,100+ games.